011B - Feed Efficiency in Swine PDF

Feed efficiency
in swine
edited by: John F. Patience

Feed efficiency in swine
This project was supported by Agriculture and Food Research Initiative Competitive Grant
no. 2011-68004-30336 from the USDA National Institute of Food and Agriculture
Feed efficiency
in swine
edited by:
John F. Patience
Wageningen Academic
P u b l i s h e r s
Buy a print copy of this book at
www.WageningenAcademic.com/feedeff
This work is subject to copyright. All rights are

reserved, whether the whole or part of the material
is concerned. Nothing from this publication
may be translated, reproduced, stored in a
computerised system or published in any form or
in any manner, including electronic, mechanical,
reprographic or photographic, without prior
written permission from the publisher,
Wageningen Academic Publishers,
P.O. Box 220, 6700 AE Wageningen,
the Netherlands,
ISBN: 978-90-8686-202-3 www.WageningenAcademic.com
e-ISBN: 978-90-8686-756-1 copyright@WageningenAcademic.com
DOI: 10.3921/978-90-8686-756-1
The individual contributions in this publication
and any liabilities arising from them remain the
First published, 2012 responsibility of the authors.
The publisher is not responsible for possible

© Wageningen Academic Publishers damages, which could be a result of content
The Netherlands, 2012 derived from this publication.
Table of contents
Preface 13
1. Herd management factors that influence whole herd feed efficiency 15

A.M. Gaines, B.A. Peterson and O.F. Mendoza
Introduction 15
Are we measuring feed conversion correctly? 16
How do we measure ourselves over time? 17
Is it time to shift our thinking? 20
What is the best way to measure feed efficiency in the sow herd? 20
Factors in a production system that could impact whole herd feed efficiency 23
Sow replacement rate 23
Timing of mortality 24
Impact of birth weight on feed efficiency 26
The effects of weaning weight on feed efficiency 28
Harvest weight 29
Pig removal strategies at marketing 31
Floor and feeder space impacts on feed efficiency 32
Conclusion 35
References 36
2. Feeding and barn management strategies that maximize feed efficiency 41

M.D. Tokach, R.D. Goodband, J.M. DeRouchey, S.S. Dritz and J.L. Nelssen
Introduction 41
Prior to entry 42
Loading the barn 46
Daily chores 48
Unloading the barn 55
Conclusions 58
References 59
3. Liquid feeding corn-based diets to growing pigs: practical considerations and use of
co-products 63
C.F.M. de Lange and C.H. Zhu
Introduction 63
Design of liquid feeding systems 64
Liquid feeding practices 66
Effects of liquid feeding corn-based feeds on growth performance and carcass
characteristics 68
Use of liquid co-products: corn distillers solubles and corn steep water 71
Use of dry high-fiber co-products: wheat shorts and dried distillers grains with solubles 75
Conclusions and implications 76
Acknowledgements 78
References 78
Feed efficiency in swine 7

Table of contents
4. Amino acid nutrition and feed efficiency 81

C.F.M. de Lange, C.L. Levesque and B.J. Kerr
Introduction 81
Whole body protein deposition and pig growth performance 82
Biology of amino acid utilization in growing pigs 84
Effect of between-animal variability on optimum dietary amino acid levels for
groups of pigs 88
Implications of phase-feeding and compensatory growth for establishing optimum
dietary amino acid levels 90
NRC approach to estimating amino acid requirements of growing-finishing pigs 93
Conclusions and implications 97
References 98
5. The influence of dietary energy on feed efficiency in grow-finish swine 101

J.F. Patience
Introduction 101
Defining and expressing feed efficiency 102
Definition of dietary energy 105
Dietary sources of energy 108
Energy systems 110
Dietary energy used for maintenance and for gain 116
Daily energy intake 120
Other considerations 122
Practical approaches to improving feed efficiency 122
Conclusion and implications 125
References 125
6. Feed processing to maximize feed efficiency 131

C.R. Stark
Introduction 131
Ingredient selection and least cost formulation 132
Feed manufacturing quality assurance 135
Feed manufacturing 138
Feed ordering and delivery 148
Conclusion 149
References 149
7. The genetic and biological basis of residual feed intake as a measure of feed efficiency 153
J.M. Young and J.C.M. Dekkers
Introduction 153
The genetic basis of residual feed intake 154
Physiological basis of residual feed intake 154
Selection experiment in Yorkshire pigs to create lines divergent in residual feed intake 156
Conclusions 163
Acknowledgements 163
References 164
8 Feed efficiency in swine

Table of contents
8. Pig breeding for improved feed efficiency 167

P.W. Knap and L. Wang
Feed efficiency: past developments 167
Feed efficiency: biological backgrounds 169
Genetic change in production traits and feed efficiency 173
Breeding for improved feed efficiency 175
Implications 178
References 179
9. Effect of climatic environment on feed efficiency in swine 183

D. Renaudeau, H. Gilbert, and J. Noblet
Introduction 183
General aspects 184
Consequences of thermal stress on feed efficiency 187
Strategies for alleviating the effects of thermal stress on feed efficiency 196
Conclusion 203
References 205
10. Fueling the immune response: what’s the cost? 211

R.W. Johnson
Introduction 211
Relationship between disease and growth performance 212
How does the immune system sense the pathogenic environment? 215
How does the immune system affect growth? 218
What does it cost to nourish the immune response? 220
Acknowledgements 220
References 220
11. Influence of health on feed efficiency 225

S.S. Dritz
Introduction 225
Direct effects of mortality 226
Chronic immune stimulation 227
Production responses to in-feed antimicrobials in multi-site production 228
Field data 229
Future advances 234
Conclusion 235
References 235
12. Physiology of feed efficiency in the pig: emphasis on the gastrointestinal tract and
specific dietary examples 239
J.R. Pluske
Introduction 239
Secretions from the lactating sow and piglet growth efficiency 239
Growth factors in colostrum and milk 242
Changes in feed efficiency associated with weaning 243

Table of contents
Relationships between dietary protein source and post-weaning diarrhea 247

ZnO as a growth-promoting compound to enhance feed efficiency after weaning 250
Conclusions 251
References 253
13. Emerging technologies with the potential to improve feed efficiency in swine 259
F.R. Dunshea
Introduction 259
Porcine somatotropin 260
Ractopamine 261
Cysteamine 263
Chromium 264
Betaine 266
Dietary neuroleptics 266
Immunization against GnRF 268
Conclusions 269
References 269

Preface
The world as we know it is changing at an accelerated pace. Continued growth of the human
population will put increasing pressure on feed supplies and food production. Coincident with a
rising population is a growing demand for pork, as the standard of living rises in many parts of the
world. Meeting this demand will be a challenge that farmers must face head-on. Pork producers
face other challenges as they adopt new technologies to produce ‘more with less.’ Unprecedented
– and unexpected – growth of the grain biofuels sector in the past decade has upset the traditional
balance of supply and demand in the grain economy. It will take some time for a new equilibrium
to be reached.
The other competitor for feed resources – human food – will also expand as the human population
grows. Nevertheless, the global pork industry has co-existed with the human food complex for
some time, so although it represents another major user of potential feed resources, it is not a
new or unfamiliar competitor.
Other trends, such as competition from other meat sources like poultry, a decline in arable land
in many historically important agricultural regions and uncertainty about the future of irrigation
in arid regions will also place greater demands on the pig industry to use feed resources more
effectively. It was in this context that Feed efficiency in swine was born.
This book evolved from the International Conference on Feed Efficiency in Swine held in
November, 2011. At that event, the speakers were charged with presenting the newest and most
current information available on feed efficiency in swine, covering everything from daily barn
management to the adoption of new technologies. This book has the same objective as that
conference and uses the speakers as its authors.
By bringing together authors with a wide array of backgrounds and roles in the pig industry, the
chapters in this book represent a similar diversity, looking at feed efficiency from perspectives in
the barn as well as in the laboratory. Feed efficiency in swine covers a broad spectrum of scientific
disciplines, including nutrition, genetics, veterinary medicine, physiology, feed processing
and many others. The result is a unique book that provides the reader with an abundance of
information on a variety of approaches to maximizing feed efficiency in the pork industry today.
I want to thank the authors, each of whom enthusiastically accepted the challenge to address
their topic thoroughly and proficiently. They validated their selection as authors by the very high
quality of the chapters they submitted. I also want to thank Abby Anderson, Holly Schuler and
Julie Roberts for their editorial assistance, and convey particular appreciation for the professional
assistance provided by Mike Jacobs and his staff at Wageningen Academic Publishers. The quality
of this book attests to their efforts.
Finally, I want to acknowledge that this book project was supported by Agriculture and Food
Research Initiative Competitive Grant No. 2011-68004-30336 from the USDA National Institute
of Food and Agriculture.
J.F. Patience, editor

1. Herd management factors that influence whole herd feed
efficiency
The Maschhoffs, LLC, 7475 State Route 127, Carlyle, IL 62231, USA; aaron.gaines@pigsrus.net
Abstract
This chapter provides a broad overview of herd management factors that influence whole
herd feed efficiency. As producers continue to look for ways to manage high feed costs, a
significant economic opportunity exists to narrow the gap between actual performance and
genetic potential by focusing on the impact of various herd management factors that could
be impacting feed efficiency. Optimization of these factors will be system dependent, and will
rely heavily on gathering and interpreting information to make informed decisions, which
will ultimately make the swine industry more efficient and competitive in the global protein
production sector. In addition to examining herd management factors that influence whole
herd feed efficiency, we will attempt to challenge old paradigms on how to measure feed
efficiency within a production system.
Introduction
The typical measurement for feed efficiency is feed per unit live weight gain (Feed:Gain) also
known as feed conversion rate. Because feed efficiency is the outcome of both feed intake and
average daily gain, nutritionists tend to focus their attention on the individual factors that impact
feed consumption and gain. An incomplete list of these individual factors would include dietary
energy level, amino acid deficiency, feed availability or wastage, etc. Although these factors are
important from a nutritionist’s point of view, one cannot dismiss other factors in a production
system that perhaps influence feed efficiency to a greater extent including genetics, disease,
environment, and management factors. These factors cannot be evaluated in isolation and must
be considered when evaluating feed efficiency within a production system.
Given today’s high feed cost environment, there has been considerable emphasis placed on feed
efficiency. This is understandable given that feed inputs represent a significant portion of overall
production costs and improving feed efficiency is one strategy to improve cost per unit gain.
Traditionally, this has been done through changes in dietary energy level; however, given current
energy costs there has been a shift to high fiber-low energy diets. This has been prompted by
direct competition for the same feedstocks as the biofuels industry, which has increased the use
of biofuel co-products such as dried distillers grains with soluble (DDGS) and other alternative
ingredients to optimize feed costs. In general, these feedstocks have been found to be lower in
energy as compared to corn. As a result, feed efficiency has inherently gotten worse with diet
formulations utilizing these ingredients.
With the increased use of high fiber diets, there is the potential to reduce carcass yield. This is
well documented in the literature as it relates to feeding high levels of DDGS continuously in the

finishing phase of growth (Stein and Shurson, 2009). As it relates to feed efficiency, this may be
of economic consequence given the fact that most finishing pigs are sold on a carcass basis. Thus,
a reduction in carcass yield inherently increases the feed per unit of carcass gain. Due to the cost
impact of feed efficiency loss, it is imperative that one considers the reduction in carcass yield
when feeding high fiber diets.
The purpose of this chapter is to challenge old paradigms regarding the measurement of feed
efficiency given today’s high feed cost environment, while considering dietary energy levels,
feedstocks being utilized, and how pigs are being sold into the marketplace. In addition, this chapter
will examine factors in a production system beyond nutrition that may impact feed efficiency with
an attempt to quantify the magnitude of these factors on whole herd feed conversion.
Are we measuring feed conversion correctly?
As previously indicated, the typical measurement for feed efficiency is feed per unit gain being
expressed on a live weight basis. Given the fact that most pigs are sold on a carcass basis, there
is a strong argument to measure feed conversion on a carcass basis as well. This is particularly
true when differences exist between live and carcass gain measurements, which can be the case
when feeding high fiber diets. To illustrate this point, we can look at the data collected in our
production system where pigs were fed varying levels of DDGS (Table 1). For example, if one
looks at live feed conversion expressed as G:F ratio, there was a significant difference among the
treatment groups. As compared to the 0% DDGS level, the G:F ratio was 1.3 and 2.1% lower for
pigs fed 15 and 30% DDGS, respectively. However, if one considers the reduction in carcass yield
with increasing DDGS level, the differences are even more apparent. As compared to the 0%
DDGS level, the G:F ratio was 2.1 and 3.8% lower for pigs fed 15 and 30% DDGS, respectively.
In this example, depending on whether one utilizes live vs. carcass feed conversion, there is a
Table 1. Impact of DDGS level on feed efficiency from weaning to market (The Maschhoffs, LLC (Trial#
200918)).
Item DDGS (%) SEM P-value
0% 15% 30%
Start weight (kg) 5.86 5.86 5.81 0.05 NS

End weight (kg) 123.6 123.7 124.2 0.32 NS
Gain:Feed1 0.378a 0.373b 0.370b 0.002 <0.001
Yield (%) 76.4a 75.7b 74.9c 0.14 0.001
Carcass end weight (kg) 94.1a 93.5ab 93.0b 0.25 0.01
Carcass Gain:Feed2 0.291a 0.285b 0.280c 0.002 <0.001
a,b,c Means within a row without a common superscript letter differ (P<0.05).
1 1.3-2.1% difference in feed efficiency when expressed on a live basis relative to 0% DDGS.
2 2.1-3.8% difference in feed efficiency when expressed on a carcass basis relative to 0% DDGS.

1. Herd management factors that influence whole herd feed efficiency
significant impact on the economic decisions within a production system as it relates to the
nutrition program. Thus, when feeding high fiber diets, one should consider the potential impact
on carcass yield and the indirect impact on carcass feed efficiency.
How do we measure ourselves over time?
As we look to measure feed efficiency over time, it is important to recognize changes that could
impact this metric. Some of the changes we have seen in recent years are changes in dietary
energy level, weaning age, market weight, and implementation of feed processing technologies.
If one looks at energy level, there has been a significant decline in dietary energy levels over the
last several years due to rising feed costs and feedstocks being utilized. The industry has also
moved to older wean age pigs in an effort to place heavier pigs at the time of weaning due to
the improvements in growth performance and livability (Dritz et al., 1996; Main et al., 2004).
Additionally, industry harvest weights have seen an upward trend for the past several years. This
has made economic sense because the additional revenue obtained for each incremental unit of
weight was almost always greater than the incremental increase in production costs. In an effort
to manage high feed costs, there has been renewed interest in feed processing technologies by
production systems due to the observable improvements in feed efficiency (Skoch et al., 1983;
Stark et al., 1994; Wondra et al., 1995). Taken together, these changes aimed at system optimization
make interpretation of production and financial records difficult over time, particularly for a
metric such as feed efficiency. Thus, it is important to attempt to account for these changes to
make better sense of production and financial data.
We know that feed efficiency will be influenced by the energy level of the diet, entry and exit
weight of the pigs, and whether the diet is further processed. Adjustment factors can be utilized
to account for these factors and their impact on feed efficiency. Equations have been generated
to allow for feed efficiency adjustments (Goodband et al., 2008). Perhaps the simplest adjustment
as it pertains to these equations is to account for differences in entry and exit weight of a group
of pigs. An example of this type of equation is shown in Equation 1, where adjustments are made
to entry and exit weight (Goodband et al., 2008).
Adjusted Feed:Gain =
observed Feed:Gain + (50 – entry weight) × 0.005 +
+ (250 – market weight) × 0.005 (1)
This equation adjusts all groups to a common entry weight of 50 lbs and an exit weight of 250
lbs. The multiplier of 0.005 represents the slope of the equation describing live feed conversion
vs. body weight on corn-soybean meal based diets with ~5% supplemental fat. The data shown
in Table 2 represents an example of adjusting feeder to finisher feed conversion in closeout data
using this equation. Feed conversion appears to be relatively similar between the growers in this
example. However, when feed conversion is adjusted to account for entry and exit weights, grower
1 has the better G:F. The better adjusted feed conversion rate for grower 1 is due to the heavier pig
weights at entry and exit. Further adjustments can be made to account for different grain sources,
dietary energy levels, and pelleted or meal diets as shown in Equation 2 (Goodband et al., 2008).

Table 2. Finishing closeout-comparison based on 2007 data (Goodband et al., 2008)1.
Item Grower 1 Grower 2
Weight in (kg) 25.9 20.4

Weight out (kg) 123.4 120.2
Mortality (%) 4.9 3.7
Average daily gain (kg) 0.853 0.848
G:F 0.352 0.350
Adjusted G:F 0.376 0.357
1 Closeouts are adjusted for initial and final weight only.
observed Feed:Gain + (50 – entry wt) × 0.005 + (250 – market weight) × 0.005
Adjusted Feed:Gain = (2)
[Grain factor – (fat level × 2)) × (1 – pellet factor)]
Grain factor = 1 for corn, 1.02 for milo, 1.18 for barley, and 1.07 for wheat
Pellet factor = the % improvement in feed efficiency due to pelleting (generally 4 to 6%)
The adjustment for energy level uses an adjustment for grain source and fat level in the diet
[grain factor – (fat level × 2)], where the grain factor is 1 for corn and fat level is the percent
fat in the diet. This equation does assume a 2% improvement in Feed:Gain for every 1% added
fat. The adjustment for pelleting is (1 − pellet factor), where the pellet factor is the percentage
improvement in feed efficiency due to pelleting. Based on the aforementioned empirical data, this
would generally be 4 to 6%. Another equation that we have utilized in our system is proposed
in Equation 3.
Adjusted F:G = ( Observed F:G + (Standardized sw – Actual sw) ×

× Slope estimate + (Standardized fw – Actual fw) × Slope estimate –
– ((Standardized el – Actual el)/ Standardized el) × Observed F:G) (3)
F:G = Feed:Gain ratio; sw = start weight; fw = final weight; el = energy level
An example using this equation is shown in Figure 4 and utilizes a similar slope estimate as
described previously. The importance of understanding the pig’s growth response to dietary
energy within your own system should be emphasized, as it will change the slope estimate. Similar
to the previous equation, further adjustments can be made to account for other factors such as
pelleting, grind size, mortality, etc.
Another factor to consider when adjusting feed efficiency from a nutritional point of view is to
account for differences in mortality, especially in a case where production systems have experienced
high mortality rates due to novel disease introduction. When making this type of adjustment, it is
critical to know the feed conversion impact for each percentage increase in mortality. In Figure 1,
assuming a mid-point mortality during the finishing phase of growth, we have attempted to quantify

Box 1. Example of feed efficiency adjustments standardized for energy level and weight based on Equation 3.
Example:
Observed Feed:Gain = 3.00

Actual start weight = 45 lbs
Actual end weight = 250 lbs
Actual energy level = 1,500 kcal/lb ME
What is the adjusted Feed:Gain if we standardize for energy level and weight?
2.71 = (3.00 + (50 – 45) × 0.005 + (250 – 275) × 0.005 – ((1,600 – 1,500) / 1,600) × 3.00)
FTF adjusted for mortality (live)

0.0108
0.0107
y = 0.00000801x + 0.00829289
0.0106 R2 = 1.00000000
Feed conversion rate
0.0105
0.0104
0.0103
0.0102
0.0101
0.0100
0.0099
0.0098
200 220 240 260 280 300 320
Final body weight (lbs)
Figure 1. Feed efficiency impact for each percentage increase in mortality for feeder to finish pigs at varying
final live body weights.1
1 Assuming the mortality occurred during the mid-point of the growth phase (i.e. 60 lbs to final live body weight).
the impact on feed conversion at different marketing end weights. It is important to note that this
type of mortality adjustment to feed conversion does not account for the disease impact and assumes
the mortality occurred during the mid-point of the growth period. The data in Table 3 represents an
example of adjusting feeder to finisher feed conversion for mortality differences in closeout data. At
the mortality rates incurred by grower 1 and 2, the adjustments in G:F were 1.6 and 1.1%, respectively.

Table 3. Finishing closeout-comparison (2007 data) with mortality adjustment (Goodband et al., 2008)1.
Item Grower 1 Grower 2
Weight in (kg) 25.9 20.4

Weight out (kg) 123.4 120.2
Mortality (%) 4.9 3.7
Average daily gain (kg) 0.853 0.848
G:F 0.352 0.350
Adjusted G:F 0.376 0.357
Mortality adjusted G:F2 0.382 0.361
1 Closeouts are adjusted for initial and final weight only.
2 Adjustment is based on mortality impact on G:F.
Is it time to shift our thinking?
Unless something changes to energy costs in the foreseeable future, it appears that dietary energy
levels will continue to decline. As previously discussed, this makes it difficult to interpret feed
conversion data over time as it is confounded due to energy level among other factors. Adjustment
factors can be used to account for these differences in energy level over time, but perhaps it is
time to shift our thinking more dramatically in regards to measuring feed efficiency. Instead
of measuring feed conversion, there is a case to simply utilize caloric efficiency in its place as a
measurement of feed efficiency. This metric considers the amount of calories required per unit of
gain, regardless of the dietary energy level fed. In Figure 2, we compare caloric efficiency vs. feed
efficiency for pigs expressed on a carcass basis during the grow-finish period. As one evaluates
whether to utilize this metric, it will require an understanding of how the pig responds to dietary
energy to establish defined targets for the various production phases. Furthermore, it will be
system specific depending on the energy system utilized. To account for other factors (i.e. feed
processing), adjustment factors may also be necessary.
What is the best way to measure feed efficiency in the sow herd?
A common practice for sow farms is to measure sow feed per year with a typical target of 1,000
to 1,090 kg/sow/year; however, this metric is not a good indicator of sow herd feed efficiency.
Thus, increasingly more production systems are utilizing sow feed per weaned pig produced,
with targets based on sow productivity. For example, to achieve 36 kg of sow feed per weaned
pig, a sow farm would need to be at 29 pigs/sow/year (P/S/Y) or better, assuming diets are fed
in meal form, as shown in Figure 3. To take this one step further, one could measure sow feed
per unit of live weight produced as shown in Figure 4. This, of course, requires the sow farm
to weigh the pigs at the time of weaning. If a production system wants to express this on a
carcass feed efficiency basis, a yield factor on the wean pig would need to be applied (i.e. 65%
of live weight).

PIC energy eff. PIC FCR

5.0 16,000
4.57
4.5 4.41 15,500
4.25
4.12
4.0 3.96 15,000
3.83
14,500
3.5
14,000
kcal/kg
3.0
F/G
15,692 13,500
15,149
2.5 14,606 13,000
14,154
2.0 13,612 12,500
13,159
1.5 12,000
1.0 11,500
193 204 214 225 235 245
Carcass end weight (lbs)
Figure 2. Caloric efficiency vs. feed efficiency expressed on a carcass basis.1

1 Source data: PIC ES 051-Tech Memo. 344; cumulative data from feeder pig wt (34.7 kg live body weight); 3,437 kcal/kg
NRC ME; 76% yield factor.
50
45
Sow feed per wean pig (kg)
40
35
30
25
20
15
10
5
0
25 25.5 26 26.5 27 27.5 28 28.5 29 29.5 30
Pigs/sow/year
Figure 3. Impact of sow productivity on sow feed per wean pig.1

1 Assuming 1,052 kg annual sow feed usage (i.e. gestation @ 2.27 kg/day; lactation 5.90 kg/day) and litter/sow/year = 2.45.

Sow feed per wean kilograms produced

8
7
6
5
4
3
2
1
0
25 25.5 26 26.5 27 27.5 28 28.5 29 29.5 30
Pigs/sow/year
Figure 4. Sow feed per wean kilograms produced (live weight basis).1
1 Assuming 1,052 kg annual sow feed usage (i.e. gestation @ 2.27 kg/day; lactation 5.90 kg/day) and litter/sow/year = 2.45.
When examining whole herd feed efficiency, the productivity of the sow herd directly determines
the number of pigs that sow feed use and costs can be spread over. Whole herd feed efficiency for
the sow can be measured using sow feed per pig marketed to include gilt development and boar
feed. As shown in Figure 5, sow productivity moderately impacts whole herd feed carcass feed
efficiency. For example, increasing P/S/Y on a given farm from 25 to 30 P/S/Y improves whole
herd carcass feed efficiency by 1.7%. Although this seems to be a small improvement given a 20%
increase in sow productivity, one needs to be mindful that sow feed only represents 10 to 12% of
whole herd feed efficiency.
Whole herd carcass FCR

4.06
4.04
4.02
4.00
3.98
3.96
3.94
3.92
25 25.5 26 26.5 27 27.5 28 28.5 29 29.5 30
Pigs/sow/year
Figure 5. Impact of sow productivity on whole herd carcass feed efficiency, assuming 1,052 kg annual sow
feed usage plus gilt development feed (i.e. 152 kg/sow) and market weight of 90 kg carcass weight.

Factors in a production system that could impact whole herd feed efficiency
Modern commercial pigs in production systems throughout the world have a much improved
potential for growth and a much better ability to convert feed to carcass gain than their past
counterparts. As it relates to improvements in feed efficiency, there has been continuous
genetic selection aimed primarily at increasing lean deposition while reducing the level of fat
in the body. Despite the genetic capability of today’s pig, most production systems struggle to
capture the full genetic potential of their animals. For example, we have found in our research
facilities that the energetic efficiency potential can be ~8% better than pigs reared under
commercial conditions (Figure 6). Interestingly, our research facilities are akin to those used
in the rest of our production system. Thus, an economic opportunity exists to narrow the gap
between actual and potential performance. In the following subsections, we will take a closer
look at some factors in a production system that should be considered when examining whole
herd feed efficiency.
Sow replacement rate
A high sow replacement rate has a direct effect on whole herd feed efficiency due to the number
of gilts needed to maintain mating volume. When producers retain more gilts to maintain
mating volume, whole herd feed conversion inherently gets worse because of the higher herd
feed consumption and (or) reduced output of pigs. Gilts generally have lower reproductive
performance (i.e. total born and farrowing rate), which lowers the number of pigs weaned.
Furthermore, mortality is typically higher for gilt vs. sow litters from birth to market as shown
in Table 4 by Smits and Collins (2009). The lower survivability of gilt offspring is likely due to
immune competence and their higher susceptibility to disease. Given what we know about the
16,000
15,000
14,000
13,000
kcal/kg gain
12,000
11,000
10,000
9,000 Research
Commercial
8,000
22.1 34.9 49.3 64.3 79.4 93.8 107.2 119.6 130.8
Carcass end weight (kg)
Figure 6. Cumulative carcass energy/kg gain for pigs reared in a commercial vs. research environment,
assuming PIC 337 × Camborough pigs and actual entry weight of 22 kg carcass weight.

Table 4. Differences between gilt and sow progeny (Smits and Collins, 2009).
Treatment Preweaning Weaner Grower Finisher
Progeny of gilts1 14.4% 5.6% 3.9% 6.7%

Progeny of sows1 11.2% 2.7% 2.2% 4.8%
P-value χ2 5.2 (0.022) χ2 10.9 (0.001) χ2 4.2 (0.041) χ2 2.5 (0.116)
1 Differences between gilt and sow (parity 3-7) progeny in respect of mortality and removal for ill thrift expressed as
a percentage of the number of pigs at the start of each growth phase.
gilt and her offspring, lowering sow replacement rate will improve whole herd feed conversion,
which is supported by Smits (2011) and shown in Table 5. These data compare two different sow
farms with differing replacement rates (i.e. 65 vs. 40%). By lowering the replacement rate, herd
feed conversion is improved by 2%. If one considers the direct effects of sow replacement with the
indirect effects of a reduced proportion of gilt progeny in the herd, feed conversion is improved
by 3%, which is shown in Table 6 (Smits, 2011). Taken together, management of the sow herd
replacement rate clearly has direct and indirect effects on the performance of their progeny and
consequently, on whole herd feed conversion.
Timing of mortality
Generally speaking, the focus on pig mortality has been centered on the period from farrowing to
weaning, as it has been well established that losing pigs during this critical period will negatively
affect the efficiency of a production system. It is during this period that the majority of the
research has been carried out and the causes and strategies to mitigate mortality have been well
defined. However, such is not the case during the grow-finish period, as there is limited data to
bring perspective to the specific effect(s) of pig mortality throughout the grow-finish period.
Nevertheless, the timing at which pig mortality occurs during the grow-finish period has a
Table 5. Predicted response in annual reproductive output and herd feed conversion efficiency (Smits, 2011).
Item1 Pigs Pigs Pigs born Progeny Cull carcass Progeny carcass HFC2
weaned mated alive sold weight (t) weight (t)
Scenario 1 18,720 2,349 21,767 15,420 127.2 1,170.6 3.86

Scenario 2 18,720 2,176 21,273 15,731 96.7 1,194.2 3.78
1 Scenario 1: 65% replacement rate, base production levels, 26% gilt matings, 82% farrowing rate, 11.3 born alive, 14%
preweaning mortality, 17% breeder mortality, culled sow weight = 175 kg hot standard carcass weight. Scenario 2:
40% replacement rate, 14% gilt matings, 85% farrowing rate, 11.5 born alive, 12% preweaning mortality, 8% breeder
mortality, culled sow weight = 210 kg cwt.
2 HFC = herd feed conversion.

Table 6. Predicted response in annual reproductive output and herd feed conversion efficiency (Smits, 2011).
Item1 Pigs Pigs Pigs Progeny Cull carcass Progeny carcass HFC2
weaned mated born alive sold weight (t) weight (t)
Scenario 1 18,720 2,349 21,767 15,420 127.2 1,170.6 3.86

Scenario 2 18,720 2,176 21,273 15,731 96.7 1,194.2 3.78
Scenario 3 18,720 2,176 21,273 15,914 96.7 1,217.7 3.74
1 Scenario 1: 65% replacement rate, base production levels, 26% gilt farrowings, 82% farrowing rate, 11.3 born alive
(gilts = 10.2, sows = 11.7), 14% preweaning mortality, 17% breeder mortality, culled sow weight = 175 kg cwt. Scenario
2: 40% replacement rate, 14% gilt farrowings, 85% farrowing rate, 11.5 born alive, 12% preweaning mortality, 89%
weaned progeny from parity 2+ sows, 12% breeder mortality, culled sow weight = 210 kg cwt. Scenario 3: As for
Scenario 2 plus an increase of 0.8 kg live weight at sale and a 0.05% reduction in weaner-grower mortality rate.
2 HFC = herd feed conversion.
significant impact on efficiency of a production system, such that losing pigs at a heavier body
weight has a bigger toll on production costs (Maes et al., 2001). In addition, as pigs die at heavier
body weights, the feed conversion ratio will be impacted more negatively for every 1% increase in
mortality, compared to pigs that die at a lighter body weight (Figure 7). Given the impact of pig
mortality on herd feed efficiency it is imperative that production systems understand the timing
and causative factors of the mortality in order to develop health management plans to minimize
these losses.
0.017
Feed:Gain
0.011
0.005
52 77 102
Live body weight (kg)
Figure 7. Impact of a 1% increase in mortality on feed conversion of pigs at different body weights, assuming
actual entry weight of 27 kg live weight and market carcass weight of 127 kg.

Impact of birth weight on feed efficiency
Variation in individual pigs’ ability to convert feed to body tissues is caused by multiple factors,
some of which are environmental in nature and some genetically programmed. One such factor
is the weight of the pig at birth. In all litter bearing species, variation exists between littermates in
all phenotypic traits at birth, and possibly the most notable is body weight. Birth weight has been
studied and reported on extensively in the literature and the reported effects on feed efficiency
are variable.
While genetic diversity is a fact of nature, many environmental factors during the prenatal period
can have a profound impact on the overall growth potential and body composition of a pig. A
brief discussion of the prenatal environment and its impact on variation in birth body weight
within and among litters is warranted to understand the causes and extent of variation that exists
in pig birth weight. Birth weight is determined in utero as a direct result of fetal nutrition. Fetal
nutrition can be affected by a variety of factors; however, when all else is held constant (sow
nutrition, disease status, etc.), variation in birth weight within the litter remains. Fetal nutrition
can be influenced by many factors other than the level and quality of feed that the sow consumes.
The number of fetuses and the position of the fetus within the uterus also largely influence the
level of nutrition that the individual fetus receives. It has been reported that as litter size increases,
average birth weight decreases (Johnson et al., 1999), due to the effects of intrauterine crowding
and the resultant increase in competition among the fetuses for maternal nutrients. It has been
known for some time that intrauterine crowding impacts fetal growth rate (Dziuk, 1968), and
the effect of crowding has been shown to manifest as early in gestation as day 30 as evidenced
by the work of Foxcroft et al. (2006). This work reported a significant decrease in myogenesis at
30 days of gestation associated with higher levels of intrauterine crowding, which could result in
lower capacity for lean growth during the pig’s life. The effects of intrauterine crowding and fetal
under-nutrition have the greatest effect on the smallest fetus in the uterus because the smallest
fetuses generally have the least amount of placental surface area attachment within the uterus.
This results in fewer nutrients from the sow being passed on to the smaller fetuses and therefore
slower prenatal growth rates. Therefore, the effects of increased nutrient intake of the sow could
have the greatest impact on the smallest fetuses, as the larger ones are already receiving adequate
nutrition for growth and development. This theory is supported by the findings of Dwyer et al.
(1994), who reported that increased sow nutrition during gestation was found to produce the
greatest relative increase in body weight in the smallest fetuses. Genetic improvement in litter size
has been quite successful and the result has been very large litters and highly productive sows. As
shown in Figure 8, the impact these larger litters have on birth weight must also be considered
in the production system economics equation, and genetic progress must provide for increased
uterine capacity to avoid negative impacts on pig birth weight as litter sizes continue to increase.
A great deal of research has been completed studying the effects of birth weight on pig growth
rate, feed intake, feed efficiency, and carcass characteristics. A vast majority of the studies have
reported that light birth weight pigs have a lower post-weaning growth rate; however, the results
for feed efficiency effects have been mixed. Multiple studies have reported a reduction in average
daily gain in light birth weight pigs compared to heavy birth weight pigs as a result of lower feed
intake with no impact on feed efficiency (Gondret et al., 2005; Quiniou et al., 2002; Wolter et al.,

2.0
1.8
1.6
1.4
Birth weight (kg)
1.2
1.0
0.8
0.6
0.4
0.2
0.0
13.2 15.4 17.6 19.8 22.1 24.3 26.5 28.7 30.9 33.1 35.3 37.5 39.7 41.9
Pigs/sow/year
Figure 8. Impact of pigs/sow/year on birth weight (Source: the Maschhoffs).
2002a). In contrast, a number of studies have reported light birth weight pigs have poorer feed
efficiency (Powell and Aberle, 1980; Gondret et al., 2006; Peterson, 2008) and Mroz et al. (1987)
reported, after conducting digestibility research, that light birth weight pigs also utilized nutrients
(i.e. energy, nitrogen, etc.) less efficiently than heavy birth weight pigs.
One possible explanation for the poorer feed efficiency observed in the literature is a reduction
in the light birth weight pigs’ overall capacity for lean tissue accretion. Muscle fiber number has
been found to decrease with birth weight in a number of studies (Nissen et al., 2003; Bee, 2004;
Gondret et al., 2006). Handel and Stickland (1987) and Dwyer et al. (1994) reported that the
number of primary muscle fibers was equal among pigs of different birth weights; however, the
number of secondary fibers was found to be lower in lighter birth weight pigs. Secondary muscle
fibers develop early in gestation, usually around day 25 to 30. As previously stated, the work of
Foxcroft et al. (2006) reported a significant decrease in myogenesis during this time associated
with intrauterine crowding, which would suggest that fetal under-nutrition is a possible cause
of depressed secondary muscle fiber development. Since the number of muscle fibers in pigs
is set at birth and subsequent muscle growth is a result of fiber hypertrophy, light birth weight
pigs could therefore have lower capacity for total lean tissue growth resulting in the deposition
of proportionally more adipose tissue than heavy birth weight pigs. This is further supported by
the work of Peterson (2008) that suggested a decrease in fat free lean percentage in light birth
weight pigs compared to heavy birth weight pigs at a common harvest weight (Figure 9). These
differences in carcass composition could ultimately result in poorer feed efficiency as adipose
tissue is more energy dense than lean and thus requires more nutrients per unit of weight to
accrete. It is well established that light birth weight pigs grow slower than their heavy birth
weight counterparts (Peterson, 2008; Table 7). When considering overall efficiency of growth
to a common body weight, light birth weight pigs will also undoubtedly be slightly less efficient
due to the extra maintenance energy required during the added days of growth. Future research

58
Carcass fat-free lean percentage

56
54
52
50
Heavy Medium Light
Figure 9. Carcass fat-free lean percentage for heavy, medium, and light birth weight pigs (Peterson, 2008).
Table 7. Effect of birth weight and weaning weight on overall growth performance (Peterson, 2008).
Item Birth weight category Weaning weight category
Heavy Light SEM1 P-value Heavy Light SEM1 P-value
Body weight (kg)

Birth 1.75a 1.31b 0.054 0.001 1.53 1.53 0.054 0.768
Weaning 7.56a 6.71b 0.377 0.011 7.41 6.87 0.377 0.090
Market 125.4 124.1 0.60 0.145 124.8 124.8 0.60 0.953
Weaning to market
Average daily gain (g) 877a 844b 11.0 0.002 859 862 11.0 0.719
Average daily feed intake (kg) 2.08 2.04 0.021 0.142 2.06 2.06 0.021 0.849
Gain:Feed 0.422 0.415 0.0027 0.056 0.417 0.420 0.0027 0.430
1 SEM = standard error of the mean.
a,b Means within a row without a common superscript letter differ (P<0.05).
should be aimed at finding management strategies that either reduce or exploit this variation in
birth weight in an effort to improve overall efficiency of production.
The effects of weaning weight on feed efficiency
The assertion that heavier pigs at weaning will grow faster and require fewer days to reach market
weight has been supported in a number of studies (Boaz and Elsley, 1962; McConnell et al.,
1987; Mahan and Lepine, 1991; Klindt, 2003). When considering the impact of weaning weight

on overall growth performance and carcass composition, the cause of the variation in weaning
weight must first be determined. Two major factors can create variation in weaning weight and
the resulting impact on growth performance is very different. The first is variation in pig potential
for growth. This can often times be attributed to the pig’s birth weight, as birth weight effects on
growth rate are evident at weaning (Peterson, 2008; Table 7). The subsequent impact on overall
growth performance and carcass composition due to differences in weaning weight caused by
differences in birth weight are similar to that described in the previous section. The second reason
for weaning weight variation is environmental impact on pre-weaning growth rate, such as access
to nutrients. Reduction in weaning weight due to lower nutrient consumption during lactation
can be compensated for after weaning if the pig is placed on proper nutrition that meets the pig’s
nutrient requirements for growth. Wattanakul et al. (2007) conducted a study wherein weaning
weights of entire litters were reduced by limiting access to the sow and thereby reducing suckling.
The authors reported that pigs with lower weaning weights due to reduced suckling actually
grew faster, consumed more feed, and had greater Gain:Feed ratios during the first 28 days post-
weaning when compared to the non-limited pigs, and growth performance was similar for the
remainder of the growth period. Similarly, Peterson (2008) reduced weaning weight of pigs with
similar birth weights by rearing pigs in litters of either 6 or 12 pigs. The resulting difference in
body weight at weaning was short lived, and the restricted pigs (reared in litters of 12) actually
grew faster after weaning, resulting in no difference in overall growth performance (average daily
gain, average daily feed intake, and gain to feed ratio) to a common harvest body weight (Table
7). These causes of weaning weight variation must be carefully considered when implementing
practices that differentially manage pigs at weaning based on body weight, as the reason for the
body weight variation may not be perfectly evident. It also must be noted that these points only
consider the growth performance of the pigs. Heavier, more robust pigs at weaning should have
a better chance of surviving the stresses that are encountered in commercial production systems
immediately after weaning. Higher levels of mortality during the nursery period negatively
affect whole herd feed conversion and profitability; thus robustness at weaning must always be a
consideration for a production system.
Harvest weight
The harvest weight at which pigs are marketed in the industry has been consistently increasing.
The reasons for this are that marketing pigs at heavier harvest weights can be a practical
means of reducing costs in a production system by producing more pounds of pork per sow or
maintaining the same level of production using fewer sows (Gil and Knowles, 2003). In addition,
producers are able to take pigs to heavier weights by taking advantage of the increased lean
deposition rates of modern genetic lines, and still producing pork with acceptable carcass and
meat quality characteristics. However, if increasing the harvest weight of pigs is a strategy that
is being considered, there are several important factors that need to be taken into account to
achieve optimum production efficiency. From a live animal performance standpoint, increases
in harvest weight of pigs above 100 kg has proven to have disadvantages mainly due to significant
reductions in the efficiency of converting feed to lean meat (Richmond and Berg, 1971; Latorre
et al., 2004). Other important factors that need to be considered are genetics, gender, in-barn
management practices, feed and housing costs, desired carcass and meat quality characteristics,
customer specifications, etc. For these reasons, the choice of the harvest weight needs to be

tailored specifically to each situation for a given production system. In general, growth rate is
decreased with increasing harvest weight, and according to Weatherup et al. (1998) and Latorre
et al. (2004) feed efficiency of finishing pigs is impaired by 0.01 kg for every 10-kg increase in
harvest weights up to ~130 kg in mixed-gender groups (Table 8). From a biological standpoint,
this happens because as pigs get heavier, the body composition changes, such that the proportion
of the deposition of body fat relative to muscle increases. Because fat requires more energy to
deposit than muscle, the quantity of feed per unit of gain increases with body weight (Figure 10).
Although there is value to be captured by taking pigs to a heavier harvest weight, it is important
that the right genetics, along with correct nutritional, management and marketing strategies are
considered to maintain profitability for a production system.
Table 8. Effect of harvest weight on the growth performance of market weight pigs (Latorre et al., 2004).
Item Harvest weight (kg) SEM1 P-value
116 124 133
Start weight (kg) 74.9 74.7 74.8 0.43 0.94

End weight (kg) 116.2c 124.4b 133.5a 1.30 0.001
Average daily gain (g) 843a 788b 769b 18.5 0.05
Average daily feed intake (kg) 2.69 2.56 2.68 0.056 0.23
Gain:Feed 0.313a 0.309a 0.287b 0.003 0.001
5.0
Fat:Muscle Feed:Gain
4.5
Fat:Muscle and Feed:Gain
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
23-45 45-68 68-91 91-113
Weight category (kg)
Figure 10. Fat to muscle ratio deposition and feed conversion ratio of grow-finish pigs across 4 weight (kg)
categories (adapted from Pond and Manner, 1974).

Pig removal strategies at marketing
When the pigs in the barn are getting close to harvest weight, pig producers have the challenging
task of consistently selecting and shipping pigs at an ideal body weight for economic optimum.
However, due to the large amount of body weight variation in a population of pigs at marketing,
it is often advocated to remove pigs at the time of market over time, selecting the heaviest animals
first, thereby allowing additional time and resources (floor and feeder space) for lighter pigs
to reach an acceptable market weight (DeDecker, 2006). Several market strategies have been
investigated with positive results in growth rate, largely caused by an increase in feed intake
(DeDecker et al., 2005; Bates and Newcomb, 1997); however, there have been instances in which
feed efficiency has also been improved which is a result of increased feed intake and improved
feed utilization by the remaining pigs in the pen (Table 9). An additional benefit when using
the right marketing strategy is the potential to reduce the total body weight variation of pigs
marketed compared to pens remaining intact, thus allowing the producer to reduce the sort loss
by shipping fewer pigs outside of the customer’s marketing grid. The best marketing strategy will
be that which maximizes economic returns, but it is important to note that it will be specific to
each production system situation. The point to keep in mind is that there is value in selecting the
right pig at the right time during marketing as it relates to herd feed efficiency.
Table 9. Effect of removal strategy at slaughter on body weight and growth performance of pigs reared within
a wean-to-finish system (DeDecker, 2002).
Item Removal treatment1 SEM2 P-value
0% 25% 50% 50% - space
Weight (kg)
Before removal 113.0 113.7 113.3 113.7 0.57 0.769
Start of test 113.0a 110.5b 105.8c 106.8c 0.65 0.001
End of test 126.0a 126.5a 122.2b 121.5b 0.74 0.001
Within-pen CV (%)
Before removal 9.31 9.47 9.60 9.45 0.37 0.959
Start of test 9.31a 8.60a 6.85b 7.86ab 0.48 0.016
End of test 9.43a 8.35ab 6.84b 8.05ab 0.54 0.032
Overall performance
Average daily gain (g) 659c 829a 834a 754b 25.5 0.001
Average daily feed intake (g) 2,795b 3,133a 3,036a 2,855b 51.0 0.001
Gain:Feed 0.24b 0.26a 0.28a 0.26a 0.006 0.001
1 Removal treatment = 0% = 52 pigs/pen, 0 removed, 0.65 m2 floor space and 4.0 cm feeder space/pig; 25% = 25%
removal, 39 pigs/pen, 13 removed, 0.87 m2 floor space and 5.4 cm feeder space/pig; 50% = 50% removal, 26 pigs/
pen, 26 removed, 1.30 m2 floor space and 8.0 cm feeder space/pig; 50%-space = 50% removal and reduced space
to that of Control.

Floor and feeder space impacts on feed efficiency
Pen resource allocation is an important consideration for modern pork production systems. Next
to feed, housing is the second largest cost center for most modern, integrated swine production
systems and therefore optimizing floor and feeder stocking rates is a critical component of
profitability.
The optimum stocking rate for a facility should maximize throughput without compromising
feed efficiency or increasing morbidity and mortality. Determining the optimum stocking rate
requires a dynamic evaluation of the biological impact floor space allocation has on the pig. Floor
space requirements must also be evaluated throughout the pig’s life because the requirement
changes as the pig grows.
Floor space requirements during the traditional nursery period (weaning to approximately eight
to 10 weeks post-weaning) are much lower than those during the finishing phase. Restricted
floor space during the nursery period has been reported to reduce growth rate; however, after the
restriction is removed (when pigs are moved to a finisher or removed from an over-stocked wean
to finish facility) the pigs will generally exhibit compensatory growth resulting in no impact on
overall growth rate or feed efficiency (Wolter et al., 2003a).
Multiple research studies have reported a reduction in average daily gain with decreasing floor
space allowances during the weaning to finish (Wolter et al., 2003b; Table 10) and finishing
production periods (Kornegay and Notter, 1984; Gonyou et al., 2006; Shull, 2010). Many of these
studies have reported this reduction in growth rate to be a result of lower daily feed intake and
some researchers have reported a reduction in feed efficiency (NCR-89, 1993; Hyun et al., 1998).
Another interesting finding in multiple research studies is a reduction in back fat associated
with lower floor spaces (Ward et al., 1997; Brumm et al., 2001; Matthews et al., 2001). When
environmental factors restrict the growth rate of pigs, the result is increased time required to reach
Table 10. Effects of floor space on wean-to-market performance (Wolter et al., 2003b).
Item Floor space1,2 SEM3 P-value
High Low
Start weight (kg) 5.5 5.5 0.01 NS

End weight (kg) 119.3 118.2 0.34 0.06
Average daily gain (g) 671 662 2.4 0.01
Average daily feed intake (g) 1,737 1,699 9.6 0.01
Gain:Feed 0.386 0.390 0.001 0.06
1 Floor space = high (0.63 m2) and low (0.31 ft2).
2 Treatment duration = 12 or 14 week postweaning of floor space restriction.

a certain body weight accompanied by an increase in the overall requirement for maintenance
energy. Therefore, it could be concluded that lowering the floor space provided to a pig and
thus reducing its average daily gain would have an adverse affect on the pigs feed efficiency by
increasing the total maintenance energy required compared to an unrestricted pig. However, this
is not the case in many studies, and is likely due to the fact that pigs housed at lower floor spaces
often have leaner carcasses. This reduction in efficiency due to slower growth rates at lower floor
spaces is likely offset by the increase in efficiency associated with proportionally more lean tissue
accretion in these slower growing pigs. This is obviously a fine balance and should be carefully
evaluated within genotypes, nutrition programs, and housing systems in order to determine
optimum stocking rates.
Lowering the amount of feeder space provided to pigs has been shown to have a similar effect
as lowering floor space. Generally, increasing the number of pigs per feeder space will reduce
feed intake and reduce average daily gain (Wolter et al., 2002b; Peterson et al., 2008; Table 11).
Similar to floor space reduction, feeder space reduction has also been shown to reduce back fat
(Peterson et al., 2008). Many different types of feeders exist in the industry and each one will
have a different optimum stocking rate. If a feeder is designed with the proper dimensions to
facilitate the natural feed intake behaviors of the pig, increasing the stocking rate on the feeder
should not have a negative impact on feed efficiency until average daily gain has been reduced
to the point that the overall maintenance energy requirement increase offsets the decrease in
body fat. Increased competition at the feeder should also be considered, especially with feeder
designs that may not be as ergonomically suited to the pig, as it could increase the amount of feed
wastage when pigs are competing for the same feeder hole. In general, feeder space is a relatively
inexpensive addition and can have a significant return on investment if it allows an increase in
throughput within a facility.
Table 11. Effects of feeder trough space on nursery growth performance (Wolter et al., 2002b).
Item Feeder trough space1 SEM2 P-value
Unrestricted Restricted
Start weight (kg) 5.4 5.4 0.01 NS

End weight (kg) 31.7 30.9 0.22 <0.05
Average daily gain (g) 474 461 3.9 <0.05
Average daily feed intake (g) 792 789 11.6 NS
Gain:Feed 0.60 0.58 0.006 NS
1 Feeder space = unrestricted (4 cm/pig) and restricted (2 cm/pig).

Out of feed events
Feed outages, or out-of-feed events, are a concerning problem in pig production systems. There
are a number of reasons for out-of-feed events at the farm, either caused by the producer or by
malfunctions in the equipment itself. In some instances, problems at the feed manufacturing
facility can also disrupt timely feed delivery to the farm. This can lead to potential problems,
some of which are documented in the literature: decreased growth rates, increased incidence of
gastric ulcers, ileitis, and hemorrhagic bowel syndrome, etc. (Brumm et al., 2005). In addition,
the production efficiency of the feed mill can be reduced as the number of feed emergency orders
is increased, disrupting planned mill production. Research carried out in this area, although
limited, shows that growth rate can be negatively impacted as a result of simulated repeated out-
of-feed events (20 h in duration) especially in the earlier phases of the growth period, whereas
pigs in the finisher phase are less affected. However, in the published research, feed efficiency of
grow-finish pigs does not seem to be impacted in spite of repeated out-of-feed events (20 h in
duration with a weekly, or up to three times per two week period frequency) if pigs are given
access to feed and enough time to recover after a feed outage (Linneen et al., 2007; Brumm et
al., 2008; Table 12 and 13). It should be borne in mind that in the research reported herein, each
of the 20 hours feed outages was conducted from 12:00 to 08:00; this does not coincide with
Table 12. Frequent out-of-feed events on the growth performance of growing-finishing pigs – weekly (Brumm
et al., 2008).
Item Out of feed events SEM1 P-value
Never Weekly
Body weight (kg)

Day 0 24.1 23.3 0.3 0.07
Day 53 70.4 65.9 0.7 0.001
Day 109 118.8 114.1 1 0.007
Average daily gain (kg)
Day 0 to 52 0.873 0.805 10 <0.001
Day 53 to 109 0.863 0.859 9 0.74
Day 0 to 109 0.867 0.832 8 0.008
Average daily feed intake (kg)
Day 0 to 52 2.00 1.87 0.026 0.003
Day 53 to 109 2.98 2.93 0.027 0.24
Day 0 to 109 2.51 2.42 0.024 0.02
Gain:Feed
Day 0 to 52 0.435 0.430 0.002 0.14
Day 53 to 109 0.290 0.293 0.002 0.29
Day 0 to 109 0.346 0.345 0.002 0.56

Table 13. Frequent out-of-feed events on the growth performance of growing-finishing pigs – 1 to 3 times
biweekly (Brumm et al., 2008).
Item Out of feed events SEM1 P-value
0× 1× 2× 3× Treatment Linear
BW (kg)
Day 0 17.8 18.1 18.6 17.8 0.4 0.38 0.76
Day 56 64.7 64.3 63.6 59.9 1.2 0.05 0.01
Day 112 117 117.6 117.5 113.8 1.2 0.13 0.10
Average daily gain (kg)
Day 0 to 55 838 826 803 753 16 0.02 0.003
Day 56 to 112 934 953 962 963 14 0.46 0.14
Day 0 to 112 887 888 883 857 8 0.08 0.03
Average daily feed intake (kg)
Day 0 to 55 1.87 1.85 1.81 1.68 0.045 0.04 0.01
Day 56 to 112 3.19 3.12 3.21 3.16 0.049 0.6 0.97
Day 0 to 112 2.53 2.49 2.51 2.42 0.04 0.3 0.11
Gain:Feed
Day 0 to 55 0.449 0.447 0.445 0.449 0.005 0.91 0.94
Day 56 to 112 0.291 0.302 0.297 0.302 0.004 0.18 0.11
Day 0 to 112 0.35 0.358 0.353 0.355 0.003 0.41 0.52
the period of highest feeding activity generally observed in pigs in commercial practice and
interpretation of this data must be done cautiously. The limited data in this area warrant further
research under commercial conditions to fully understand the effect of out-of-feed events on the
growth performance of grow-finish pigs.
Conclusion
With today’s high feed cost environment, there will continue to be a lot of emphasis placed on
feed efficiency. Thus, it is important that we measure feed conversion correctly. Due to the fact
that most pigs are sold on a carcass basis, there is a strong argument to express feed efficiency on
a carcass instead of a live basis. This becomes increasingly more important given the increased
usage of low energy-high fiber diets. Producers should consider adjusting feed conversion
data for factors such as body weight, energy level, and feed processing to improve system level
interpretation and analysis. Alternatively, one might consider utilizing caloric efficiency to
measure feed efficiency. When measuring feed conversion in the sow herd one needs to include
the impact of sow productivity. As producers continue to look for ways to manage high feed
costs, a significant economic opportunity exists to narrow the gap between actual performance

and genetic potential by focusing on the impact of various herd management factors such as,
sow replacement rate, mortality, birth weight, weaning weight, floor space allowance, marketing
strategy, and out-of-feed events that could be impacting feed efficiency.
References
Bates, R. O., and M. D. Newcomb. 1997. Removal of market ready pen mates improved growth rate of
remaining pigs. J. Anim. Sci. 75(Suppl 1):247 (Abstr).
Bee, G. 2004. Effect of early gestation feeding, birth weight, and gender of progeny on muscle fiber
characteristics of pigs at slaughter. J. Anim. Sci. 82:826-836.
Boaz, T. G., and F. W. H. Elsley. 1962. The growth and carcass quality of bacon pigs reared to different weights
at 56 days old. Anim. Prod. 4:1-24.
Brumm, M. C., M. Ellis, L. J. Johnston, D. W. Rozeboom, and D. R. Zimmerman. 2001. Interaction of swine
nursery and grow-finish space allocations on performance. J. Anim. Sci. 79:1967-1972.
Brumm, M., J Marchant-Forde, R Marchant-Forde, and B. Richert. 2005. Out-of-feed events in grow-finish
pigs: causes and consequences. Nebraska Swine Reports.
Brumm, M., S. L. Colgan, and J. J. Bruns. 2008. Effect of out-of-feed events and diet particle size on pig
performance and welfare. J. Swine Health Prod. 16(2):72-80.
DeDecker, J. M. 2002. Effects of space allowance in a wean-to-finish system and pig removal strategies at
market on the growth performance and variation in performance of pigs. M. S. Thesis University of
Illinois, Urbana.
DeDecker, J. M. 2006. Management factors affecting the growth of pigs and the impact of pig removal
strategies at market on growth performance and production efficiencies. PhD Diss. Univ. of Illinois,
Urbana.
DeDecker, J. M., M. Ellis, B. F. Wolter, B. P. Corrigan, S. E. Curtis, E. N. Parr, and D. M. Webel. 2005. Effects
of proportion of pigs removed from a group and subsequent floor space on growth performance of
finishing pigs. J. Anim. Sci. 83:449-454.
Dritz, S. S., K. Q. Owen, J. L. Nelssen, R. D. Goodband, and M. D. Tokach. 1996. Influence of weaning age
and nursery diet complexity on growth performance and carcass characteristics and composition of
high-health status pigs from weaning to 109 kilograms. J. Anim Sci. 74:2975-2984.
Dwyer, C. M., N. C. Stickland, and J. M. Fletcher. 1994. The influence of maternal nutrition on muscle fiber
number development in the porcine fetus and on subsequent postnatal growth. J. Anim. Sci. 72:911-917.
http://jas.fass.org/cgi/ijlink?linkType=ABST&journalCode=animalsci&resid=72/4/911.
Dziuk, P. J. 1968. Effect of number of embryos and uterine space on embryo survival in the pig. J. Anim.
Sci. 27:673-676.
Foxcroft, G. R., W. T. Dixon, S. Novak, C. T. Putman, S. C. Town, and M. D. A. Vinsky. 2006. The biological
basis for prenatal programming of postnatal performance in pigs. J. Anim. Sci. 84(E. Suppl.):E105-E112.
Gil, P. and A. Knowles. 2003. An industry guide to the production of heavier pigs. BPEX. Meat Livest.
Comm. 24.
Gondret, F., L. Lefaucheur, H. Juin, I. Louveau, and B. Lebret. 2006. Low birth weight is associated with
enlarged muscle fiber area and impaired meat tenderness of the longissimus muscle in pigs. J. Anim.
Sci. 84:93-103.

Gondret, F., L. Lefaucheur, I. Louveau, B. Lebret, X. Pichodo, and Y. Le Cozler. 2005. Influence of piglet
birth weight on postnatal growth performance, tissue lipogenic capacity and muscle histological traits
at market weight. Livest. Prod. Sci. 93:137-146.
Gonyou, H. W., M. C. Brumm, E. Bush, J. Deen, S. A. Edwards, R. Fangman, J. J. McGlone, M. Meunier-
Salaun, R. B. Morrison, H. Spoolder, P. L. Sundberg, and A. K. Johnson. 2006. Application of broken-
line analysis to assess floor space requirements of nursery and grower-finisher pigs expressed on an
allometric basis. J. Anim. Sci. 84:229-235.
Goodband, R. D., M. D. Tokach, S. S. Dritz, J. M. DeRouchey, and J. L. Nelssen. 2008. Feeding and feeder
management influences on feed efficiency. Allen D. Leman Swine Conference; Saint Paul, Minnesota;
September 20-23. Pages 20-27 in Kansas State University – University of Minnesota Nutrition Workshop:
Focus on Feeding Efficiency.
Handel, S. E., and N. C. Stickland. 1987. Muscle cellularity and birth weight. Anim. Prod. 44:311-317.
Hyun, Y., M. Ellis, G. Riskowski, and R. W. Johnson. 1998. Growth performance of pigs subjected to multiple
concurrent environmental stressors. J. Anim. Sci. 76:721-727.
Johnson, R. K., M. K. Nielsen, and D. S. Casey. 1999. Responses in ovulation rate, embryonal survival, and
litter traits in swine to 14 generations of selection to increase litter size. J. Anim. Sci. 77:541-557.
Klindt, J. 2003. Influence of litter size and creep feeding on pre-weaning gain and influence of pre-weaning
growth on growth to slaughter in barrows. J. Anim. Sci. 83:2434-2439.
Kornegay, E. T., and D. R. Notter. 1984. Effects of floor space and number of pigs per pen on performance.
Pig News Info. 5:23-33.
Latorre, M. A., R. Lazaro, D. G. Valencia, P. Medel, and G. G. Mateos. 2004. The effects of gender and
slaughter weight in the growth performance, carcass traits, and meat quality characteristics of heavy
pigs. J. Anim. Sci. 82:526-533.
Linneen, S. K., S. S. Dritz, R. D. Goodband, M. D. Tokach, J. M. DeRouchey, and J. M. Nelssen. 2007. Effects
of frequent out-of-feed events on growth performance of nursery and grow-finish pigs. J. Anim. Sci.
85:2043-2047.
Maes D., A. Larriestra, J. Deen, and R. Morrison. 2001. A retrospective study of mortality in grow-finish
pigs in a multi-sire production system. J. Swine Health Prod. 9(6):267-273.
Mahan, D. C., and A. J. Lepine. 1991. Effect of pig weaning weight and associated nursery feeding programs
on subsequent performance to 105 kilograms body weight. J. Anim. Sci. 69:1370-1378.
Main, R. G., S. S. Dritz, M. D. Tokach, R. D. Goodband, and J. L. Nelssen. 2004. Increasing weaning age
improves pig performance in a multisite production system. J. Anim Sci. 82:1499-1507.
Matthews, J. O., L. L. Southern, T. D. Bidner, and A. M. Persica. 2001. Effects of betaine, pen space, and
slaughter handling method on growth performance, carcass traits, and pork quality of finishing barrows.
J. Anim. Sci. 79:967-974.
McConnell, J. C., J. C. Eargle, and R. C. Waldorf. 1987. Effects of weaning weight, co-mingling, group size,
and room temperature on pig performance. J. Anim. Sci. 65:1201-1206.
Mroz, Z., A. Lipiec, C. Wenk, and M. Kronauer. 1987. The effects of birth weight and dietary lysine between
21 and 84 days of age on the subsequent performance of pigs to 225 days of age. Livest. Prod. Sci.
17:351-363.
NCR-89 Committee on Confinement Management of Swine. 1993. Space requirements of barrows and gilts
penned together from 54 to 113 kilograms. J. Anim. Sci. 71:1088-1091.

Nissen, P. M., V. O. Danielson, P. F. Jorgensen, and N. Oksbjerg. 2003. Increased maternal nutrition of sows
has no beneficial effects on muscle fiber number or postnatal growth and has no impact on meat quality
of the offspring. J. Anim. Sci. 81:3018-3027.
Peterson, B. A. 2008. Effects of birth and weaning weight on variation in growth performance parameters
and carcass characteristics and composition of pigs. PhD Diss. Univ. of Illinois, Urbana.
Peterson, B. A., M. Ellis, R. Bowman, O. Mendoza, A. Rojo, and B.F. Wolter. 2008. The effects of feeder space
and feed form on pig growth performance and carcass yield in a commercial wean-to-finish facility. J.
Anim. Sci. 86(E. Suppl.)E39.
Pond, W. G., and J. A. Manner. 1974. Swine production in temperate and tropical environments. W. H.
Freeman and Company, San Francisco, USA.
Powell, S. E., and E. D. Aberle. 1980. Effects of birth weight on growth and carcass composition of swine. J.
Anim. Sci. 50:860-868.
Quiniou, N., J. Dagorn, and D. Gaudre. 2002. Variation in piglets’ birth weight and consequences on
subsequent performance. Livest. Prod. Sci. 78:63-70.
Richmond, R. J., and R. T. Berg. 1971. Tissue development in swine as influenced by live weight, breed, sex
and ration. Can. J. Anim. Sci. 51:31-38.
Shull, C. M. 2010. Effect of floor space in the nursery and grow-finish periods on the growth performance
of pigs. M.S. Thesis. Univ. of Illinois, Urbana.
Skoch, E. R., S. F. Binder, C. W. Deyoe, G. L. Allee, and K. C. Behnke. 1983. Effects of pelleting conditions
on performance of pigs fed a corn-soybean meal diet. J. Anim. Sci. 57:922-928.
Smits, R. J. 2011. Impact of the sow on progeny productivity and herd feed efficiency. Recent Advances in
Anim. Nutrition-Australia 18:61-67.
Smits, R. J., and C. L. Collins. 2009. Progeny reared by their birth dam do not outperform progeny
crossfostered to a similar parity dam. Manipulating Pig Production XII: p. 143.
Stark, C. R., K. C. Behnke, J. D. Hancock, S. L. Traylor, and R. H. Hines. 1994. Effect of diet form and fines
in pelleted diets on growth performance of nursery pigs. J. Anim. Sci. 72(Suppl.1):214 (Abstr.).
Stein, H. H., and G. C. Shurson. 2009. Board-Invited Review: The use and application of distillers dried
grains with solubles in swine diets. J. Anim. Sci. 87:1292-1303.
Ward, T. L., L. L. Southern, and T. D. Bidner. 1997. Interactive effects of dietary chromium tripicolinated
and crude protein level in growing-finishing pigs provided inadequate and adequate pen space. J. Anim.
Sci. 75:1001-1008.
Wattanakul, W., J. A. Rooke, A. H. Stewart, P. R. English, and S. A. Edwards. 2007. Effect of milk deprivation
during the lactation period on performance and digestive enzyme activities of the piglets following
weaning. Animal. 1:381-387.
Weatherup, R. N., V. E. Beattie, B. W. Moss, D. J. Kilpatrick, and N. Walker. 1998. The effect of increasing
slaughter weight on the production performance and meat quality of finishing pigs. Anim. Sci. 67:591-600.
Wolter, B. F., M. Ellis, B. P. Corrigan, and J. M. DeDecker. 2002a. The effect of birth weight and feeding
of supplemental milk replacer to piglets during lactation on preweaning and postweaning growth
performance and carcass characteristics. J. Anim. Sci. 80:301-308.
Wolter, B. F., M. Ellis, B. P. Corrigan, J. M. DeDecker, S. E. Curtis, E. N. Parr, and D. M. Webel. 2003a. Impact
of early postweaning growth rate as affected by diet complexity and space allocation on subsequent
growth performance of pigs in a wean-to-finish production system. J. Anim. Sci. 81:353-359.

Wolter, B. F., M. Ellis, B. P. Corrigan, J. M. DeDecker, S. E. Curtis, E. N. Parr, and D. M. Webel. 2003b. Effect
of restricted postweaning growth resulting from reduced floor and feeder-trough space on pig growth
performance to slaughter weight in a wean-to-finish production system. J. Anim. Sci. 81: 836-842.
Wolter, M. F., M. Ellis, S. E. Curtis, E. N. Parr, and D. M. Webel. 2002b. Effects of feeder-trough space and
variation in body weight within a pen of pigs on performance in a wean-to-finish production system.
J. Anim. Sci. 80:2241-2246.
Wondra, K. J., J. D. Hancock, K. C. Behnke, R. H. Hines, and C. R. Stark. 1995. Effects of particle size and
pelleting on growth performance, nutrient digestibility, and stomach morphology in finishing pigs. J.
Anim. Sci. 73:757-763.

2. Feeding and barn management strategies that maximize feed
efficiency
Department of Animal Sciences and Industry, Kansas State University, 246 Weber Hall,
Manhattan, KS 66506-8028, USA; mtokach@ksu.edu
Abstract
The stockperson is responsible for the daily care and welfare of pigs in the barn. Through their
actions, they also can influence overall feed efficiency. Prior to loading pigs, thoroughly cleaning
and operating the facility in an all-in, all-out manner has shown to improve pig performance
and feed efficiency. Removing feed from previous groups of pigs and repairing feeding and
ventilation equipment influence closeout feed efficiency. When loading the barn, pigs should
not be sorted into narrow weight categories. With weanling pigs, mat feeding can help increase
feed consumption and reduce mortality immediately after weaning; however, prolonged mat
feeding will result in feed wastage. After loading, daily chores that influence overall feed efficiency
include individual pig treatment and timely euthanasia, ensuring water and feed availability,
feeding the appropriate diet, managing the air quality and environmental temperature, properly
adjusting feeders, and handling pigs in a positive manner. Removing a portion of the pigs from
all pens during initial marketing events can result in feed savings while maximizing weight
produced from the facility. Withdrawing feed prior to market also can result in feed savings.
Proper handling during loading and transport to the processor to minimize mortality also can
influence closeout feed efficiency. The stockperson can play a large role in improving overall feed
efficiency by how they manage their day-to-day activities in the barn. This chapter will focus on
these activities and their impact on feed efficiency.
Introduction
As discussed in other chapters, genotype, gender, market weight and dietary factors, such
as energy and amino acid levels, diet form (e.g. pelleting or particle size), and additives (e.g.
ractopamine) are the major drivers of differences in feed efficiency among production systems.
However, many factors within production systems also lead to differences in feed efficiency.
Even after standardizing genetics, facilities, feeders, diets, weights and as many other variables
as possible, feed efficiency and growth rate can remain highly variable within an individual
production system. Once this variation is measured and noted, the question is how to reduce it.
For discussion of the actions that people in the barn can do on a day-to-day basis to influence
feed efficiency, we will separate barn management into four phases of a barn turn that people
in the barn can do: (1) prior to entry; (2) while loading the barn; (3) through daily chores, and
(4) while unloading the barn at marketing time. Each of these four phases of barn management
provides unique opportunities to influence feed efficiency. There are also some activities that are

emphasized by some producers that are not as important as they believe. These action items that
should be moved down the priority list or eliminated will also be discussed.
Prior to entry
The majority of modern swine production systems around the world manage their growing
facilities on an all-in, all-out basis where all of the pigs are removed before another group of
pigs enter the facility. This management style has greatly improved performance by decreasing
horizontal transfer of disease from one group of pigs to the next (Cargill and Banhazi, 1998). All-
in, all-out production also provides a unique opportunity for the barn manager to start anew with
each group. The main areas for focus prior to entry are thoroughly cleaning the barn, maintenance
and repair of equipment, and ensuring feed bins are empty before delivering new feed.
Thoroughly clean the barn
The primary objective of barn cleaning practices is to lower the dose of infectious pathogens
that can be transmitted from one group of pigs to the next. Environmental contamination is an
important contributor to bacterial and viral infections. For example, Davies et al. (1999) found
that 27% (7/26) of samples obtained from a fully slatted finishing floor just prior to placement
of pigs were found to be positive for salmonella. Cargill and Banhazi (1998) found that cleaning
barns between groups of pigs was the most important component of all-in, all-out production.
It led to improved pig performance and a greater reduction in respirable dust particles, viable
bacteria counts, and gram positive bacteria counts than gained by adopting all-in, all-out
production without cleaning barns (Table 1).
It has been well documented that animal performance is increased in ‘clean vs. dirty’ environments
(Renaudeau, 2009). Pigs reared in a dirty environment had a 10% reduction in average daily gain
(ADG) (0.78 vs. 0.87 kg) and 18% reduction in feed intake (1.86 vs. 2.28 kg) as compared with
Table 1. Influence of all-in, all-out management and cleaning between batches of pigs on growth performance
and air quality measurements (adapted from Cargill and Banhazi, 1998).
Item1 All-in, all-out; All-in, all-out; Continuous flow

cleaned not cleaned
Average daily gain (g) 658a 619b 610b

Airborne dust (mg/m3) 1.80 2.31 2.51
Respirable particles (mg/m3) 0.201a 0.265b 0.29b
Viable bacteria (CFU × 103/m3) 132a 177b 201b
Gram positive bacteria (CFU × 103/m3) 82a 109b 122b
1 CFU = colony forming units.
a,b Groups without a common superscript letter differ at P<0.05.

2. Feeding and barn management strategies that maximize feed efficiency
pigs reared in a clean environment. The influence of sanitation on pig performance appears to
impact feed intake and thus growth rate to a greater extent than feed efficiency.
Cleanliness is probably responsible for a large percentage of the growth performance benefits
from all-in, all-out production (Amass et al., 2001). Fortunately, most swine pathogens only
survive for a brief amount of time outside the host in the absence of organic materials or moisture.
Under experimental conditions, up to 99% of the bacteria can be removed by cleaning alone.
Removal of visible organic matter removes 90% of bacteria from the environment. Another 6 to
7% of bacteria are killed by disinfectants with a final 1 to 2% killed by fumigation (Morgan-Jones,
1987). Letting the barn dry between washing and loading with pigs also plays a role in pathogen
load. When we cannot let the facility dry, viruses have the opportunity to survive for extended
periods of time. For example, PRRS can survive in water for up to 11 days; however, when dried
it dies quickly (Pirtle and Beran, 1996).
When washing, one of the frequent errors producers make is not adequately cleaning feeders and
waterers or not removing disinfectant and water from feeders or waterers. Because many feeders
and waterers are not easily removed for cleaning, other methods must be used to remove water
and dry them. Some producers use leaf blowers to remove the water from feeders and waterers
that are not easily moved for cleaning.
Basic hygiene practices to decrease pathogen transmission from group to group include: (1)
Building materials that are easy to clean. Rough surfaces such as concrete are more difficult to
clean than smooth surfaces such as wire and plastic. Smooth nonporous surfaces will provide
easier removal of fecal matter and faster drying. (2) Thorough cleaning and removal of organic
matter such as feces and feed. In general, organisms are protected against disinfection agents by
organic materials such as pus, serum, or feces. (3) Proper use of disinfectants including correct
dilution and application. Diluting the disinfectant below its proper dosage or applying disinfectant
to manure covered floors render it ineffective. (4) Lastly, proper downtime and drying of rooms
is vital to minimizing pathogen load.
A survey of hygiene practices on 129 French farms indicated several practices associated with
decreased residual contamination in nurseries (Madec et al., 1999). The practices included
dampening of the rooms immediately after moving the pigs out of the room. The researchers
hypothesized that dampening prevented dying of the fecal matter and increased the ease
and thoroughness of cleaning. Using a detergent was also recommended and associated with
decreased residual contamination. However, in another study evaluating the impact of using
a detergent, the researchers were unable to detect any impact on residual contamination after
thorough washing (Kihlstrom et al., 2001). This indicates that using a detergent may improve the
ease of cleaning; however, if cleaning procedures are thorough, detergents may not impact the
final amount of residual contamination.
Several other studies indicated that thorough cleaning and removal of organic matter resulted
in less residual contamination (Amass et al., 2001; Kihlstrom et al., 2001). Additionally, greater
distances between the surface of the slurry and the floor were associated with less residual
contamination. The authors attributed this risk factor to splash back and recontamination during

the cleaning process. Finally, factors associated with disinfectant usage such as proper dilution
and application were important. Commonly available disinfectants vary widely in their ability to
neutralize viruses such as porcine circovirus type 2 (PCV2; Figure 1). Royer et al. (2001) evaluated
11 commonly used disinfectants in swine farms and research laboratories. These included several
disinfectant classes (products) tested: ethanol (alcohol), iodine (Weldol), phenol (1-Stroke, Tek-
Trol), quaternary ammonium (Roccal D Plus, Fulsan), oxidizing agent (Clorox, VirkonS), alkali
(NaOH), and chlorhexidine (Nolvasan). The mean titer after disinfection ranged from 5.2 for the
chlorhexidine to 1.6 for the oxidizing agent VirkonS (log 10 scale). This compares to the control
titer without disinfection of 6.0. The log 10 scale indicates that the reduction from 6 to 5 results
in a 90% reduction, from 6 to 4 a 99% reduction, from 6 to 3 a 99.9% reduction and from 6 to 1
a 99.99% reduction of the virus. There are two important points to remember from this study:
1. PCV2 is a small enveloped virus similar to Parvovirus and thus, difficult to neutralize with
disinfectants.
2. This study was conducted under controlled laboratory conditions and designed for maximum
disinfectant activity. Disinfectant activity may be less effective in the field setting.
Although our knowledge on proper cleaning and disinfecting procedures for swine facilities
is not complete, considerable research has improved our knowledge base in the past 10 years.
In addition to the PCV2 disinfectant evaluation, this includes evaluation of farrowing house
cleaning protocols, boot bath cleaning disinfectants and procedures, and methods to rapidly
evaluate surface contamination in swine facilities (Amass et al., 2001; Kelly et al., 2001; Kihlstrom
et al., 2001; Amass, 2004; Martin et al., 2008).
Maintenance and repair of equipment
The best time to conduct maintenance is when the barn is empty. Most good producers have a
mental list of items that must be done between groups, but it is helpful to keep a log of items that
need to be fixed, replaced, or serviced when the barns are empty. These could include items such
as greasing bearings on augers or repairing waterers, gates, feeders, inlets, curtains, or insulation.
Titer after disinfection (log10)
7
6.0
6 5.2
5 4.4 4.3 4.3 4.2 3.9
4 3.6
3.3 3.0
3 2.3
2 1.6
1
0
N trol
n
&R
Et ol
Te ol
Fu l
1- an
ke
Vi H
S
o
ca
ro
sa
on
aO
d
Tr
ro
ls
oc
el
ha
on
lo
va
k-
rk
St
N
W
D
R
C
ol
Figure 1. Reduction in infectivity of PCV2 after a 10 min disinfectant exposure (Royer et al., 2001).

From a feed efficiency perspective, maintenance of feed handling is of utmost importance. This
would include fixing any leaking feed bins, broken feed lines, feeder adjustment rods or other
feeder parts (Figure 2). In order to understand the importance of proper maintenance on feed
equipment, it may be helpful to consider the amount of feed passing through a single feeder or
barn on an annual basis. For example, a 1,200 head barn with 48 pens (28 pigs per pen) will have
24 fenceline feeders. If feed efficiency is 2.8 and the pigs gain 100 kg during the finishing period,
each pig would consume 280 kg of feed. Thus, 15,680 kg of feed would pass through each feeder
and a total of 376,320 kg of feed would be used in the barn during a single turn. If 2.8 groups
of pigs are fed in the barn each year, the quantity of feed passing through a feeder and through
the entire barn increases to 43,904 kg and 1,053,696 kg, respectively. Thus, each feeder handles
almost 44 tonnes of feed and over 1,000 tonnes are used in the barn annually. If diet cost is $280
per metric ton, the value of feed used per feeder and barn would be greater than $12,000 and
$295,000 annually. Repairing feed handling equipment to save a small portion of this expense
pays big dividends.
Other equipment, such as watering, ventilation, heating, and cooling equipment, also should be
checked and repaired as needed. If pigs are above or below their thermoneutral zone because of
equipment malfunctions, feed efficiency and growth rate will be negatively impacted.
Before pigs are loaded into the barn, equipment also should be checked to ensure the settings are
correct for starting pigs. Ventilation controllers, temperature probes, fans, and curtains should be
checked to make sure they are operational and that temperatures and ventilation are set for the
Figure 2. Feed spills.

appropriate number and weight of pigs. Before the pigs arrive, the barn should be pre-warmed
if possible and be in the thermo neutral zone for the weight of the pigs. Feeders and feed drops
should be adjusted as needed to ensure the feed is flowing to the entire length of the feeder. In
some cases, feed drops are placed near the end of the feeder or too low in the feeder and the entire
length of the feeder cannot be used by the pigs (Figure 3). Before pigs are placed, the proper diet
should be in the feeders and all watering devices should be checked to ensure they are operating
correctly and with adequate flow.
Ensure feed bins are empty
One consequence of phase feeding and all-in, all-out production is that feed for late finisher pigs
is often left in the feeder after they are marketed. The lysine requirement for late finishing pigs
is likely 40% lower than the requirement of the new pigs entering the barn. If a diet containing
lysine levels typically fed to late finishing pigs is fed to lighter growing pigs, growth rate and
feed efficiency can be 12 and 20% poorer, respectively, than if the pigs were fed to their lysine
requirement (Friesen et al., 1994). Thus, if low lysine, late finisher feed is left in the bin, growth
rate and feed efficiency would be expected to be impaired at the beginning of the next group.
Cleaning of bins between groups of pigs also provides one of the only opportunities that producers
have to ensure that feed buildup is not occurring on sides of bins. This buildup can be the source
of mycotoxins, but can also lead to bridging, poor feed flow ability, and out of feed events after
the subsequent group of pigs is placed.
Loading the barn
Sorting pigs
A common practice of many producers world-wide is to sort pigs by weight as they enter the
facility to decrease variation within a pen. Although sorting to have uniform weight pigs in the
pen is aesthetically pleasing, it is detrimental to pig growth performance (Gonyou et al. 1986b;
O’Quinn et al., 2001). Unsorted pigs have a wider weight range at placement, but grow faster than
pigs that are sorted to be uniform in weight groups (O’Quinn et al., 2001; Table 2). The reason for
Figure 3. Wrong placed feed drops preventing the use of the entire length of the feeder by the pigs.

this difference in growth performance is thought to be due to the increased aggression in pens
with uniform pig weights as compared to pens with variable weight pigs (Francis et al., 1996).
Using individual pig weights within pens, O’Quinn et al. (2001) found that the medium weight
pigs in the sorted pens grew slower than medium weight pigs in the unsorted pens. Although
weight variation within a pen is less at placement when pigs are sorted into uniform groups,
weight variation at market is similar to unsorted pens that have wide variation at placement
(Tindsley and Lean, 1984; O’Quinn et al., 2001).
There are legitimate reasons for sorting pigs, such as for split sex feeding or to feed the lightest
pigs on one feed line and the heaviest pigs on another feed line. However, unless the pigs are
split for these reasons, pigs should not be sorted when loading the barn. Pigs should be ‘gate
cut’ into pens to allow the normal variation within each pen. As will be discussed later in this
chapter, having the entire variation in pig weights within each pen also plays an important role
in marketing as it allows pigs to be removed from all pens at the initial marketing event.
Stocking
During placement, another common practice in many production systems is to stock pens at a
greater density than normal, in order to leave one or two pens (commonly referred to as ‘pull’
or ‘hospital’ pens) to allow for sick or unthrifty pigs to be sorted at a later time. Although this
practice can help ensure that sick pigs receive proper access to feed and water and protection from
pen mates, it can be troublesome from a performance standpoint if not handled correctly. For
example, if too much space remains open and these ‘pull’ pens are not filled in a timely manner,
pigs in the other pens will be overcrowded and performance will be reduced. Also, some feeders
are designed to be shared between two pens with an open trough between the pens. These feeders
can waste considerable amounts of feed if stocking density is much greater on one side of the
feeder (regular pen) than the other (‘pull pen’).
Table 2. Sorting pigs at placement reduces overall growth rate and final body weight (adapted from O’Quinn
et al., 2001).
Item1 Sorted pens Average2 Unsorted
Heavy pens Medium pens Light pens
ADG (g) 0.94 0.92 0.91 0.92 0.94

ADFI (g) 2.67 2.66 2.73 2.69 2.70
Feed:Gain ratio 2.85 2.93 3.02 2.93 2.88
Final weight (d 91) (kg) 123.4 117.8 113.2 118.1 119.9
1 ADG = average daily gain; ADFI = average daily feed intake.
2 Average of heavy, medium, and light pigs from sorted pens.

Mat feeding
When weaned pigs are loaded into a nursery or wean-to-finish barn, producers often mat feed
to help pigs find feed after weaning and improve the transition to dry feed. When mat feeding,
a small amount of feed (e.g. 500 g/pen) is placed on a board or mat. Feed is usually placed
on the mat two or three times per day. When done correctly, mat feeding can improve growth
performance and feed efficiency by reducing the number of ‘fallout’ pigs that fail to find feed
immediately after weaning (Potter et al., 2010). Because the only benefit of mat feeding is to
help pigs find feed quickly after weaning, there is no benefit to mat feeding if it does not reduce
morbidity or mortality. Providing too much feed or for too long a period can result in poor feed
efficiency (Potter et al., 2010).
Daily chores
Daily care and management of pigs in the barn is the greatest responsibility of a barn manager
and an area where that person can have the greatest impact on feed efficiency. The herdsman has
the responsibility to meet the daily needs of each individual pig every day. The basic needs of the
pig can be simplified to feed, water, environment (air and temperature), and health.
Access to quality feed
In many production systems, the diet fed to pigs is controlled by someone other than the person
in the barn. Feed budgets are often used to provide the appropriate quantity of each diet to pigs
before they are switched to the next diet. Improper budgeting can lower pig performance, impair
feed efficiency, and increase cost. By under-budgeting diets, pigs are fed below their amino acid
requirement which will decrease pig performance (Main et al., 2008). By over-budgeting diets,
pigs are fed above their amino acid requirements which simply increases feed cost and reduces
profitability (Main et al., 2008). Many feed companies have tools available to assist with feed
budgeting. A simple feed budget calculator is available at: www.KSUswine.org.
The first step in providing quality feed is ensuring that the correct diet was ordered and delivered
from the feed mill. Also, the feed then must be made available to pigs in the barn. Feed access can
be interrupted by a variety of manners, such as late delivery, bridging in feed bins, plugged feeders,
or inadequate feeder space. Although the out-of-feed events can reduce growth performance,
there is little evidence that they impact feed efficiency (Brumm et al., 2008).
Design and adjustment of feeders
Feeder design and adjustment can influence feed efficiency. In 1989, Taylor and Curtis compared
11 nursery feeders to determine the degree of feed wastage. They found that feed wastage ranged
from 1.7 to 11%. Relatively few studies comparing feeder designs have been published since
that time. Gonyou and Lou (2000) compared 12 feeders selecting dry and wet/dry feeders that
provided either a single or multiple feeding spaces. They found no difference in growth or feed
efficiency between the single and multiple space feeder design. Pigs fed with wet/dry feeders had
5% greater ADG and average daily feed intake (ADFI) than pigs fed with dry feeders, but there

was no difference in feed efficiency. Gonyou and Lou were testing classes of feeders and did not
have enough replications of any one feeder to directly compare one feeder with another.
In order to understand proper feeder design, it is useful to consider how pigs waste feed while
eating, their eating speed, and their space needs while eating. Gonyou (1999) determined that
most feed wastage occurs during four main pig behaviors: (1) when feed falls from the pigs’
mouth while they are eating, (2) as pigs back away from the feeder with feed in their mouth, (3)
during fights at the feeder, and (4) while stepping in and out of the feeder. Thus, feeder designs
and management strategies that minimize these behaviors will reduce feed wastage. For example,
designs that decrease fighting and increase length of uninterrupted feeding bouts will reduce
feed wastage. Designs that recapture feed that falls from the pigs’ mouth and minimizes stepping
in the feeder also reduce wastage. Full partitions that separate feeding spaces reduce pig-to-pig
interaction during feeding and prolong feeding time (Baxter, 1991). Partitions also encourage
pigs to stand perpendicular to the feeder (Baxter, 1991), which reduces the amount of feed that
drops from the pigs’ mouth to the floor and increases the amount that drops back into the feeder.
With meal (mash) diets, pigs fed with a wet/dry feeder eat faster than those fed with a dry feeder
(Gonyou and Lou, 2000). This difference in eating speed is greatly reduced or eliminated when
diets are pelleted. This is because eating speed is greater for pelleted diets than meal diets when
fed in dry feeders. Thus, more feeder spaces are required for pigs fed meal diets in dry feeders
than when fed pelleted diets or when fed mash diets in a wet/dry feeder.
The speed at which pigs consume feed dictates the maximum number of pigs that can eat from a
single feeding space and, thus, the number of feeding spaces required for a pen of pigs (Gonyou,
1999). Using total eating duration and 80% occupancy rate, Gonyou (1999) estimated 11 to 12
pigs could be fed from each feeding space when fed meal diets from dry feeders. For wet/dry
feeders with meal diets, maximum estimated stocking rate ranged from 10 to 15 pigs per space.
These stocking rates are considerably greater than the traditional recommendations of 4 to 5 pigs
per feeding space, but are supported by data from other studies (Bates et al., 1993; McGlone et
al., 1993; Morrow and Walker, 1994).
The pigs’ spatial requirements dictate the width and depth of the feeding space. Width of the
pigs’ shoulders dictates width of the feeding space. The desired width of the feeding space can be
determined by adding 10% to shoulder width at the maximum pig weight. The shoulder width
can be predicted by the equation (width, cm = 6.1 × BW, kg0.33; Petherick, 1983). Thus, if pigs are
marketed at 110 kg, feeder width should be at least 32 cm. If pigs are marketed at 130 kg, feeder
width should be at least 34 cm. For pigs marketed at 150 kg, the width of the feeder space should
be at least 35 cm. Feeder depth is dictated by the slope of the front of the feeder and size of the
pig’s head at their heaviest weight in the barn. For pigs marketed at greater than 120 kg, the depth
from the front of the feeder lip to where feed exits the feeder should be at least 25 cm to allow
pigs to eat comfortably while perpendicular to the feeder.
Even well designed feeders can waste feed if they are not properly adjusted. Liptrap et al. (1985)
was one of the first to demonstrate that feed wastage was influenced by feeder plate setting. In three

experiments, they found that feed wastage increased by 1.7 to 5.4% with increasing feeder opening.
However, in all three of their experiments, daily gain also increased as feeder opening increased.
In nursery pigs, Smith et al. (2004) compared feeder openings of 9.2 to 31.5 mm for crumbled
diets in a 6-space dry feeder (Table 3). These settings resulted in pan coverage ranging from
6 to 93%. A second criterion was to stock pens with 16, 20, and 24 pigs per feeder. As pan
coverage increased, growth rate and final body weight increased with 11.8 mm opening (12.3%
pan coverage) adequate when pens were stocked with 16 to 20 pigs; however, 17.9 mm opening
(43.7% pan coverage) was required when pens contained 24 pigs. Interestingly, feed efficiency
was not significantly influenced by feeder pan coverage; however, feed efficiency was 2.1%
numerically poorer for the greatest pan coverage (93%) as compared to 12 to 44% pan coverage.
These data demonstrate that the ideal pan coverage for optimal growth performance is greater
than previously thought and is influenced by the number of pigs per feeder space.
Most of the grow-finish research has found similar responses to the nursery experiment with a
greater impact of feeder adjustment on average daily gain than on feed efficiency. In four of the
eight grow-finish experiments (Table 3), growth rate improved as feeder opening and, thus pan
coverage, increased. Feed efficiency was only significantly improved in two of the experiments.
The average numerical improvement in feed efficiency due to decreasing pan coverage was 2.9%
with optimal pan coverage of approximately 40 to 60%.
Increasing pan coverage to increase growth performance appears to be more important with
younger pigs (<70 kg; Myers et al., 2010a; Bergstrom, 2011) than with older pigs. Extra pan
Table 3. Recent experiments determining the influence of feeder adjustment on pig performance.
Reference Stage Feeder Feed pan Significant Improvement in

type coverage (%) response1 Feed:Gain ratio
Min Max
Smith et al., 2004 nursery dry 6 93 ADG, ADFI 2.1%

Duttlinger et al., 2008 finisher dry 26 79 - 3.1%
Duttlinger et al., 2008 finisher dry 24 78 ADG 2.6%
Bergstrom, 2011 Exp. 1 grower dry 9 79 ADFI 2.2%
Bergstrom, 2011 Exp. 1 grower wet/dry 35 65 ADG, ADFI -1.6%
Bergstrom, 2011 Exp. 2 finisher dry 25 83 - 0.0%
Bergstrom, 2011 Exp. 2 finisher wet/dry 53 82 ADG, ADFI 5.7%
Bergstrom, 2011 Exp. 3 finisher wet/dry 63 83 ADG, ADFI 0.4%
Myers et al., 2010a finisher dry 28 75 ADFI, F:G 3.9%
Myers et al., 2010b finisher dry 43 87 ADFI, F:G 4.9%
Average improvement in finisher Feed:Gain ratio by decreasing pan coverage: 2.9%
1 ADG = average daily gain; ADFI = average daily feed intake, F:G = Feed:Gain ratio.

coverage appears to have little impact on feed efficiency with younger, lighter pigs. Conversely,
excess feeder pan coverage appears to have the greatest negative effect on feed efficiency with
older, heavier pigs (Myers et al., 2010a). The difference in response at different body weights may
be due to the differences in eating speed. Younger pigs eat more slowly and require more total
eating time than older pigs (Figure 4; Hyun et al., 1997). The difference in eating speed means
that younger pigs require more feeding time when stocked at the same density on the same feeder
than older pigs. Thus, if the feeder is adjusted too tightly, they may not have enough time to work
the feeder and thus cannot consume enough feed to maximize growth performance. Conversely,
older pigs that eat faster have more time to work the feeder and consume their daily requirement.
As reviewed by Brumm and Gonyou (2001), feed access appears to alter feeding speed with
ranges of 22 g/min for a restrictive feeder (Walker, 1991) to 37 g/min for pigs eating from a more
spacious feeder (Gonyou et al., 1992).
Water availability
Water is often a forgotten or under-appreciated nutrient. Access to water appears to be the most
critical factor with water flow rate and quality less important for growth performance and feed
efficiency. Pigs are adept at adapting their drinking time based on flow rate to achieve their
needed water intake. Numerous recommendations exist for minimum water flow rates; however,
Brumm and Mayrose (1991) demonstrated that, even in summer heat without sprinkling for
cooling, flow rate could be as low as 250 ml/min for finishing pigs with 22 pigs per nipple without
altering growth performance. The low flow rates caused more social disruptions indicating that
pig behavior was influenced, but pigs apparently adjusted their drinking time to maintain intake.
From a practical perspective, flow rate should be 0.5 to 1 liter/min for 10 to 20 kg pigs and 1 to 2
liters per minute for finishing pigs (Defra, 2003).
For drinker numbers, a general recommendation is that one nipple drinker should be provided
for every 15 pigs and one bowl drinker should be provided for every 30 pigs (Defra, 2003).
110
Eating duration per day (min)
100
90
80
70
60
50
25 35 45 55 65 75 85
Body weight (kg)
Figure 4. Eating duration (minutes per pig per day) as influenced by body weight (adapted from Hyun et al., 1997).

Providing multiple drinkers in a pen reduces the possibility of limiting water intake when a
drinker becomes plugged or somehow unusable.
Most recent research on water and water devices has centered on the impact on water waste
and manure volume. In a series of experiments as reviewed by Brumm and Gonyou (2001),
researchers demonstrated that providing water via wet/dry feeders reduced wastage compared
with nipple drinkers; swinging drinkers reduced water use compared with nipple drinkers and
bowl drinkers decreased water use compared with swinging drinkers. Although not directly
compared in the same experiment, bowl drinkers and wet/dry feeders appear to have similar
water use. They both also had the lowest water disappearance of the delivery devices tested.
No differences in performance were found in most experiments except in one trial where pigs
provided water through a bowl drinker had a slight improvement (2.49 vs. 2.55, P<0.01) in feed
efficiency compared with pigs provided water from a swinging drinker. Patience et al. (2004) also
demonstrated that nursery feed efficiency tended to be improved (1.49 vs. 1.54, P<0.10) with a
bowl drinker compared with a nipple drinker. The reasons that feed efficiency would be improved
with a bowl or cup drinker are not known.
Providing a clean water source is important as it can be a source of bacteria and viruses (Thacker,
2001). Standards for other water quality criteria have been set for total dissolved solids, pH,
hardness, sulfates, nitrates, and other contaminants (Thacker, 2001). In reality, numerous studies
have demonstrated that pigs adapt to even high levels of most of these other water measures with
little impact on performance. However, many swine producers are often concerned with loose
stool and diarrhea with poor quality water when pig performance is not affected (Thacker, 2001).
Manage the environment
The person in the barn has the greatest responsibility for daily management of air quality and
environmental temperature. The impact of environmental temperature on feed intake and growth is
well documented and covered in detail by Renaudeau et al. (2012; Chapter 9 in this book). In brief,
as ambient temperature drops below the lower critical temperature, feed intake increases. Because
of the increase in thermoregulatory requirement, feed efficiency becomes poorer. Nienaber et al.
(1990) demonstrated that feed efficiency was 14% and 35% poorer for pigs housed at 4 or 11 °C,
below their lower critical temperature, respectively, than pigs housed within their thermoneutral
zone (2.75 vs. 3.14 vs. 3.72 F:G). Clearly, housing pigs below their lower critical temperature greatly
increases feed intake and maintenance requirements which results in poorer feed efficiency.
Heat stress also can have a negative impact on feed efficiency due to the increased energy
requirement for thermoregulation. In finishing pigs, growth rate is often affected to a greater extent
due to hot temperatures than cold due to the negative impact on feed intake. In a meta-analysis,
Renaudeau et al. (2011) mathematically described the relationship between temperature and
pig performance at different body weights. From their data, the maximum ambient temperature
before daily gain and feed efficiency are influenced can be calculated. These data indicate that
daily gain is negatively affected before feed efficiency as temperature increases (Table 4). Their
review indicated that the effect of temperature on daily gain is greater in today’s pigs than those
in the past. This indicates that genetic changes for increased lean growth may have led to pigs

Table 4. Maximum ambient temperature (°C) before daily gain or feed efficiency are impacted due to high
temperatures.
Body weight (kg) Daily gain Feed efficiency
10 29.3 35.6
15 28.2 34.1
20 27.4 33.0
25 26.7 32.1
30 26.2 31.4
35 25.7 30.8
40 25.3 30.2
45 24.9 29.8
50 24.6 29.4
55 24.3 29.0
60 24.1 28.6
65 23.8 28.3
70 23.6 28.0
75 23.4 27.7
80 23.2 27.5
85 23.0 27.2
90 22.9 27.0
with reduced capacity to cope with heat stress. The daily reduction in gain for each degree Celsius
increase in temperature from 20 to 30 °C was 12, 18, and 25 g/d for publications before 1990, 1990
to 1999, and 2000 to 2009, respectively for a 50 kg pig (Renaudeau et al., 2011).
In modern swine facilities, high ambient temperatures are often a greater risk than low
temperatures. Maintenance of fans, sprinklers, cool cells, and any other heat dissipation equipment
is essential to minimize the effect of high ambient temperature on pig performance. Sprinklers
must be checked, flushed, and if needed, replaced to ensure they are operational before summer
temperature increases. Stir fans, maximum ventilation fans, and fan controllers also should be
checked. It may be helpful to run all equipment simultaneously in all rooms of a facility to ensure
that motors and the electrical system meet needs before high temperatures test their capabilities.
Besides temperature, there is increasing evidence that air quality affects pig performance. Data
from Murphy et al. (2000) illustrates a negative relationship between airborne bacteria numbers
and growth rate. This negative relationship was found in three separate analyses with different
phases of production in the farm to farm comparisons; however, viable airborne bacteria
concentrations were also correlated with stocking density differences across the farms making it
difficult to totally separate the impact of stocking density and air quality. The evidence suggests
that methods to lower airborne bacteria may improve pig performance. Recent commercial
studies with electronic space charge systems (electrostatic particulate ionization) indicate that

reducing airborne particulate matter lowers hydrogen sulfide and ammonia concentrations and
increases pig performance. Several research studies in commercial poultry facilities and with
experimental pathogen loads have validated the ability of these electrostatic space charge systems
to lower dust, ammonia, and airborne pathogens (Ritz et al., 2006). However, more studies under
controlled environments are needed to verify these results in pigs.
Individual pig treatment and timely euthanasia
Another key job of the stockperson is to check each pig each day for signs of disease or discomfort
and to ensure that they can easily access feed, water, and fresh air. Pigs showing signs of disease
should be promptly treated as per veterinarian-client-patient relationship. Any pig that shows
no sign of improvement or has no prospect for improvement after two days of intensive care
should be humanely euthanized, unless there are special circumstances (National Pork Board,
2011). Any severely injured or non-ambulatory pigs with the inability to recover should be
euthanized immediately. These guidelines are for the welfare of the individual pig, but also can
serve to improve the welfare and limit mortality in other pigs by reducing the spread of infection.
Reducing mortality through early detection and prompt intervention will improve feed efficiency
of the overall group. Although the impact varies with timing of mortality in the finishing stage,
each 1% increase in mortality worsens closeout feed efficiency by approximately 0.5%.
Pig handling
In addition to daily observation, the person in the barn also can dictate the level of fear that pigs
have to human interaction. Several experiments have demonstrated that pigs that are fearful of
humans due to repeated negative handling have poorer performance. In a four week study, Boyce
et al. (2001) demonstrated that pigs exposed to negative handling (shocking the animal with a
battery-operated electric prod if the pig approached the person) for 10 s per day had poorer feed
efficiency (3.19 vs. 2.75) than pigs that received positive treatment (stroking the animal) with pigs
that experienced no handling intermediate (2.89). The response was mainly due to a numerical
reduction in daily gain for the negatively handled pigs.
Earlier research by Gonyou et al. (1986a) also demonstrated that similar minimal (no handling)
or positive handling (kneeling and stroking the pig when it approached) had improved ADG
and feed efficiency compared with pigs that received negative (approaching and touching pig on
forehead with gloved hand) or adverse (shocking pigs with electric prod) handling. The negative
effects were most pronounced during the first 3 weeks of the experiment with no difference
between treatments during the last 4 weeks of their 10 week experiment. Hemsworth et al. (1987)
found that alternating between pleasant and unpleasant handling resulted in similar performance
to pigs that received continual unpleasant handling. Black et al. (2001) concluded that minimal
human exposure appears to be the best treatment for maximizing growing pig performance
under commercial conditions. Certainly, there is ample evidence that negative handling must be
avoided, not only from a pig performance perspective, but an animal welfare perspective as well.

Unloading the barn
A barn manager can do an excellent job of managing pigs throughout the growing period, but
their job is not completed until the pigs are marketed and the facility is cleaned and prepared
for another group of pigs. The marketing strategy, length of feed withdrawal before market, and
handling of pigs while loading and during transport also affects closeout feed efficiency.
Marketing pigs from all pens
Pig weight is the major factor influencing market price and gross revenues in a production system.
Because individual pig growth rate varies, the range in weight from the lightest to heaviest pig
within a barn can be over 50 kg at market. Therefore, pigs are often marketed over time. The
heaviest pigs in the barn are sold first, allowing more space and time for lighter pigs to achieve
the desired market weight. Removing the heaviest pigs from a pen increases the growth rate of
remaining pigs in the pen (Woodworth et al., 2000) with part of the response due to increased
floor and feeder space (DeDecker et al., 2005). The remaining portion of the response may be due
to changing social dynamics when dominant pigs are removed from the pen.
After the first marketing event, improvement in the growth rate of pigs remaining in the pen will
result in more total weight sold from the barn. This compensatory growth also results in feed
savings as feed efficiency of pigs remaining in the barn is improved.
The magnitude of growth response to removing pigs from the barn appears to vary due to days
that pigs remain in the barn and stocking density. DeDecker et al. (2005) removed either 25% (16
of 52 pigs) or 50% (26 of 52 pigs) of the pigs in a pen 19 d before market and found that growth
rate and feed efficiency improved for the remaining pigs in the pen (Table 5). In their experiment,
removing 25% of the pigs from the pen improved growth rate of remaining pigs to such an extent
that total weight marketed per pen was greater than for pens without any pigs marketed on day
19. The improvement in feed efficiency led to a savings of 7.7 kg of feed per pig marketed when
25% of the pigs were removed 19 days before market. Removing 50% of the pigs led to further
numerical improvements in performance of the remaining pigs and resulted in a savings of 23.9
kg of feed per pig; however, their compensatory growth was not enough to maintain total weight
sold per pen.
Jacela et al. (2009) conducted two experiments to determine if removing a lower percentage of
pigs than DeDecker et al. (2005) would elicit a similar benefit in subsequent performance. They
observed that removing pigs 2 or 4 pigs from a pen of 25 pigs 15 days before marketing improved
growth and feed efficiency resulting in overall feed savings without reducing total weight sold. In
the second experiment, Jacela et al. (2009) tested several marketing strategies (Table 6). Starting
with 25 pigs per pen, they marketed 2 pigs per pen 20 days before final marketing with an
additional 2, 4, or 6 pig per pen (8, 16, or 24%) marketed 10 days later. All remaining pigs were
marketed on d 20. Like DeDecker et al. (2005), Jacela et al. (2009) found that total weight sold
per pen can be maintained and feed savings per pig increased when a portion of the pigs per pen
are removed before the final market date.

Table 5. Influence of pen unloading strategy on pig performance and feed savings (adapted from DeDecker
et al., 2005).
Initial pigs per pen 52 52 52

Pigs removed 0 13 (25%) 26 (50%)
Space per pig after removal (m2) 7.0 9.4 14.0
Initial pig weight (kg) 113.0 113.7 113.3
Initial weight of remaining pigs (kg) 113.0 110.5 105.8
Final weight of remaining pigs (kg) 126.0 126.5 122.2
Initial weight removed (kg) 0 1,623.1 3,176.9
Removed average weigth per pig (kg) 0 124.9 122.2
Marginal days on feed 19 19 19
Marginal average daily gain (g) 659 829 834
Marginal Feed:Gain ratio 4.24 3.76 3.63
Weight marketed per pen (kg) 6,536.2 6,556.4 6,334.6
Weight marketed per pig placed (kg) 125.7 126.1 122.2
Feed savings per pig placed
kg per pen - 401.1 1,244.5
kg per pig - 7.7 23.9
Table 6. Influence of pen unloading strategy on pig performance and feed savings during the last 20 days
prior to market (adapted from Jacela et al., 2009).
Pigs/pen 25 25 25 25 25
Removed on d 0 0 2 2 2 2
Removed on d 10 0 0 2 4 6
Space/pig (m2) 0.67 0.72 0.80 0.88 0.98
Total pen gain (kg) 458.1 465.8 459.0 445.9 442.3
Total pen feed (kg) 1,320.0 1,261.0 1,250.6 1,226.1 1,168.0
Total wt marketed (kg) 3,127.6 3,139.3 3,123.5 3,123.9 3,122.6
Marginal average daily gain (kg) 0.92 1.01 1.04 1.06 1.11
Marginal average daily feed intake (kg) 2.64 2.74 2.84 2.92 2.92
Marginal Feed:Gain ratio 2.88 2.71 2.73 2.75 2.64
Feed savings per pig placed
kg/pen - 59.0 69.4 93.9 152.0
kg/pig - 2.4 2.8 3.8 6.1

To capture the growth and feed efficiency benefit for the greatest majority of pigs in the barn, a
portion of the pigs should be removed from every pen at the initial marketing event. If certain
pens have several pigs removed and other pens have none or very few pigs removed, the benefit
will be limited to those pens where pigs were removed. Although we are focusing on the feed
savings in our feed efficiency discussion, the value gained by increasing the weight of remaining
pigs in the pen should not be overlooked. This weight gain is highly valuable for any producer
marketing to a processor with discounts for lightweight pigs.
Feed withdrawal prior to market
Withdrawing feed from pigs before market will decrease weight of intestinal contents resulting in
decreased risk of accidental laceration of the gastrointestinal tract during evisceration (Kephart
and Mills, 2005; Frobose et al., 2011). It can also significantly increase feed savings and carcass
yield. However, care must be taken to not extend the feed withdrawal period too long or carcass
weight will also be reduced.
In two experiments, Kephart and Mills (2005) tested feed withdrawal periods of 6 or 24 h (Exp.
1) or 6, 16, and 24 h before slaughter (Exp. 2). Feed withdrawal decreased feed intake in both
experiments and resulted in an approximate feed savings of 2 kg/pig for 24 h withdrawal. To
the contrary, they also found a 1 kg reduction in carcass weight with 24 h feed withdrawal in
both experiments. The negative impact on carcass weight was not observed with 16 h of feed
withdrawal, but feed savings were also not as great.
In two similar experiments, Frobose et al. (2011) tested withdrawal times of 7, 24, 36, and 48 h
(Exp. 1) or 7, 12, 24, and 36 h (Exp. 2) before slaughter. Similar to the results of Kephart and Mills
(2005), feed withdrawal led to important feed savings (Table 7). In contrast to the earlier trials,
carcass weight was not reduced until feed withdrawal was greater than 24 h. From these data, it
appears that feed withdrawal of 12 to 16 h can be safely used without reducing carcass weight.
Extending the feed withdrawal past 16 h has the potential for greater feed savings, but may result
in reduced carcass weight. Kephart and Mills (2005) also suggested that the negative effect of
longer feed withdrawal times may be more pronounced in hot weather than in cold weather.
Withdrawing feed before market is easily implemented when selling the last loads of pigs in the
barn; however, it is difficult to implement when other pigs remain in the barn. Special facility
designs, such as sort barns, allow feed withdrawal to be practiced on all pigs marketed from the
barn. Without these designs, pigs remaining in the barn are subjected to repeated out of feed
events, which can lower weight gain of pigs remaining in the barn.
Handling pigs while loading
Mortalities that occur during the marketing process or during transport to the processor are
more expensive to the producer than during any other stage of production. This is because 100%
of production costs, including all feed, have been used with no value returned to the production
system. In a review of 23 trials conducted in the USA, Ritter et al. (2009) found that losses at the
processing plant were approximately 0.69% with 0.25% being from dead pigs and 0.44% from

Table 7. Effect of feed withdrawal on finishing pig carcass weight and feed savings (adapted from Frobose
et al., 2011).
Feed withdrawal in hours
7 24 36 48
Experiment 1
Yield (%) 74.4 76.1 76.3 76.4
Carcass weight (kg) 95.8 95.5 93.8 93.1
Feed savings (kg) - 2.5 4.3 5.0
Experiment 2
Yield (%) 75.3 75.5 76.1 77.0
Carcass weight (kg) 91.6 92.9 92.4 91.1
Feed savings (kg) - 0.4 1.7 2.9
non-ambulatory pigs. The summary indicated an additional 0.11% of pigs were non-ambulatory
losses at the farm. These losses were estimated to cost the USA pork industry over $46 million.
Although some transportation loss is unavoidable, proper handling and care during loading
and movement will reduce such losses. Transport losses can be reduced by proper loading ramp
and alley floor design, quiet handling with minimal or no use of electric prods, proper stocking
density during transport, avoiding hauling during extreme weather conditions, and controlling
ambient temperature in the transportation vehicle (Grandin, 2010). Moving market weight pigs
in small groups (4 vs. 8 pigs) during loading reduced loading time, physical signs of stress during
loading and unloading, and transport losses at the packing plant (Berry et al., 2009). Procedures
outlined in the National Pork Board’s Transportation Quality Assurance Handbook (2008)
should be followed when loading and transporting market pigs to minimize losses during the
marketing process.
Conclusions
In conclusion, the stockperson is responsible for the daily care and welfare of pigs in the barn
and can clearly influence feed efficiency for a group of pigs. From before loading the barn to
implementing the final marketing program, stock people play a key role in determining success
or failure of a group of pigs’ performance. Key periods when their actions have the greatest
impact on overall feed efficiency include: (1) daily care and treatment of pigs to minimize
mortality; (2) proper maintenance and upkeep of feeding and ventilation equipment; (3)
managing the environment temperature and air quality in the barn; (4) adjustment of feeders;
and (5) implementation of the appropriate marketing strategy. Although other action items can
influence growth and feed intake, the stockperson in the barn can have their greatest impact on
feed efficiency by focusing on these areas.

References
Amass, S. F. 2004. Diagnosing disinfectant efficacy. J. Swine Health Prod. 12:82-83.

Amass, S. F., D. Ragland, and P. Spicer. 2001. Evaluation of the efficacy of a peroxygen compound, Virkon
S, as a boot bath disinfectant. J. Swine Health Prod. 9:121-123.
Bates, R. O., S. L. Tilton, J. C. Rea, and S. Woods. 1993. Performance of pigs stocked at either 5 or 10 per
feeder space in grow-finish. Univ. of Missouri Swine Day Research Report, Columbia.
Baxter, M. R. 1991. The design of the feeding environment for pigs. Page 150-158 in Manipulating Pig
Production III. E.S. Batterham, ed. Australian Pig Science Association, Albury, New South Wales.
Bergstrom, J. R., 2011. Effects of birth weight, finishing feeder design, and dietary astaxanthin and
ractopamine HCl on the growth, carcass, and pork quality characteristics of pigs; and meta-analyses
to improve the prediction of pork fat quality. Doctoral Thesis. Kansas State University, Manhattan.
Accessed at: http://krex.k-state.edu/dspace/bitstream/2097/9218/3/JonathanBergstrom2011.pdf
accessed on February 27, 2012.
Berry, N. L., M. Ritter, E. Brunton, W. Stremsterfer, B. Hoag, J. Wolfe, N. Fitzgerald, M. Porth, D. Delaney,
and T. Weldon. 2009. Effects of moving market pigs in different group sizes during loading on stress
responses and transport losses at the packing plant. J. Anim. Sci. 87(E-suppl. 3):46(Abstr.).
Black, J. L., L. R. Giles, P. C. Wynn, A. G. Knowles, C. A. Kerr, M. R. Jones, A. D. Strom, N. L. Gallagher,
and G. J. Eamens. 2001. A review – factors limiting the performance of growing pigs in commercial
environments. Pages 9-36 in Manipulating Pig Production VIII. P.D. Cranwell, ed. Australian Pig
Science Association, Werribee, Victoria.
Boyce, J. M., P. H. Hemsworth, J. L. Barnett, E. Jongman, and L. Beveridge. 2001. The effects of handling and
space allowance on performance of growing pigs. Page 48 in Manipulating Pig Production VIII. P.D.
Cranwell, ed. Australian Pig Science Association, Werribee, Victoria.
Brumm M. C., S. L. Colgan, and K. J. Bruns. 2008. Effect of out-of-feed events and diet particle size on pig
performance and welfare. J. Swine Health Prod. 16:72-80.
Brumm, M. C., and H. W. Gonyou. 2001. Effects of facility design on behavior and feed and water intake.
In: Swine Nutrition. A.J. Lewis and L.L. Southern, ed.. CRC Press, Boca Raton, FL.
Brumm, M. C., and V. B. Mayrose. 1991. Nipple drinkers for finishing pigs. Nebraska Swine Report EC91-
219. Univ. of Nebraska Cooperative Extension, Lincoln, p. 41.
Cargill, C., and T. Banhazit. 1998. The importance of cleaning in all-in/all-out management systems. Page
15, volume 3 in Proc. Int. Pig Vet. Soc. Congr., Birmingham.
Davies P., J. Funk, and W. E. M. Morrow. 1999. Fecal shedding of Salmonella by a chort of finishing pigs in
North Carolina. J. Swine Health Prod. 7(5):231-234.
DeDecker, J. M., M. Ellis, B. F. Wolter, B. P. Corrigan, S. E. Curtis, E. N. Parr, and D. M. Webel. 2005. Effects
of proportion of pigs removed from a group and subsequent floor space on growth performance of
finishing pigs. J. Anim. Sci. 83:449-454.
Defra Code of Recommendations for the Welfare of Livestock. 2003. http://www.defra.gov.uk/publications/
files/pb7950-pig-code-030228.pdf accessed on January 4, 2012.
Duttlinger, A. W., S. S. Dritz, M. D. Tokach, J. M. DeRouchey, J. L. Nelssen, and R. D. Goodband. 2008.
Effects of feeder adjustment on growth performance of growing and finishing pigs. Kansas Swine
Industry Day Report of Progress 1001. Kansas State University, Manhattan.
Francis, D. A., G. I. Christinson, and N. F. Cymbaluk. 1996. Uniform or heterogeneous weight groups as
factors in mixing weanling pigs. Can. J. Anim. Sci. 76:171-176.

Friesen, K. G., J. L. Nelssen, R. D. Goodband, M. D. Tokach, J. A. Unruh, D. H. Kropf, and B. J. Kerr. 1994.
Influence of dietary lysine on growth performance, carcass characteristics and carcass composition of
high-lean growth gilts fed from 34 to 72.5 kilograms. J. Anim. Sci. 72:1761-1770.
Frobose, H. L., N. W. Shelton, S. S. Dritz, L. N. Edwards, K. J. Prusa, M. D. Tokach, J. M. DeRouchey, R. D.
Goodband, and J. L. Nelssen. 2011. The effects of feed-withdrawal time on finishing-pig characteristics
in a commercial environment. J. Anim. Sci. 89(E-Suppl. 2):130(Abstr.).
Gonyou, H. 1999. Feeder and pen design to increase efficiency. Adv. Pork Prod. 10:103-113.
Gonyou, H. W., and Z. Lou. 2000. Effect of eating space and availability of water in feeders on productivity
and eating behavior of grower-finisher pigs. J. Anim. Sci. 78:865.
Gonyou, H. W., R. P. Chapple, and G. R. Frank. 1992. Productivity, time budgets and social aspects of eating
in pigs penned in groups of five or individually. Appl. Anim. Behav. Sci. 34:291.
Gonyou, H. W., P. H. Hemsworth, and J. L. Barnett. 1986a. Effects of frequent interactions with humans on
growing pigs. Appl. Anim. Behav. Sci. 16:269-278.
Gonyou, H. W., K. A. Rohde, and A. C. Echeverri. 1986b. Effects of sorting pigs by weight on behavior and
productivity of growing/finishing pigs. J. Anim. Sci. 63:163-164.
Grandin, T. 2010. Welfare of pigs during transport. National Pork Board Factsheet. Accessed at http://www.
extension.org/pages/27248/welfare-of-pigs-during-transport on January 9, 2012.
Hemsworth, P. H., J. L. Barnett, and C. Hansen. 1987. The influence of inconsistent handling by humans on
the behaviour, growth, and corticosteroids of young pigs. Appl. Anim. Behav. Sci. 17:245-252.
Hyun, Y. M., M. Ellis, F. K. McKeith, and E. R. Wilson. 1997. Feed intake pattern of group-housed growing-
finishing pigs monitored using a computerized feed intake recording system. J. Anim. Sci. 75:1443.
Jacela, J. Y., S. S. Dritz, M. D. Tokach, J. M. DeRouchey, R. D. Goodband, and J. L. Nelssen. 2009. Economic
impact of removing pigs before marketing on the remaining pigs’ growth performance. Kansas Swine
Industry Day Report of Progress 1020, pp 262-269. Kansas State University, Manhattan.
Kelly, J. A., S. F. Amass, and D. Ragland. 2001. Analysis of Lightning and BioClean tests for assessment of
sanitation. J. Swine Health Prod. 9:207-224.
Kephart, K. B., and E. W. Mills. 2005. Effect of withholding feed from swine before slaughter on carcass and
visceral weights and meat quality. J. Anim. Sci. 83:715-721.
Kihlstrom S., W. E. M. Morrow, P. Davies, and G. Luginbuhl. 2001. Assessing the progressive decontamination
of farrowing crate floors by measuring the decrease in aerobic bacteria. J. Swine Health Prod. 9:65-69.
Liptrap, D. O., G. L. Cromwell, D. E. Reese, H. J. Monegue, T. S. Stahly, G. R. Parker. 1985. Effect of feeder
adjustment on feed wastage and pig performance. J. Anim. Sci. 61(Suppl. 1):110.
McGlone, J. J., T. Hicks, R. Nicholson, and C. Fumuso. 1993. Feeder space requirement for split sex or mixed
sex pens. Texas Tech Univ. Agriculture Science Report T-5-327.
Madec, F., F. Humbert, G. Salvat, and P. Maris. 1999. Measurement of the residual contamination of post-
weaning facilities for pigs and related risk factors. J. Vet. Med. 46, 37-45.
Main, R. G., S. S. Dritz, M. D. Tokach, R. D. Goodband, J. L. Nelssen, and J. M. DeRouchey. 2008. Effects
of feeding growing pigs less or more than their lysine requirement in early and late finishing on overall
performance. Prof. Anim. Sci. 24:76-87.
Martin, H., M. F. Le Potier, P. Maris. 2008. Virucidal efficacy of nine commercial disinfectants against
porcine circovirus type 2. The Vet. J. 77:388-393.
Morgan-Jones, S. 1987. Practical aspects of disinfection and infection control. Page 144 in Disinfection in
Veterinary and Farm Animal Practice. A. H. Linton ed. Blackwell Scientific Publications, Oxford.
Morrow, A. T. S., N. and Walker. 1994. Effects of number and siting of single-space feeders on performance
and feeding behavior of growing pigs. J. Agric. Sci. 122:465.

Murphy, T., C. Cargill, and J. Carr. 2000. The effects of stocking density on air quality. Proc. 16th Int. Pig Vet.
Soc. September, 2000, p. 326.
Myers, A.J., R.D. Goodband, M.D. Tokach, S.S. Dritz, J.R. Bergstrom, J.M. DeRouchey, and J.L. Nelssen.
2010a. The effects of feeder adjustment on growth performance of finishing pigs. Kansas Swine Industry
Day Report of Progress 1038. Kansas State University, Manhattan.
Myers, A.J., R.D. Goodband, M.D. Tokach, S.S. Dritz, J.R. Bergstrom, J.M. DeRouchey, and J.L. Nelssen.
2010b. The effects of feeder space and adjustment on growth performance of finishing pigs. Kansas
Swine Industry Day Report of Progress 1038. Kansas State University, Manhattan.
National Pork Board PQA Plus Manual. 2011. Version 1.2. Accessed at: http://www.pork.org/filelibrary/
PQAPlus/PQAPlusEdBook.pdf on February 25, 2012.
National Pork Board Transportation Quality Assurance Handbook. 2008. Version 4. Accessed at: http://
www.pork.org/filelibrary/TQA/manual.pdf on January 9, 2012.
Nienaber, J. A., T. P. McDonald, G. L. Hahn, and Y. R. Chen. 1990. Eating dynamics of growing-finishing
swine. Transactions of the Am. Soc. Agric. Eng. 33:2011-2018.
O’Quinn, P. R., S. S. Dritz, R. D. Goodband, M. D. Tokach, J. C. Swanson, J. L. Nelssen, and R. E. Musser.
2001. Sorting growing-finishing pigs by weight fails to improve growth performance or weight variation.
J. Swine Health Prod. 9:11-16.
Patience, J. F., A. D. Beaulieu, and D. A. Gillis. 2004. The impact of ground water high in sulfates on the
growth performance, nutrient utilization, and tissue mineral levels of pigs housed under commercial
conditions. J. Swine Health Prod. 12:228-236.
Petherick, J. C. 1983. A note on allometric relationships in Large White × Landrace pigs. Anim. Prod. 36:497.
Pirtle, E. C, and G. W. Beran. 1996. Stability of PRRS virus in the presence of fomites commonly found on
farms. J. Am. Vet. Med. Assoc. 208:390-392.
Potter, M. L., S. S. Dritz, M. D. Tokach, J. M. DeRouchey, R. D. Goodband, and J. L. Nelssen. 2010. Effects
of mat-feeding duration and different waterer types on nursery pig performance in a wean-to-finish
barn. Kansas Swine Industry Day Report of Progress 1038. Accessed at: http://krex.k-state.edu/dspace/
bitstream/2097/6547/1/Swine10pg62-71.pdf on December 13, 2011.
Renaudeau D. 2009. Effect of housing conditions (clean vs. dirty) on growth performance and feeding
behavior in growing pigs in a tropical climate. Trop. Anim. Health Prod. 41:559-563.
Renaudeau, D., H. Gilbert and J. Noblet, 2012. Effect of climatic environment on feed efficiency in
swine. Pages 183-210 in Feed efficiency in swine. J.F. Patience, ed. Wageningen Academic Publishers,
Wageningen, the Netherlands.
Renaudeau, D., J. L. Gourdine, and N. R. St-Pierre. 2011. A meta-analysis of the effects of high ambient
temperature on growth performance of growing-finishing pigs. J. Anim. Sci. 89:2220-2230.
Ritter, M. J., M. Ellis, N. L. Berry, S. E. Curtis, L. Anil, E. Berg, M. Benjamin, D. Butler, C. Dewey, B.
Driessen, P. DuBois, J. D. Hill, J. N. Marchant-Forde, P. Matzat, J. McGlone,P. Mormede, T. Moyer, K.
Pfalzgraf, J. Salak-Johnson, M. Siemens, J. Sterle, C. Stull, T. Whiting, B. Wolter, S.R. Niekamp, and A.
K. Johnson. 2009. Review: Transport losses in market weight pigs: I. A review of definitions, incidence,
and economic impact. Prof. Anim. Sci. 25:404-414.
Ritz, C. W., B. W. Mitchell, B. D. Fairchild, M. Czarick III, and J. W. Worley. 2006. Improving In-House Air
Quality in Broiler Production Facilities Using an Electrostatic Space Charge System. J. Appl. Poultry
Res. 15:333-340.
Royer, R. L., P. Nawagitgul, P. G. Halbur, and P. S. Paul. 2001. Susceptibility of porcine circovirus type 2 to
commercial and laboratory disinfectants. J. Swine Health Prod. 9:281-284.

Smith, L. F., A. D. Beaulieu,, J. F. Patience, H. W. Gonyou, and R. D. Boyd. 2004. The impact of feeder
adjustment and group size-floor space allowance on the performance of nursery pigs. J. Swine Health
Prod. 12:111-118.
Taylor, I., and S. Curtis. 1989. Nursery feeders: Researchers find feed waste ranges from 2% to 11% in tests.
National Hog Farmer. May 15, 1989.
Thacker, P. A. 2001. Water in Swine Nutrition. in Swine Nutrition. A. J. Lewis and L. L. Southern, eds. CRC
Press, Boca Raton, FL, USA.
Tindsley, W. E. C., and I. J. Lean. 1984. Effect of weight at allocation on production and behavior in fattening
pig. Appl. Anim. Behav. Sci. 12:79-92.
Walker, N. 1991. The effects on performance and behavior of number of growing pigs per mono-place
feeder. Anim. Feed Sci. Technol. 35:3.
Woodworth, J. C., S. S. Dritz,M. D. Tokach, R. D. Goodband, and J. L. Nelssen. 2000. Examination of the
interactive effects of stocking density and marketing strategies in a commercial production environment.
J. Anim. Sci. 78(Suppl. 2):56. (Abstr.).

3. Liquid feeding corn-based diets to growing pigs: practical
considerations and use of co-products
Department of Animal and Poultry Science, University of Guelph, 50 Stone Road East, Guelph,
ON N1G 2W1, Canada; cdelange@uoguelph.ca
Abstract
Liquid feeding has many potential benefits over conventional dry feeding of pigs, such as
improved gut health, use of inexpensive liquid co-products from the food and biofuel industry,
flexibility and ease of feed delivery, and manipulation of feeding value of ingredients with
enzymes and microbial inoculants. These benefits can result in improved growth performance
and feed efficiency, reduce the reliance on feeding antibiotics and improve public views on pork
production and pork products. In the province of Ontario, Canada, about 20% of growing-
finishing pigs are currently raised on liquid feeding systems and experience has been gained
with liquid feeding corn-based diets. Based on growth performance of high health status pigs,
there is limited benefit of liquid feeding corn-based diets to growing-finishing pigs. This is in
contrast to European findings, where swine liquid feeding research is more focused on wheat-
and barley-based diets. Recent research shows that liquid feeding allows for an effective use of
liquid corn distillers solubles and corn steep water. In general, and when used at 15% or less of
feed dry matter content, the use of corn distillers solubles and corn steep water does not result
in major changes in pig growth performance, or carcass and meat quality. The feeding value
of wheat shorts appears improved in liquid fed pigs. There is potential to further enhance the
value of feed ingredients for the pig by steeping with enzymes and controlled fermentation with
microbial inoculants. Uncontrolled (proteolytic) fermentation which contributes to reduced feed
palatability and nutritional value of liquid feeds can be minimized via control of feed pH and
lactic acid content. When using liquid feeding systems, pigs should have access to an additional
source of water. Management of liquid feeding systems requires computer and engineering skills
and attention to detail, especially when using co-products with variable nutrient content.
Introduction
Liquid feeding of swine is an emerging technology in North America, while it is common practice
in other regions of the world, such as Western Europe and Brazil. Liquid feeding systems typically
include a central feed mixing tank and a series of pumps and pipes to deliver liquid feed, with a
dry matter content of approximately 23%, to individual feeding troughs. Feed preparation and
delivery is generally computer controlled.
Liquid feeding has several potential benefits over conventional dry feeding, such as improved gut
health, use of inexpensive co-products from the food and bio-fuel industry, flexibility and ease
of feed delivery, and manipulation of feeding value of ingredients with enzymes and microbial
inoculants (Scholten et al., 1999; Brooks et al., 2001; Van Winsen et al., 2001; Missotten et al.,
2010). These benefits can result in improved growth performance and feed efficiency, reduce the

reliance on feeding antibiotics and improve public views on pork production and pork products
(Jensen and Mikkelsen, 1998; Brooks et al., 2001; Van Winsen et al., 2001; Missotten et al., 2010).
However, these potential benefits have largely been established in European research and in pigs
fed barley- and wheat-based diets, and may not be applicable to corn-based diets.
In the province of Ontario, Canada, about 20% of growing-finishing pigs are currently raised on
liquid feeding systems and practical experience has been gained with liquid feeding corn-based
diets. Key drivers for the implementation of liquid feeding technology in Ontario have been
European heritage of pork producers, establishment of reliable swine liquid feeding equipment
suppliers, and the use of high-moisture corn and inexpensive co-products from the food and
bio-fuel industries (SLFA, 2012). At the University of Guelph (Guelph, ON, Canada) research
is conducted to explore the benefits of liquid feeding corn-based diets and to characterize the
feeding value of locally available co-products.
In this chapter, the value of liquid feeding corn-based diets to growing pigs and the use of both
liquid and high-fiber containing dry co-products is discussed. First a brief overview is provided
of alternative liquid feeding systems and practical aspects of liquid feeding management.
Design of liquid feeding systems
Around the world many different liquid feeding systems are used, ranging from simple and
manual liquid feed mixing and delivery systems to highly mechanized and automated systems
with sensors in feeding troughs for monitoring feed consumption and computer controlled liquid
feed preparation and delivery (e.g. Columbus et al., 2006; Big Dutchman, 2012; Hampshire, 2012;
Weda, 2012). Key aspects of commercial liquid feeding systems are (1) a central feed mixing
tank and residue tanks to separate different batches of feed, (2) tanks for controlled steeping
or fermentation of liquid feed ingredients or mixed liquid feeds, (3) delivery of liquid feed to
feeding troughs and (4) design of feeding troughs. Few controlled studies have been conducted to
evaluate each of these aspects individually and much of the development of liquid feeding systems
is based on observational studies and experiences obtained under commercial conditions.
Most liquid feeding systems are computer controlled and have one central feed mixing tank in
which liquid feed is prepared in batches of about 500 to 3,000 kg. The feed mixing tank is generally
placed on load cells to control dosing of individual feed ingredients into the tank and delivery of
mixed feeds to individual troughs. The use of tanks for the controlled steeping and fermentation
of liquid feed ingredients is not (yet) very common, but may present opportunities to improve
the feeding value of liquid feeds or individual feed ingredients (e.g. Jensen and Mikkelsen, 1998;
Brooks et al., 2001; Canibe et al., 2007; Missotten et al., 2010; Zhu et al., 2011). The latter requires
increased capacity to store liquid feeds or liquid feed ingredients and will be discussed in more
detail later in this chapter.
Many options are available to deliver liquid feed to feeding troughs and may involve the use of
water, rubber plugs or high-pressure air to separate batches of feed and to propel feed through
feed lines. Choices should be made based on the amount of feed that is handled, the number of
different feeds that are made, required control of feed hygiene and cross-contamination between

3. Liquid feeding corn-based diets to growing pigs
different batches of feeds, required accuracy of delivering targeted amounts of feed to individual
troughs, and ease of operation. Obviously, feed hygiene and the accuracy of feed delivery are more
critical for newly-weaned pigs and individually-fed sows than for large groups of growing-finishing
pigs. In many systems for growing-finishing pigs, feed lines are always filled with feed. As a result,
mixing of different feeds occurs when different feeds are delivered to troughs through a common
feed line. Generally this is not a concern, especially if large amounts of feed are handled at regular
intervals. The residual feed in feed lines may be circulated back into the liquid feed mixing tank
and accounted for when dosing feed ingredients for the next batch of feed. In order to reduce
cross-contamination, some systems include feed residue tanks for temporary feed storage, or have
separate – and sometimes parallel – feed lines for the different types of feed, thereby reducing or
eliminating the need to circulate feed back into the mixing tank. In several systems, water is used
to rinse feed lines, and water – rather than feed – is left in the feed lines between feedings. Rubber
plugs may be used to separate batches of feed and rinse water, further reducing the risk of cross-
contamination and improving the accuracy of feed delivery. In a few systems, including the Big
Dutchman HydroAir™ liquid feeding system that is used at the University of Guelph, high-pressure
air is used to move feed to feeding troughs. In the University of Guelph system the feed mixing tank,
with a capacity of only 75 kg, is pressurized as well and a new batch of feed is prepared for each
feed delivery to individual troughs. In this system, water that is use for rinsing is disposed. Because
of these features, small amounts of feed can be delivered accurately and cross-contamination
between different feeds is minimized. It should be noted that it is difficult to completely avoid
cross-contamination between feeds, especially when using a single feed mixing tank. It is therefore
suggested to not deliver feed additives that present a risk of generating residues in pork products
(e.g. feed medication that has a pre-slaughter withdrawal time) through the liquid feeding system.
The two basic designs of feeding troughs are long troughs that allow all pigs in a pen to eat
simultaneously and short ad libitum feeding troughs that are equipped with sensors. Both designs
have advantages and disadvantages and the choice between designs is largely based on personal
preference. Excellent pig performance has been achieved on both trough designs (e.g. MLC,
2005; Zhu et al., 2010). When long troughs are used, pigs are generally fed distinct meals and
feeding space (e.g. the number of pigs per trough and maximum width of pigs at the shoulder)
should be respected. Long troughs allow easy inspection of pigs at the time of feeding, as well
as modest feed intake restriction to optimize feed efficiency and carcass leanness. Even though
feeding levels are controlled in meal-fed pigs, in carefully managed systems high feed intake
levels can be achieved that are similar to those observed in conventional dry feeding systems and
feed wastage is minimized (De Lange et al., 2006; Zhu et al., 2010). When using short ad libitum
feeding troughs, the amount of feeding space can be reduced, providing one feeding space for
every 3 or 4 growing-finishing pigs. In this system, care should be taken that the feed sensors in
troughs are set and function properly, to ensure that liquid feed is always available to the pigs.
In both trough designs, it is critical that feed troughs are empty at least once daily to avoid feed
spoilage and spillage. Combinations of these two basic designs may be considered. For example,
if long troughs do not provide sufficient feeding space for all pigs in a pen, at each feeding time
two batches of feed may be delivered with only a small time interval between batches. This allows
the more dominant pigs to eat at the first feed delivery and the more subordinate pigs to eat at
the second feed delivery. Moreover, feed sensors may be used in long troughs to monitor feed
consumption (Columbus et al., 2006).

Liquid feeding practices
In Ontario, a survey of feeding practices on 25 commercial swine units with liquid feeding systems,
and an additional 10 units with conventional dry feeding systems serving as controls, was completed
in 2004 (Braun and De Lange, 2004). On all farms, corn-based feeds were used. This survey yielded
information about liquid feeding practices, including feed and food safety, accuracy of liquid
delivery, dry matter contents of liquid feeds and water usage, feeding levels and frequencies, feed
processing, and the use of liquid feed ingredients. The survey summary can be accessed on the
website of the Swine Liquid Feeding Association (SLFA, 2012) and includes results of detailed
analyses of nutrients, including heavy metals, mycotoxins, and some microbial characteristics of
liquid feeds and ingredients. The use of feed ingredients is addressed later in this chapter. In this
section, some of the results of the survey are presented and liquid feeding practices are discussed.
In the Ontario survey, no feed or food safety issues were identified that are unique to liquid
feeding. The survey showed that the prevalence of salmonella was lower on farms with liquid
feeding systems than dry feeding systems, which was a confirmation of European findings (Van
der Wolf et al., 2001; Beal et al., 2002).
The survey showed that liquid feed delivery is generally accurate, based on analyzed dry matter
content and nutrient profiles of feed samples taken from the liquid feed mixing tank, the second
feed valve, and the second-last feed valve in the feed line. However, on two of the farms with
older liquid feeding systems, substantial differences were observed in the mineral content of
feed samples taken from the beginning and end of the feed line. Largely because of technological
improvements of liquid feeding equipment, this un-mixing of the feed is less of a concern with
modern systems. Moreover, practical means to reduce feed separation are to use ingredients with
increased viscosity, such as corn distillers soluble or corn steep water, to keep heavy (mineral)
particles in suspension longer, and to control particle size of mineral sources.
The optimum dry matter content of liquid feed varies with the pig’s stage of growth, feed composition,
environmental conditions and water quality. In newly-weaned pigs, physical feed intake capacity
generally limits nutrient intake and the dry matter content of feed should be maximized (Geary et
al., 1996; Russell et al., 1996). When feeding young pigs, the physical pumping capacity of the liquid
feeding system generally determines the upper limit for feed dry matter content; at high liquid feed
dry matter contents, the risk of blocked or burst feed lines increases. In contrast, in finishing pigs
growth performance and feed utilization appears largely independent of feed dry matter content,
when it is varied between 20 and 30% (Gill, 1988; Russell et al., 1996). An important consideration
for growing-finishing pigs is manure volume, which is minimized by maximizing feed dry matter
content. However, additional water may be required at the onset of disease, high environmental
temperatures, failure of the liquid feeding system, or when co-products with extreme mineral levels
are included in the feed. Co-products such as whey, whey permeate, corn distillers solubles and corn
steep water are high in sodium or potassium contents, which should be considered carefully when
formulating diets and can substantially increase the pigs’ requirements for water. According to Gill
(1988), water consumption was increased by 24% and to a water-to-feed dry matter ratio of 3.64
when the potassium level in the feed was raised from 0.83 to 1.71% of feed dry matter by adding

potassium carbonate. Based on these considerations, liquid fed pigs should always be allowed access
to a separate source of good quality water.
For liquid feeding growing-finishing pigs, there appears little benefit of feeding pigs more than
3 meals daily. At the University of Guelph, high levels of feed dry matter intake and growth rates
are achieved when feeding growing-finishing pigs three equal meals at 06:00, 12:00 and 17:00 h
(Zhu et al., 2010). In that study and when feeding dry corn-based feeds, feed intakes were similar
and growth rates of liquid fed pigs were higher than those on conventional ad libitum dry feed
from single space feeders and with 8 pigs per pen (1,114 vs. 1,030 g/d; 33 to 116 kg body weight;
P<0.01). In liquid fed starter pigs between 7 and 25 kg body weight, Woods (2008) achieved
similar levels of feed intake when pigs were fed 6 times daily at 3 h intervals between 06:00 and
21:00 pm when compared with feeding 12 times daily a 2 h intervals.
Liquid feeding research conducted in the UK suggests that the optimum fineness of grinding of
barley- and wheat-based feed may be slightly larger in liquid feeding than in conventional dry
feeding (MLC, 2005). These observations are consistent with observations in growing-finishing
pigs that were liquid fed high-moisture corn-based diets (Zhu et al., 2010). In that study, reducing
the mean particle size in high-moisture corn from 918 to 355 μm did not affect (P>0.10) daily
body weight gain (1,031 vs. 1,027 g/d for course vs. fine between 34 and 116 kg body weight)
or gain:feed (0.463 vs. 0.466 on a dry matter basis), while the incidence of stomach ulcers was
increased in pigs fed the corn with the smaller particle size. High-moisture corn used in that study
was ground through screen sizes with 6/32” vs. 3/32”, and corn samples were dried after grinding
and before measuring particle size.
A key consideration is the cleanliness of liquid feeding systems. Common to all liquid feeding
systems is the development of a bio-film with microbes on the surface of liquid feed tanks, feed
lines and feeding troughs. Therefore, in all liquid feeding systems some fermentation occurs (Brooks
et al., 2001; Canibe and Jensen, 2003; Missotten et al., 2010). Provided that a stable and favorable
microbial population is established, which is dominated by lactic acid producing bacteria, this is
not a concern and frequent cleaning of the liquid feeding system is not recommended (e.g. Brooks
et al., 2001; MLC, 2003). In fact, during fermentation lactic acid and butyric acid can be generated,
which may promote gut health and development, and can benefit pig growth performance (Brooks
et al., 2001; Canibe et al., 2007; Missotten et al., 2010). However, over time the abundance of yeast
or unfavourable bacteria in the bio-film may increase and this can affect the nature of fermentation,
increase the generation of amines and other unfavorable compounds, compromise feed quality and
palatability, and lead to loss of nutrients such as synthetic amino acids (Pedersen et al., 2002; Niven
et al., 2006a; Canibe et al., 2007). Liquid feed quality is likely to be maintained, and loss of synthetic
amino acids minimized, when lactic acid levels in the liquid feed are higher than 75 mM, or when
the pH of liquid feed is maintained below 4.5. It should be noted that feed quality is more likely to
be compromised when liquid feeding newly-weaned pigs, because of feed composition, high room
temperature, relative low volumes of liquid feed that are moved through the system, and inconsistent
feed intakes. The latter may contribute to the often variable growth performance response to (partly)
fermenting feeds, especially in studies involving newly-weaned piglets (e.g. Jensen and Mikkelsen,
1998; Missotten et al., 2010). When feed quality is compromised, the liquid feeding system should
be thoroughly cleaned between batches of pigs, (MLC, 2003), preferably by washing the system with

both acids and bases for controlling yeast and other unfavourable microbes. The use of microbial
inoculants has been suggested to better control the nature of fermentation and, as result, improve
animal performance (Russell et al., 1996; Jensen and Mikkelsen, 1998; Scholten et al., 1999; Brooks
et al., 2001; Scholten, 2001; Canibe et al., 2007). The latter is of special interest when fermenting
liquid feed ingredients to enhance feeding value, which will be discussed later in this chapter.
An important feature of modern computerized liquid feeding system is the control and monitoring
of feed intake and feeding programs. Practices such as split-gender feeding, phase- or blend-
feeding, introduction of new (liquid) feed ingredients, or modest feed intake restriction prior to
slaughter, are easily implemented when using modern liquid feeding systems. Moreover, sudden
changes in feed intake that may indicate onset of disease may be identified quickly. However,
management of liquid feeding systems requires computer and engineering skills, to operate the
system or to fix minor technical problems, and attention to details, to monitor the system and the
quality of incoming ingredients that can be variable in nutrient content.
Effects of liquid feeding corn-based feeds on growth performance and carcass

characteristics
In Western Europe, the beneficial effect of liquid feeding barley- and wheat-based feeds on
growth performance of starting and growing-finishing pigs has been well established (e.g. Jensen
and Mikkelsen, 1998; Brooks et al., 2001; MLC, 2005; Table 1). However, some care should
be taken when extrapolating these results to corn-based feeds. An important aspect of liquid
feeding is soaking of ingredients, which is likely to be more beneficial when using ingredients that
contain higher levels of fiber and endogenous phytase activity, such as barley and wheat. When
compared to ground wheat and barley, ground corn has rather unique physical and nutritional
Table 1. Pig performance and carcass characteristics of pigs that were fed a barley- and wheat-based diets
in either liquid or dry form.1
Feeding system SE Dif P-value
Liquid Dry
Initial body weight (kg) 34.14 34.23 1.29 ns

Final body weight (kg) 103.00 102.90 0.753 ns
Gain (g/d) 796 754 9.6 <0.001
Feed intake (87% DM) (kg/d)2 2.22 2.39 0.021 <0.001
Feed:Gain 2.27 2.53 0.027 <0.001
Carcass dressing (%) 73.96 74.62 0.207 <0.01
Backfat (mm) 11.45 11.39 0.304 ns
1 MLC (2005); Stotfold research unit.
2 DM = dry matter.

characteristics, including high bulk density, rapid settling when suspended in water, low water-
binding capacity, high starch and low fiber, and no endogenous phytase activity.
In initial studies at the University of Guelph, there was no performance benefit of liquid feeding
dry or high-moisture corn-based diets when compared to conventional ad libitum feeding of
pelleted diets with similar ingredient compositions (Table 2). These data suggest that liquid
feeding dry ground corn is equivalent to feeding pelleted corn-based feeds. However, actual
liquid feeding practices – such as soaking time, feeding management and feed wastage – should
be considered when comparing responses to liquid feeding observed in different studies and
research locations. For example, in the MLC (2005) study barley and wheat were allowed to
steep for several hours prior to feeding. In contrast, in the University of Guelph study, liquid feed
with dry corn was dispensed to the troughs within minutes after feed preparation, while high-
moisture corn was stored mixed with water in a 2 to 1 ratio for 1 to 7 days prior to feeding. Liquid
feeding steeped high-moisture corn only yielded a numerical, but not significant, improvement
in feed efficiency of 5% when compared to liquid feeding dried corn. In a subsequent study at the
University of Guelph, the interactive effect of fineness of grinding dried corn and feeding method
was evaluated. In this study, corn used for pigs on both conventional dry feeding and liquid
feeding was not passed through a pellet mill, and a positive effect of liquid feeding on growth rate
was observed (Table 3). The improvement in growth rate was attributed to a numerical increase in
feed intake and a significant improvement in feed efficiency. It should be noted that the observed
improvement in feed efficiency of 4% was not supported by an improvement in fecal organic
matter digestibility and is, therefore, likely the result of reduced feed wastage. The difference in
response to liquid feeding corn-based diets between the initial and subsequent studies may in
part be attributed to increased experience with liquid feeding practices. Experience with liquid
feeding is needed to maximize feed intake and minimize feed wastage, by adjusting feed intake
levels for individual pigs over time. Unfortunately, in the latter study, no performance response
to fineness of grinding corn (mean particle size 652 vs. 525 μm) was observed, and the study
lacked sensitivity to adequately assess the interactive effect of feeding method and fineness of
grinding of corn.
When liquid feeding corn-based diets, only minor changes in carcass and meat quality
characteristics were observed (Table 2). It should be noted that carcass dressing percentage appears
lower in liquid than in dry feeding (Table 1 and 2). Most of the meat quality measurements that
were taken (objective Minolta colour evaluation of loin and ham, subjective structure, firmness
and wetness scores) were not influenced by dietary treatment. However, 24 h pH values for loin
samples was highest for pigs fed the dry corn via the liquid feeding system, but this was not
related to treatment effects on loin drip loss. Subjective marbling scores and color scores were
highest for pigs that were liquid fed high-moisture corn. In this experiment, no concerns were
observed with gut health for any of the treatments. Based on behavior observations taken during
week 6 of the growing-finishing period, pigs raised on liquid feeding systems spend more time
lying and less time nosing other pigs than pigs on the conventional dry feeding system (De Lange
et al., 2006). The latter is likely related to the feeding scheme, rather than the method of feed
delivery. Pigs on the liquid feeding system showed little activity between feedings.

Table 2. Effect of feeding corn-based diets in different forms on growth performance and aspects of carcass
and meat quality of growing-finishing pigs.1
Feeding system SEM2 P2

(n=8)
Conventional, Liquid, Liquid,
dry pelleted dried corn high-
feed moisture
corn
Growth performance and routine carcass measurements

Initial body weight (kg) 23.5 23.7 23.4 1.34 0.95
Final body weight (kg) 104.7 105.8 104.2 2.19 0.16
Gain (kg/d) 983 1013 1014 12.4 0.19
Dry matter intake (kg/d) 2.27 2.33 2.24 0.06 0.63
Extra water usage (l/pig/d) 4.63b 0.32a 0.21a 0.16 0.00
Feed:Gain (DM basis) 2.31 2.30 2.21 0.04 0.26
Carcass dressing (%) 82.2ab 80.4a 82.5b 0.45 0.03
Probe back fat, mm3 16.7 17.0 17.1 0.84 0.90
Probe loin muscle depth (mm3) 55.6 52.2 54.2 1.06 0.18
Estimated carcass lean yield (%)3 61.2 60.9 61.0 0.44 0.80
Meat quality characteristics (1 pig per pen)
24 h loin pH 5.53a 5.58b 5.55ab 0.01 0.05
Loin drip loss (%) 11.4 10.6 11.6 0.67 0.68
Subjective evaluations
Marbling score 1.51a 1.11a 1.93b 0.10 0.01
NPPC color score 2.78a 2.84ab 2.97b 0.03 0.03
Japanese color score 2.49a 2.83ab 2.89b 0.08 0.01
1 Columbus et al. (2006) and De Lange et al. (2006); 8 pens per treatment with 8 pigs per pen. Liquid feeds were
prepared by mixing a pelleted and crumbled supplement (containing all feed ingredients except corn) with water
and ground dry corn or ground and steeped high-moisture corn; the water to feed dry matter ratio was 2.5 to 1.
Liquid fed pigs were fed equal meals four times daily, at 06:00, 10:00, 14:00 and at 18:00 h; at feeding all pigs were
able to eat simultaneously and trough sensors were used to monitor liquid feed delivery. In the conventional dry
feeding system, pigs were fed ad libitum pelleted feeds from single space feeders. For all pigs additional water was
available from nipple drinkers.
2 SEM represents standard error of treatment means; P is the probability of feeding system effect.
3 Canadian carcass grading system.
a,b Treatment means followed by differing superscripts differ (P<0.05).

Table 3. Effects of conventional dry feeding and liquid feeding of corn-based diets on growth performance
and aspects of carcass quality of growing-finishing pigs1.
Feeding system SEM2 P2

(n=12)
Dry Liquid
Initial body weight (kg) 33.7 32.3 0.281

Final body weight (kg) 115.0 118.4 0.84 0.012
Gain (kg/day) 1.030 1.114 0.02 0.004
Feed intake (kg/day)3 2.72 2.80 0.04 0.069
Feed:Gain3 2.64 2.52 0.05 0.041
Apparent fecal organic matter digestibility (%) 87.0 87.8 0.45 0.243
Carcass weight (kg)4 93.0 93.1 1.86 0.932
Probe back fat (mm)4 18.4 19.0 1.23 0.485
Probe loin muscle depth (mm)4 60.6 60.5 1.87 0.947
Estimated lean yield (%)4 60.8 60.5 0.54 0.463
1 Zhu et al. (2010); 12 pens per treatment with 8 pigs per pen. Liquid feeds were prepared by mixing a pelleted and
crumbled supplement (containing all feed ingredients except corn) with water and ground dry corn; the water to
feed dry matter ratio was 2.5 to 1. Liquid fed pigs were fed equal meals three times daily, at 06:00, 12:00 and at 17:00
h; at feeding all pigs were able to eat simultaneously and trough sensors were used to monitor liquid feed delivery.
In the conventional dry feeding system, pigs were fed ad libitum feed a mix of the pelleted and crumbled supplement
and ground corn from single space feeders. For all pigs additional water was available from nipple drinkers.
2 SEM represents standard error of treatment means; P is the probability of feeding method effect.
3 As fed basis (88% dry matter).
Use of liquid co-products: corn distillers solubles and corn steep water
Corn distillers’ solubles (CDS) and corn steep water (CSW) are liquid co-products from ethanol
production and the wet corn-milling industry, respectively, and are typically dried and included
in dried distillers grains with solubles (DDGS) and corn gluten feed. Drying of these products
may reduce nutrient bio-availability (e.g. Fontaine et al., 2007) and represents an energy cost. The
use these liquid co-products was explored.in research at the University of Guelph.
Fresh CDS was analyzed to contain on average 30% dry matter and, on a dry matter basis, 22%
protein, 19% fat, 8.4% ash, 1.4% phosphorus, 10% starch, and about 6% soluble sugars (Braun and
De Lange, 2004). Initial on-farm experiences with liquid CDS were not positive, largely because
of poor palatability. Inoculation of CDS, which is largely sterile, with a mixture of microbes
was believed to enhance palatability of this product. In a simple and small-scale fermentation
unit, fermentation conditions were established to enhance the nutritional value of CDS (Squire,
2005). Standardizing the initial pH to 6 and inoculation with both Lactobacillus acidophilus and
Bacillus subtilus resulted in the most effective fermentation. This was based on decline of pH,

appearance of beneficial lactic acid and other volatile fatty acids, and on microbial evaluation of
the fermented material. These practices were also effective on a larger scale and when feed was
prepared for a pig performance study in 2,500 liter fermentation tanks (Figure 1).
In a preliminary study where growing pigs were fed diets that contained 0, 7.5, 15.0 and 22.5%
non-fermented CDS, it was established that feed palatability was reduced when the inclusion
level exceeded 15.0% (Squire, 2005). In a full-scale growing pig performance study, feeding
non-fermented CDS resulted in a significant reduction in growth performance, while growth
performance of pigs fed the fermented product was not different from the control (Table 4).
Apparently, fermenting CDS provided more beneficial lactic acid and lactic acid producing bacteria
to the pigs, which enhanced feed and nutrient utilization. The apparent fecal digestibility of energy
and protein was slightly reduced in pigs fed fermented CDS (Table 4), which might be due to the
generation of volatile organic acids that are largely lost during drying of liquid samples for nutrient
analyses. The digestibility of fat was increased in pigs fed CDS containing feed, indicating that fat
in CDS is highly digestible. Only pigs fed the control and the non-fermented CDS were raised to
full slaughter weight. Feeding CDS did not impact routine carcass measurements that determine
carcass value (Table 4). However, among the various measurements that were taken to evaluate
carcass and loin meat quality (subjective color scores: NPPC and Japanese color score, firmness
and wetness score, subjective marbling score, actual fat depths at grading site, loin eye area, Minolta
fat and lean colour, drip loss, 24 h pH in loin muscle), only the pH at 24 h post-mortem differed
between the two treatments (Table 4). The higher pH in the pigs fed fresh CDS coincided with a
trend towards reduced drip loss from loin samples. In this study, we did not observe a positive
effect of feeding fermented feed on gut microflora, in terms of lactobacillus bacteria count in the
gut or balance between lactobacillus bacteria and coliform bacteria. It appears difficult to influence
6 30
5 25
Lactic acid (g/kg)
4 20
3 15
pH
2 10
pH, Ferm
pH, Control 5
1
Lactic acid, Ferm
Lactic acid, Control
0 0
0 1 2 3 9 16
Time (days)
Figure 1. Lactic acid content and acidity (pH) of fermented (Ferm) and non-fermented (Control) corn distillers
solubles used in a pig performance study (Squire, 2005).

Table 4. Growth performance, apparent fecal nutrient digestibility and carcass quality of pigs fed a corn
and soybean meal based control diet, or a corn and soybean meal based diet with either non-fermented or
fermented corn distillers’ solubles at 15% of feed dry matter. All diets were fed in a liquid form1 (Squire et al.,
2004, 2005).
Treatment
Control Corn distillers’ solubles SEM2

(n=6)
Non-fermented Fermented
Growth performance
Initial body weight (kg) 23.5 23.3 23.4 1.6
Final body weight (kg) 50.1a 47.5b 48.6ab 0.6
Average daily gain (g/d) 952a 858b 898ab 22
Average daily feed intake (kg/d)3 1.62a 1.49b 1.61a 0.03
Feed:Gain2 1.70 1.73 1.80 0.03
Apparent fecal nutrient digestibility
Energy (%) 81.6ab 82.5a 79.9b 0.60
Crude protein (%) 72.5a 73.2a 69.3b 0.85
Crude fat (%) 80.9b 85.4a 85.4a 0.82
Carcass and meat quality
Final body weight (kg) 106.5 107.0 0.79
Carcass dressing (%) 82.1 82.6 0.31
Probe back fat, mm4 16.6 17.1 0.39
Probe loin muscle depth (mm)4 54.3 53.7 0.86
Carcass lean yield (%)4 61.2 60.9 0.19
24 h loin pH 5.74b 5.80a 0.01
Loin drip loss (%) 9.63 8.83 0.29
1 6 pens per treatment with 10 pigs per pen.
2 SEM = standard error of the treatment means.
3 Dry matter basis.
a,b Values within rows followed by different superscripts differ (P<0.05).
microflora in the lower gut when feeding fermented feed to growing-finishing pigs with a mature
digestive system. Alternatively, fresh CDS may have some pre-biotic properties that are lost during
fermentation. Both non-fermented and fermented CDS can be used in pig feed, provided that
ingredient costs off-set the changes in feed efficiency and body weight gain.
Untreated CSW is commonly used to feed cattle, but pigs may utilize nutrients in this product
more efficiently. Samples of CSW were analyzed to have a pH of 4.3, contain 45% dry matter and,
on a dry matter basis, 50% crude protein, 2% lysine, 18.0% ash, 5% potassium, 3.3% phosphorus

(of which about 80% is bound in phytate), 1.5% magnesium, 0.5% crude fat and 20% lactic acid
(Niven et al., 2006b). The high lactic acid content may be of benefit for reducing pathogen load
and stimulating gut development in young pigs. In a preliminary finishing pig performance study,
the inclusion of 5% CSW (dry matter basis) in the feed numerically enhanced body weight gain
and Feed:Gain (1.10 vs. 1.17 kg/d, SE 0.03, and 2.46 vs 2.30 kg/kg, SE 0.05), while the inclusion
of 10% CSW in the feed numerically reduced pig performance (Niven et al., 2006b). Studies
were then conducted to explore means to enhance the nutritional value of CSW. In controlled
small-scale in vitro studies CSW could not be fermented, most likely because of the antimicrobial
properties of the high lactic acid content. In an attempt to release phytate-bound phosphorus,
samples of CSW were steeped with a commercially available phytase (Natuphos™; BASF, Florham
Park, NJ, USA) at 4 inclusion levels (125, 250, 500 and 750 FTU per kg CSW dry matter) and
at two temperatures (40 °C and 50 °C) that were similar to temperatures during initial storage
of CSW at the processing plant. The rate of phytate-phosphorus release was increased (P<0.05)
with both temperature and phytase inclusion level. At 50 °C and with 750 FTU per kg dry matter,
appearance of soluble phosphorus was maximized at approximately 25 g phosphorus per kg dry
matter (Niven et al., 2006b).
In a subsequent and larger scale finishing pig study, the impact of feeding phytase treated CSW
at different inclusion levels was then evaluated (Table 5). In this study, carcass value was not
influenced by CSW inclusion level. The results show a linear negative effect of inclusion level of
phytase-treated CSW on growth performance, but the drop in performance was most apparent
Table 5. Pig performance and carcass characteristics of pigs that were fed liquid diets containing varying
inclusion levels of phytase-treated corn steep water (CSW).1
Inclusion level of CSW SEM2 P-values

(n=4)
0% 7.5% 15% 22.5% Linear Quadratic
Initial BW (kg) 69.1 68.8 68.8 69.3 0.26 0.560 0.157

Final BW (kg) 108.3 104.6 107.7 103.1 1.36 0.079 0.754
BW gain (g/d) 1,191a 1,080a 1,063a 899b 29 0.0002 0.450
Feed intake (kg/d)3 2.759a 2.491ab 2.584ab 2.293b 0.075 0.008 0.868
Feed:Gain3 2.33a 2.30a 2.42ab 2.55b 0.05 0.002 0.147
Carcass weight (kg) 86.3 82.7 83.4 80.5 1.59 0.060 0.828
Probe back fat (mm)4 18.1 18.7 18.0 17.1 0.66 0.326 0.294
Probe loin muscle depth (mm)4 58.2 58.9 56.4 58.3 0.99 0.723 0.574
Estimated lean yield (%)4 60.3 60.3 60.5 60.1 0.47 0.897 0.633
1 De Lange et al. (2006); 4 pens per treatment and 8 pigs per pen.
2 SEM = standard error of the treatment means.
3 Feed dry matter basis.
a,b Values followed by different superscripts differ (P<0.05).

when the dietary inclusion level of CSW exceeded 15%. The negative effect on performance at the
highest CSW inclusion levels can in part be attributed to the high levels of minerals, especially
potassium, in the product (Gill, 1988). Cost-effective means to reduce mineral levels or to reduce
its negative effect on pig performance should be explored further. Preliminary studies indicate
that manipulation of the balance among minerals, for example by adding calcium chloride to
the feed, is a means of alleviating some of the negative effects of feeding high potassium levels
(Guimaraes et al., 2010). Provided that ingredient costs off-set the changes in feed efficiency and
body weight gain, moderate levels of CSW can be used effectively in swine feed. At inclusion
levels of approximately 5% CSW may improve pig performance, which may be attributed to the
high lactic acid content in the product.
Use of dry high-fiber co-products: wheat shorts and dried distillers grains
with solubles
As mentioned earlier, simply soaking feed ingredients in water may improve their nutritional
value. Moreover, liquid feeding allows more extensive interactions of exogenous and microbial
inoculants with feed ingredients. It has, for example, been shown that steeping corn with phytase
results in near complete release of phytate-phosphorus (Niven et al., 2006b) and that only small
amounts of added phytase are needed to improve phosphorus utilization in starter pigs fed corn-
based feed (Columbus et al., 2010a,b).
In University of Guelph studies, the use of wheat shorts and DDGS in liquid feeding has been
explored. In a finishing pig performance study, it was shown that liquid feed containing 40%
wheat shorts resulted in improvements in feed efficiency when compared to conventional dry
feeding (Table 6). In this experiment, pigs received a pelleted and crumbled supplement that
was mixed with un-pelleted wheat shorts and water at the time of liquid feed preparation and
delivered immediately to the pigs, whereas pigs on conventional feeding received a blend of
crumbled supplement and un-pelleted wheat shorts that was prepared in the feed mill. Feed
ingredient compositions were identical for the two feeding systems. These data illustrate that
the improvement in nutritional value of wheat shorts-based feed is larger in liquid feeding than
conventional dry feeding of un-pelleted wheat shorts. In this study, liquid feeding reduced
estimated carcass lean yield and probe loin muscle depth, and increased probe back fat thickness
(Table 6). It appeared that the higher growth rate achieved in pigs on liquid feeding increased
body fat deposition, indicating increased availability of energy from liquid feeding a wheat
shorts-based feed, and reduced estimated carcass lean content. Unfortunately, attempts to steep
or ferment wheat shorts in water and to study the response to added enzymes and microbial
inoculants had to be abandoned. Wheat shorts swell quickly when mixed with water, which
caused blockage of feed lines in the liquid feeding system.
In simple in vitro studies, the effects of using varying combinations of fiber degrading enzymes
and microbial inoculants on changes in pH, appearance of sugars, organic acids when steeping
and fermenting DDGS or wheat shorts have been explored, as outlined by Niven et al. (2007).
The combination of fiber degrading enzymes and microbial inoculants was explored to ensure
that some of the simple sugars that may be released during fermentation of feed ingredients
and that cannot be metabolized by pigs, like xylose and arabinose, are converted to organic

Table 6. Effects of conventional dry feeding and liquid feeding of diets containing 40% wheat shorts on
growth performance, and aspects of carcass quality of finishing pigs.1
Feeding system SEM2 (n=12) P2
Dry Liquid
Initial body weight (kg) 72.0 72.1 1.40 0.98

Final body weight (kg) 117.7 120.1 0.47 0.002
Gain (kg/d) 1.04 1.18 0.02 <0.001
Feed intake (kg/d)3 3.44 3.47 0.07 0.45
Feed:Gain3 3.32 2.96 0.04 <0.001
Standardized carcass weight (kg)4 91.5 91.6 1.86 0.23
Probe back fat (mm)4 17.3 18.8 0.35 0.003
Probe loin muscle depth (mm)4 62.0 59.5 0.52 0.001
Estimated lean yield (%)4 61.3 60.5 0.17 0.001
1 University of Guelph; Unpublished; 12 pens per treatment with 8 pigs per pen. Liquid feeds were prepared by
mixing a pelleted and crumbled supplement (containing all feed ingredients except wheat shorts) with water and
un-pelleted wheat shorts; the water to feed dry matter ratio was 2.5 to 1. In the conventional dry feeding system,
pigs were fed ad libitum from single space feeders a blend of crumbled supplement and un-pelleted wheat shorts
that was prepared in the feed mill. For all pigs additional water was available from nipple drinkers.
2 SEM represents standard error of treatment means; P is the probability of feeding method effect.
3 As fed basis (88% dry matter).
acids (De Lange, 2000). In these in vitro studies, positive additive effects of using enzymes and
microbial inoculants on the generation of lactic acid were shown (Figure 2). In a subsequent
and preliminary liquid feeding study, steeping DDGS with enzymes and microbial inoculants
improved feed intake and growth performance of finishing pigs fed feeds containing 30% DDGS
(Zhu et al., 2011; Table 7). In this study, feed efficiency was improved when using enzymes
but was not influenced by the use of microbial inoculants. However, numerically the best feed
efficiency was achieved when using both enzymes and microbial inoculants. These observations
illustrate the potential for improving the nutritional value of high fiber co-products through
liquid feeding and the use of exogenous enzymes and microbial inoculants. However, further
work is required to optimize steeping and fermentation conditions, before this technology can
be applied on commercial farms.
Conclusions and implications
Based on growth performance of high health status pigs, there is only limited benefit of liquid
feeding corn-based feeds to growing-finishing pigs. This is in contrast to European findings,
where swine liquid feeding research is more focused on wheat- and barley-based feed. Liquid
feeding allows for an effective use of co-products, such as corn distillers solubles, corn steep

2.5
SE 0.09
Lactic acid
Acetic acid
2.0
Content (%)
1.5
1.0
0.5
0.0
Control +Enz. +Inoc +Enz. + Inoc.
Figure 2. Lactic and acetic acid content (%) in the supernatant of corn distillers solubles mixed with water
(17% dry matter) in large scale fermentation tanks; the treatments represent: no added enzymes and
inoculants (Control), with added xylanase and glucanase (+Enz.), with added Pediococcus inoculant (+Inoc)
or its combination (+Enz+Inoc.). Effects of adding both enzymes and inoculants were significant (P<0.05)
while no interactive effect was observed (P>0.10; Zhu et al., 2011).
Table 7. Effects of treating DDGS with enzymes and microbial inoculant on growth performance of finishing
pigs that are liquid fed diets containing 30% DDGS.1
Treatment1 SEM2 P2
(n=6)
Control +Enz. +Inoc. +Enz.+ Enz. Inoc. Enz. ×
Inoc. Inoc.
Initial body weight (kg) 64.0 64.7 65.8 65.3 1.94 0.984 0.543 0.748
Final body weight (kg) 118.8 117.9 118.8 118.4 0.51 0.503 0.832 0.776
Gain (kg/d) 1.03 1.12 1.09 1.18 0.03 0.007 0.041 0.948
Feed intake (kg/d)3 2.40 2.53 2.55 2.60 0.05 0.082 0.031 0.391
Feed:Gain3 2.34 2.26 2.35 2.19 0.05 0.022 0.476 0.446
1 Derived from Zhu et al. (2011); 6 pens per treatment with 8 pigs per pen. Liquid feeds were prepared by mixing a
pelleted and crumbled supplement (containing all feed ingredients except DDGS) with water and DDGS; the water
to feed dry matter ratio was 2.5 to 1. The DDGS was mixed with water steeped in one of 4 fermentation tanks: DDGS
and water only (Control), with added xylanase and glucanase (+Enz.), with added Pediococcus inoculant (+Inoc) or
its combination (+Enz+Inoc.). For all pigs additional water was available from nipple drinkers.
2 SEM represents standard error of treatment means; P is the probability of main effects and interaction between
main effects.
3 100% dry matter basis.

water, wheat shorts and corn distillers grains with solubles. In general and when used at 15% or
less of feed dry matter content, corn distillers solubles, and corn steep water does not result in
major changes in pig growth performance, or carcass and meat quality. The feeding value of wheat
shorts appears increased in liquid fed pigs. There is potential to further enhance the feeding value
of feed ingredients for pigs by steeping with enzymes and controlled fermentation. Quality of
liquid feed can be maintained, and the loss of synthetic amino acids in (partly) fermented liquid
feed can be minimized, via control of feed pH and lactic acid content. When using liquid feeding
systems, pigs should have access to an additional source of water. Management of liquid feeding
systems requires computer and engineering skills and attention to detail, especially when using
co-products with variable nutrient content.
Acknowledgements
The swine liquid feeding research program at the University of Guelph is supported by a number
of organizations, including Swine Liquid Feeding Association, Natural Sciences and Engineering
Research Council of Canada, Ontario Pork, the Ontario Ministry of Agriculture Food and Rural
Affairs (OMAFRA) Food Safety Research Program, OMAFRA & University of Guelph Research
Partnership Program, Swine Innovation Porc, AB Agri. Ltd., Agribrands Purina Canada Inc.,
Big Dutchman, B.S.C. Nutrition, Chris Hansen Animal Health and Nutrition, Casco Inc./Corn
Products International, Daco laboratories Inc., Dwyer Manufacturing Ltd., Farmix Ltd., Grand
Valley Fortifiers, Furst McNess Co. Ltd. Great Lakes Nutrition/Wallenstein Feeds, Kenpal Farm
Products Inc. and Lallemand Specialties Inc./Institute Rosell A number of students, post-doctoral
fellows and technical support staff have contributed to this program, including J. Zhu, D. Wey, S.
Niven, M. Or-Rashid, J. Squire, D. Columbus, D. Woods, J. Guimaraes, M. Rudar and K. de Ridder.
References
Beal, J. D., S. J. Niven, A. Campbell, and P. H. Brooks. 2002. The effect of temperature on the growth and
persistence of Salmonella in fermented liquid pig feed. International Journal of Food Microbiology
79:99-104.
Big Dutchman. 2012. Pig feeding systems. http://www.bigdutchman.de/en/pig/products/feeding-systems.
html (accessed February 1, 2012).
Braun, K. and C. F. M. De Lange. 2004. Co-products used in swine liquid feeding: aspects of food safety
and nutritional value. Pages 205-222 in Proc. 2004. Eastern Nutrition Conference. Animal Nutrition
Association of Canada, Ottawa, Canada. K2P 1P1 (and www.slfa.ca).
Brooks, P. H., J. D. Beal, and S. J. Niven. 2001. Liquid feeding for pigs: potential for reducing environmental
impact and improving productivity and food safety. Pages 49-63 in Recent Advances in Animal Nutrition
in Australia, Vol. 13, ed by Corbett J. L. University of New England, Armidale, Australia.
Canibe, N. and B. B. Jensen. 2003. Fermented and nonfermented liquid feed to growing pigs: effect on
aspects of gastrointestinal ecology and growth performance. J. Anim. Sci. 81:2019-2031.
Canibe. N., O. Holberg, J. H. Badsberg, and B. B. Jensen. 2007. Effect of feeding fermented liquid feed and
fermented grain on gastrointestinal ecology and growth performance in piglets. J. Anim. Sc. 85:2959-2971.

Columbus, D., J. Zhu, D. Woods, J. Squire, E. Jeaurond, and C. F. M. De Lange. 2006. On-farm experience
with swine liquid feeding: research unit at Arkell Swine research station at the University of
Guelph. Pages 189-198 in Proc. 2006 London Swine Conference, London, Ontario, Canada (www.
londonswineconference.ca).
Columbus, D., S. J. Niven, C. Zhu, and C. F. M. De Lange. 2010a. Phosphorus utilization in starter pigs fed
high-moisture corn-based liquid diets steeped with phytase. J. Anim. Sci. 88:3964-3976.
Columbus, D., S. J. Niven, C. Zhu, J. R. Pluske and C. F. M. De Lange. 2010b. Body weight gain and nutrient
utilization in starter pigs that are liquid-fed high-moisture corn-based diets supplemented with phytase.
Can. J. Anim. Sci. 90:45-55.
De Lange, C. F. M. 2000. Characterization of the non-starch polysaccharides in feeds. Pages 77-92 in Feed
evaluation – principles and practice. P.J. Moughan, M.W.A. Verstegen and M. Visser-Reyneveld, ed.
Wageningen Pers, Wageningen, the Netherlands.
De Lange, C. F. M., C.H. Zhu, S. Niven, D. Columbus, and D. Woods. 2006. Swine Liquid Feeding:
Nutritional Considerations. Pages 37-50 in Proc. 27th Western Nutrition Conference. Department of
Animal Science, University of Manitoba, Winnipeg, MB, Canada (and www.slfa.ca).
Fontaine, J., U. Zimmer, P. J. Moughan, and S. M. Rutherfurd. 2007. Effect of heat damage in an autoclave
on the reactive lysine contents of soy products and corn distillers dried grains with solubles. Use of
the results to check on lysine damage in common qualities of these ingredients. J. Agric. Food Chem.
55:10737-10743.
Geary, T. M., P. H. Brooks, D. T. Morgan, A. Campbell, and P. J. Russell. 1996. Performance of weaner pigs
fed ad libitum with liquid feed at different dry matter concentrations. J. Sci. Food Agric. 72:17-24.
Gill, B. P. 1988. Water use by pigs managed under various conditions of housing, feeding and nutrition.
Ph.D. Thesis. Seale-Hayne College, Plymouth, UK.
Guimaraes, J., C. L. Zhu, D. Wey and C. F. M. De Lange. 2010. Calcium chloride reduces the negative impact
of feeding high potassium and co-product containing diets to finishing pigs. J. Anim. Sci. Vol. 88 (Suppl.
2; Abstracts):379.
Hampshire, 2012. Liquid feeding systems for pigs. http://www.hampshirefeedingsystems.com (accessed
February 1, 2012).
Jensen, B. B. and L. L. Mikkelsen, L.L. 1998. Feeding liquid diets to pigs. Pages 107-126 in Recent
Advances in Animal Nutrition. P. C. Garnsworthy and J. Wiseman, eds. Nottingham University Press,
Loughborough, UK.
Missotten J. A., J. Michiels, A. Ovyn, S. De Smet, and N. A. Dierick. 2010. Fermented liquid feed for pigs.
Arch. Anim. Nutr. 64:437-66.
MLC (Meat and Livestock commission). 2003. General guidelines on liquid feeding for pigs. British Pig
Executive, Stoneleigh Park, Kenilworth, Warwickshire, CV8 2TL, UK.
MLC (Meat and Livestock commission). 2005. Finishing pigs – systems research. Final report to Defra,
August 2005. British Pig Executive, Stoneleigh Park, Kenilworth, Warwickshire, CV8 2TL, UK.
Niven, S. J., J. D. Beal and P. H. Brooks. 2006a. The effect of controlled fermentation on the fate of synthetic
lysine in liquid diets for pigs. Anim. Feed Sci.Techn. 129:304-315.
Niven, S. J., C. Zhu, D. Columbus, O. Izquirde and C. F. M. De Lange. 2006b. Chemical composition and
phosphorus release of corn steep water during phytase steeping. J. Anim. Sci. 84 (Suppl. 1; Abstracts):429.
Niven, S. J., C. Zhu, D. Columbus, and C. F. M. De Lange. 2007. Impact of controlled fermentation and
steeping of high-moisture corn on its nutritional value for pigs. Livest. Prod. Sci. 109:166-169.
Pedersen, A. O., N. Canibe, I. D. Hansen, and M. D. Aaslying. 2002. Fermented Liquid Feed for Finishers –
Pelleted Feed, vol. 1. The National Committee for Pig Production, Copenhagen.

Russell, P. J., T. M. Geary, P. H. Brooks, and A. Campbell. 1996. Performance, water use and effluent output
of weaner pigs fed ad libitum with either dry pellets or liquid feed and the role of microbial activity in
the liquid feed. J. Sci. Food Agric. 72:8-16.
Scholten, R. H. J., C. M. C. Van der Peet-Schwering, M. W. A. Verstegen, L. A. den Hartog, J. W. Schrama,
and P. C. Vesseur. 1999. Fermented co-products and fermented compound diets for pigs: a review. Anim.
Feed Sci. Tech. 82:1-19.
Scholten, R. 2001. Fermentation of liquid diets for pigs. Ph.D. Thesis. Wageningen University, the
Netherlands.
SLFA (Swine Liquid Feeding Association). 2012. www.slfa.ca (Accessed February 1, 2012).
Squire, J. M. 2005. Fermentation of an alternative feedstuff for use in swine liquid feeding. M.Sc. Thesis.
Department of Animal and Poultry Science University of Guelph. Guelph, ON, Canada.
Squire, J. M., C. Zhu, and C. F. M. De Lange. 2004. Fermentation of an alternative feedstuff for us in swine
liquid feeding: Condensed corn distillers solubles. Can. J. Anim. Sci. 84 (Abstracts):782.
Squire, J. M., C. L. Zhu, E. A. Jeaurond, and C. F. M. De Lange. 2005. Condensed corn distillers’ solubles in
swine liquid feeding: growth performance and carcass quality. J. Anim. Sci. 83 (Suppl. 1; Abstracts):165.
Van der Wolf, P. J., W. B. Wolbers, A. R. W. Elbers, H. M. J. F. Van der Heijden, J. M. C. C. Koppen, W. A.
Hunneman, F. W. van Schie, and M. J. M. Tielen. 2001. Herd level husbandry factors associated with the
serological Salmonella prevalence in finishing pig herds in the Netherlands. Vet. Microbiol. 78:205-219.
Van Winsen, R. L., B. A. P. Urlings, L. J. A. Lipman, J. M. A. Snijders, D. Keuzenkamp, J.M.H. Verheijden,
and F. van Knapen. 2001. Effect of fermented feed on the microbial population of the gastrointestinal
tracts of pigs. Appl. Environ. Microbiol. 67:3071-3076.
Weda. 2012. Liquid feeding. http://www.weda.de/Produkte/Fluessigfuetterung.aspx (accessed February 1,
2012).
Woods, T. D. 2008. Liquid feeding of newly weaned pigs with corn and soybean meal based diets. M.Sc.
Thesis. Department of Animal and Poultry Science University of Guelph. Guelph, ON, Canada.
Zhu, C. H., D. Wey, R. Friendship, and C. F. M. De Lange. 2010. Impact of fineness of grinding of corn and
high-moisture corn on growth performance, carcass characteristics and stomach lesions in liquid-fed
growing-finishing pigs. Proc. 21st International Pig Veterinary Society Congress. July 2010, Vancouver,
Canada.
Zhu, C. L. M. Rudar, D. Wey, and C. F. M. De Lange. 2011. Glucanase, xylanase and microbial inoculants
improve feeding value of DDGS for liquid-fed finishing pigs. J. Anim. Sci. 89 (Suppl. 1; Abstracts):78.

4. Amino acid nutrition and feed efficiency
C.F.M. de Lange1, C.L. Levesque1 and B.J. Kerr2
1Department of Animal and Poultry Science, University of Guelph, 50 Stone Road East, Guelph,
ON N1G 2W1, Canada; cdelange@uoguelph.ca

2Agricultural Research Service, US Department of Agriculture, 2110 University Boulevard, Ames,
IA 50011-3310, USA
Abstract
In growing pigs, whole body protein deposition (PD) is the main determinant of dietary amino
acid requirements and is closely associated with lean tissue growth, feed efficiency, and carcass
quality. In North America, the typical mean PD for barrows and gilts between 25 and 125 kg body
weight is approximately 135 g/d, but PD is known to be highly variable between and within groups
of pigs. The use of dietary amino acids for PD involves digestion, absorption, and post-absorptive
metabolism; biological processes which are all influenced by factors associated with the animal
(pig genotype, physiological state, and health status) and the environment (diet composition,
thermal and physical environment) and should be considered when establishing optimum
dietary amino acid levels for groups of pigs. Mathematical models, such as the NRC (2012), are
now available in which the biology of dietary amino acid utilization for PD is represented which
can be used to estimate amino acid requirements of groups of pigs maintained in a relatively
stress and disease-free environment. For predicting amino acid requirements of groups of pigs,
between-animal variation in growth performance and nutrient utilization should also be taken
into account. In addition, nutritional history and compensatory growth should be considered
when establishing optimum dietary amino acid levels for maximum profit. Some examples are
provided in this chapter to illustrate the effect of gender and feeding high-fiber co-products on
amino acid requirements of growing-finishing pigs.
Introduction
In growing pigs, whole body protein deposition (PD) is the main determinant of dietary amino
acid requirements and is closely associated with lean tissue growth, feed efficiency, and carcass
quality (NRC, 2012). Therefore, relationships among amino acid intake, PD, and pig growth
performance must be considered carefully when optimizing feeding programs for individual
growing-finishing pig barns. For the development of phase-feeding programs and for determining
optimum slaughter weights, changes in PD and feed intake with increasing body weight (i.e. PD
and feed intake curves) should also be considered.
The utilization of dietary amino acids for PD involves digestion, absorption, and post-absorptive
metabolism; biological process which are all influenced by factors associated with the animal (pig
genotype, physiological state, and health status) and the environment (diet composition, thermal,
and physical environment; Moughan, 1999; Rakhshandeh and De Lange, 2011; NRC, 2012).
Amino acids also serve as precursors for bioactive compounds that influence gene expression
and endocrine function (Averous et al., 2003). Optimum dietary amino acid levels will, therefore,

vary among groups of pigs, and an understanding of the biological principles that contribute to
amino acid utilization is required for optimizing feeding programs.
In this chapter, determinants of dietary amino acid requirements of growing- finishing pigs are
briefly reviewed. In addition, practical considerations for establishing optimum dietary amino acid
levels for groups of growing-finishing pigs are discussed, including between-animal variability,
phase-feeding, compensatory growth, and cost-benefit analyses. Finally, the mathematical
modeling approach that is described in the recent publication Nutrient Requirements of Swine
(NRC, 2012) is used to estimate amino acid requirements of growing pigs at various levels of
performance and under varying environmental conditions. The approach used in NRC (2012) is
a revision of NRC (1998) and is based on an extensive review of the scientific literature.
Whole body protein deposition and pig growth performance
Given the varying definitions of (fat-free) lean tissue growth and potential biases in its estimation
from carcass measurements, PD is considered a more objective and universal measure than lean
tissue growth, and may be used to characterize growth rates and growth potentials of pigs (NRC,
2012). In North America and according to NRC (2012), the typical mean PD for barrows and
gilts between 25 and 125 kg body weight is approximately 135 g/d. However, PD is highly variable
among and within groups of pigs and has been reported to exceed 200 g/d in entire male pigs of
superior genotypes that are managed in a stress and disease-free environment (Schinckel and De
Lange, 1996; NRC, 2012).
The data presented in Table 1 show that a 10% increase in mean PD represents an improvement
in feed efficiency of approximately 5.5% and an increase in estimated carcass lean yield of just
Table 1. Estimated effect of a 10% improvement in whole body protein deposition (PD) on growth performance
and carcass characteristics in growing-finishing pigs between 25 and 125 kg body weight.
Mean PD (g/d)
135 148.5
ME intake2 (kcal/d) 7,231 7,239

Feed intake (kg/d) 2.31 2.31
Body weight gain (g/d) 859 907
Body lipid deposition (g/d) 261 251
Feed:Gain 2.69 2.54
Probe back fat (mm)1 18.3 16.1
Carcass lean content (%)1 49.0 50.6
1 Estimated based on the partitioning of energy intake for growth and prediction of carcass characteristics as outlined
in NRC (2012).
2 ME = metabolizable energy.

over 3%. For generating these estimates it is assumed that energy intake is not affected by PD
and that intake of essential nutrients, including amino acids, does not limit pig performance.
These data illustrate the importance of improving the genetic capacity of the pig for PD, and of
optimizing environmental conditions and feeding programs to allow expression of this genetic
capacity. Impacts of pig genotypes and environmental conditions on pig growth performance are
described in detail elsewhere in this book and are not addressed in this chapter.
It is well established that entire males have a higher mean PD than gilts and barrows, and that entire
males are able to maintain high rates of PD up to higher body weights (Figure 1). Gender effects
on PD curves, combined with feed intake curves, provide the basis for the development of split-
gender feeding programs. They also provide an incentive to use entire males for pork production,
as reported by Dunshea (2012) in Chapter 13 of this book. It should be stressed, however, that
the effect of gender on PD curves varies among lines of pigs (Schinckel and De Lange, 1996). It
has also been established that at lower body weights, PD is generally not determined by the pig’s
genetic capacity for PD, but by intake of either energy or essential nutrients. Moehn and De Lange
(1998) and Moughan et al. (2006) suggested that the capacity for PD is relatively constant and
independent of body weight up to approximately 80 kg body weight, after which it will gradually
decline towards zero as pigs reach maturity. Therefore, in starting and growing pigs that are fed
diets that are not limiting in essential nutrients, energy intake appears to limit expression of
maximum PD, and these pigs are in the so-called ‘energy intake dependent’ phase of PD as long
as PD increases with body weight (Figure 1). Whether PD is determined by the pig’s maximum
180
Whole body protein deposition (g/d)
160
140
120
100
80
Entire males
60 Gilts
Barrows
40
25 50 75 100 125 150
Body weight (kg)
Figure 1. Typical whole body protein deposition curves in barrows, gilts and entire male pigs according to
NRC (2012).

PD or energy intake may have consequences for the rate and extent of compensatory growth,
which will be discussed later in this chapter.
Estimates of PD may be obtained by closely monitoring growth rates and changes in chemical
body composition; the latter being estimated from physical body measurements such back fat
thickness and loin depth (Schinckel and De Lange, 1996; Schinkel et al., 2010). Alternatively,
estimates of PD may be derived from accurate measurements of energy intake and live body
weight gain, and modeling the partitioning of available energy intake over and above maintenance
energy requirements between PD and whole body lipid deposition (LD) (Van Milgen et al., 2008;
NRC, 2012). For this modeling approach, assumptions are required about gut fill, maintenance
energy requirements, energy requirements for PD and LD, as well as relationships between PD
and accretion of water and ash (De Lange et al., 2003; Van Milgen et al., 2008). This modeling
approach is based on the principles that accretion of water and ash in the pig’s body is closely
related to PD, and that estimates of PD are obtained by matching observed live body weight gain
with model estimated live body weight gain. Live body weight gain is estimated from changes in
gut fill, PD, LD, water, and ash accretion; LD is predicted from partitioning of available energy
intake over and above maintenance energy requirements between PD and LD. This modeling
approach to estimate PD at the various stages of growth has been used in NRC (2012) to interpret
a large number of conventional, empirical amino acid requirements studies, and was used to
model amino acid requirements of growing pigs.
Biology of amino acid utilization in growing pigs
The main biological processes that influence amino acid utilization and contribute to dietary amino
acid requirements for PD in growing pigs are summarized in Table 2. Estimated requirements
for lysine, other key amino acids, and nitrogen (N) for some of these processes are presented in
Table 3; values which can be used to model the dietary amino acid requirements of individual pigs
in a relatively disease and stress-free environment, and at varying levels of feed intake and PD.
These processes are described briefly in this section. For a more detailed description the reader
is referred to Moughan (1999), Stein et al. (2007) and NRC (2012).
Standardized ileal digestibility (SID) coefficients are used routinely to estimate the bioavailability
of amino acids in swine feed ingredients (Stein et al., 2007). Estimates of SID coefficients for
amino acids in a wide range of feed ingredients are summarized in NRC (2012). The SID values
reflect that amino acids that disappear from the hindgut are not available for metabolism by
the pig – hence the term ‘ileal digestibility’ – and are corrected for the basal ileal endogenous
amino acid losses – hence the term ‘standardized’. The SID values are corrected for the basal
ileal amino acid losses only, largely because it is technically too difficult to routinely measure
total ileal endogenous amino acid losses (Stein et al., 2007). The main benefit of using SID over
conventional apparent ileal digestibility coefficients is that SID values are more likely to be
additive among feed ingredients, which is an important prerequisite in diet formulation (Stein
et al., 2007). It should be noted that in some instances SID values overestimate bioavailability of
amino acids (Columbus and De Lange, 2012). This occurs, for example, when amino acids are
digested and absorbed in a chemical form that renders them unavailable for metabolism, which
applies in particular to lysine that is present in heat treated feed ingredients such as dried milk

Table 2. Main biological processes that determine amino acid utilization for whole body protein deposition
in growing pigs.
Item Short description
Bioavailability of dietary amino acids and Proportion of ingested amino acids and nitrogen
nitrogen that can be digested and absorbed in a chemical
form that renders them available for metabolism,
including protein synthesis.
Endogenous gut amino acid losses Net loss of endogenous amino acids into the gut,
representing the balance between secretions into
gut (to facilitate nutrient digestion and absorption,
and maintain gut integrity) and re-absorption.
Fermentative amino acid losses Loss of amino acids due to fermentable fiber intake
induced microbial fermentation.
Integument amino acid losses Amino acids that are lost with skin and hair.
Amino acid needs for synthesis of non- Use of amino acids for the production of non-protein
protein compounds compounds that are required for essential body
functions. Examples are the use of tryptophan for
producing serotonin and cysteine for producing
glutathione.
Minimum amino acid catabolism Minimum rate of breakdown of amino acids, which is
associated with whole body protein turnover, and
occurs even when pigs are fed amino acid-free diets.
Inevitable amino acid catabolism Minimum rate of breakdown of absorbed available
dietary amino acids, which determines the maximum
marginal efficiency of using absorbed and available
amino acid intake for amino acid retention in body
protein.
Preferential amino acid catabolism The breakdown and utilization of amino acids as a
source of energy when energy intake is insufficient
to support the pig’s desired rate of body protein
deposition.
Whole body protein deposition Retention of amino acids in deposited body protein.
products and dried distillers grains plus solubles (DDGS). Fontaine et al. (2007) showed that
more than 15% of lysine in DDGS is non-reactive, which is a reflection of chemical unavailability.
Another example is the negative impact of fermentable fiber intake on threonine bioavailability
(Zhu et al., 2005; Libao-Mercado et al., 2006, 2009), which may be attributed to the impact of
fiber-induced enteric microbial fermentation and the use of high threonine containing mucus
proteins as nitrogen and energy sources for enteric microbes (Columbus and De Lange, 2012).

Table 3. Estimated lysine (Lys), methionine (Met), methionine plus cysteine (Met+Cys), threonine (Thr),
tryptophan (Trp) and nitrogen (N) requirements for various biological functions, including whole body
protein deposition (PD).1
Item Lys Met Met+Cys2 Thr Trp N3
Basal gut losses (g/kg dry matter intake) 0.459 0.129 0.197 0.242 0.059 1.034
Fermentative losses (mg/g fermentable 0 0 0 4.2 0 0
fiber intake)
Integument losses (mg/kg body weight0.75) 4.50 1.05 5.76 3.35 0.94 16.7
Minimum+inevitable catabolism (% of 25.0 27.0 39.7 22.04 39.0 15.0
available intake)
Content in PD5 (g/kg) 71.0 20.0 30.4 37.5 9.1 160.0
1 Derived from NRC (2012); these values allow for the estimation of standardized ileal digestible amino acid intake
requirements for PD of individual growing pigs in a relatively disease and stress-free environment.
2 Cysteine is considered a conditionally essential amino acid and can be synthesized from methionine only. Therefore,
methionine may satisfy requirements for cysteine and diets should be formulated to meet requirements for both
methionine and methionine plus cysteine.
3 Nitrogen may be required for the synthesis of non-essential amino acids when insufficient amounts are supplied in
diets. The rate of inevitable catabolism of N-containing compounds is not well understood, but in corn and soybean
meal based diets N intake is likely to become limiting before some essential amino acids, such as phenylalanine and
leucine.
4 This represents the rate of inevitable threonine catabolism when pigs are fed diets devoid of fermentable fiber;
increased intake of fermentable fiber will reduce the post-absorptive efficiency of threonine utilization.
5 A different content of amino acids in PD is assumed for ractopamine induced PD.
Endogenous gut amino acid losses, integument amino acid losses, and minimum amino acid
catabolism are the main contributors to maintenance amino acid requirements. Under some
conditions the use of some amino acids for the production of non-protein compounds may be
quantitatively important. The latter applies, for example, to tryptophan and methionine plus
cysteine requirements of pigs during inflammation (Le Floc’h et al., 2009; Rakhshandeh, 2011;
Rakhshandeh and De Lange, 2011). Currently, insufficient quantitative information is deemed
available to generate reasonable estimates of increases in maintenance requirements for these
amino acids in pigs that are exposed to disease causing organisms. Traditionally, maintenance
amino acid requirements have been related to the metabolic body weight of pigs (i.e. (kg body
weight)0.75; NRC, 1998). Unfortunately, this traditional approach ignores endogenous gut amino
acid losses which are the main contributor to maintenance amino acid requirements. Therefore,
dietary factors such as feed dry matter intake have a major impact on maintenance amino acid
requirements (Moughan, 1999; Van Milgen et al., 2008; NRC, 2012). Moreover, traditional
estimates of maintenance amino acid requirements are generally lower than endogenous ileal
amino acid losses, in particular for lysine (NRC, 2012).

The inefficiency of utilizing available amino acid intake over and above maintenance amino acid
requirements for amino acid retention in PD can be attributed largely to inevitable amino acid
catabolism (Moughan, 1999; Tables 2 and 3). When obtaining estimates of inevitable catabolism,
care should be taken to differentiate it from preferential catabolism, which occurs when energy
intake limits PD, and catabolism of amino acids that are supplied in excess of requirements. The
latter is of concern when pigs are fed constant dietary levels of amino acids over a wide body
weight range, because dietary amino acid requirements – expressed as dietary concentrations –
decline with increasing body weight. Estimates of inevitable catabolism of lysine and threonine
have been obtained from observations on individual pigs and in closely controlled serial slaughter
studies (e.g. Bikker et al., 1994; Moehn et al., 2000; De Lange et al., 2001b; Moehn et al., 2004),
and are about 25% of available amino acid intake. Based on these studies, the rate of inevitable
amino acid catabolism is rather constant over a wide range of amino acid intake levels (Figure
2), appears to be largely independent of BW, and increases slightly with improvements in pig
performance potential. Basically, for every 1 g increase in maximum PD, the rate of inevitable
lysine catabolism is reduced by 0.2% (Moehn et al., 2004; NRC, 2012). Estimates of inevitable
catabolism of amino acids other than lysine and threonine may be obtained by carefully fitting
requirements observed in empirical amino acid requirements studies with model estimated
requirements whereby all other determinants of amino acid requirements are considered (NRC,
2012). It should also be noted that the efficiency of amino acid utilization for PD is always lower
in groups of pigs than in individual pigs, which can be attributed largely to between-animal
variability in amino acid requirements and will be discussed in more detail later in this chapter.
40
36.9
SEM = 1.1
Total threonine catabolism
(% of available intake)
30
24.3 25
23.5
21.8
20
15.1
10
0
60 70 80 90 100 120
Threonine intake (% of estimated requirements)
Figure 2. Estimates of inevitable threonine catabolism in growing pigs at varying threonine intake levels;
values are derived from serial slaughter observations on individually housed pigs that were fed casein and
cornstarch based diets (De Lange et al., 2001b). Observed catabolism at the highest threonine intake level also
includes catabolism of threonine intake above requirements for maximum whole body protein deposition.

Based on a review of the available literature, NRC (2012) estimated the mean amino acid
composition of PD in growing pigs (Table 3). This amino acid composition reflects amino acid
profiles of a wide range of unique proteins (e.g. actin and myosin in muscle tissue, collagen in
connective tissue, and skin) and the relative contribution of these proteins to PD. In situations
where the contribution of specific proteins to PD is increased, the amino acid composition of
PD will be altered. The latter applies especially to pigs that are fed ractopamine (Paylean™) to
increase muscle protein accretion, and is accounted for in NRC (2012) by using a unique amino
acid profile for ractopamine induced PD. Additional information is needed to develop a more
mechanistic approach for representing pig type and environmental effects on variation in the
amino acid composition of PD.
Effect of between-animal variability on optimum dietary amino acid levels for

groups of pigs
In commercial pig barns, the coefficient of variation for average daily gain among individual
pigs at similar body weights is often more than 10% (Pomar et al., 2003; Morel et al., 2008). This
reflects variability in feed intake and PD within groups of pigs and, quite likely, variability in
nutrient and energy requirements for maintenance and other body functions. As a consequence,
there will always be between-animal variation in nutrient requirements for maximum growth
performance, which must be considered when establishing the optimum dietary amino acid
levels for groups of pigs.
Between-animal variability is one of the main contributors to the diminishing marginal efficiency
of amino acid utilization for growth when dietary amino acid levels approach requirements for
maximum performance in groups of pigs (Baker, 1986; Moehn et al., 2000; De Lange et al., 2001a).
This is illustrated in Figure 3, where the typical growth responses of groups of pigs to varying
amino acid intake levels (Figure 3a) is contrasted with the response to varying threonine intake
levels observed in individually housed pigs (Figure 3b; De Lange et al., 2001b). A diminishing
marginal efficiency of nutrient utilization implies that profit is not maximized when feeding dietary
amino acid levels required for maximizing growth performance of groups of pigs. For maximum
profits, the optimum dietary amino acid level should be chosen based on maximizing the difference
between variable costs (determined by levels of expensive amino acids, such as lysine, in the diet)
and generated value (determined by changes in growth rate, feed efficiency, and carcass quality).
Feeding strategies for maximizing profits may also be established by using stochastic models
in which between-animal variability and cost-benefit analysis are considered explicitly (Pomar
et al., 2003; Morel et al., 2008). However, the use of such models requires substantially more
computing power than conventional deterministic models that represent the average for a group
of animals. Moreover, not only variability in the various traits (PD, feed intake, maintenance
energy, and nutrient requirements, etc.), but also co-variation among these traits should be
considered. Estimates of co-variation are generally not available or not very reliable (Pomar et
al., 2003; Morel et al., 2008).
Some relatively simple principles may be adopted to accommodate between-animal variability

for established feeding programs. For example, in Figure 4 the relationship between dietary lysine

a: Groups of pigs and extended time periods
PD (g/d)
Amino acid intake
b: Individual pigs and short time periods

160
140
120
PD (g/d)
100
80
SE 4.4; 40-70 kg BW
60
40 60 80 100 120 140
Threonine intake (% of estimated requirements)
Figure 3. Typical relationship between dietary nutrient levels and growth responses observed in: (a) groups
of animals and (b) a dose-response study were pigs were housed individually (De Lange et al., 2001b; data
presented in Figure 2 are from the same study).
level and the proportion of pigs of which dietary lysine requirements are met is illustrated for
groups of pigs with coefficients of variation for PD of 10 and 20%. This figure shows that when
the coefficient of variation for PD is 10%, the dietary lysine level should be increased by 5% to
meet the lysine requirements of 70% of the pigs, relative to levels that are needed to meet the
requirements of the average pig within the group. This increase, or adjustment, is 10% when the
coefficient of variation for PD is 20%. Based on financial analyses with a stochastic pig growth
model, Pomar et al. (2003) suggested that dietary nutrient levels should be targeted to meet the
nutrient requirements of approximately 70% of the pigs. Therefore, an adjustment may be applied
to the estimated amino acid requirements of the average pig in a group, in order to establish

7.5
CV 10%
CV 20%
Available lysine content (% of diet)
7.0
6.5
6.0
5.5
5.0
0 20 40 60 80 100
Proportion of pigs for which lysine requirements are met (%)
Figure 4. Impact of coefficient of variation for whole body protein deposition (PD; 10 vs. 20%) on the
relationship between dietary standardized ileal digestible lysine level and the proportion of pigs for which
lysine requirements for PD are met (body weight is 80 kg; PD is 140 g/d; feed intake is 90% of voluntary daily
digestible energy intake according to NRC, 1998; diet digestible energy content 14 MJ/kg).
amino acid requirements of groups of pigs. This adjustment should reflect both the amount
of between-animal variability and the proportion of pigs for which nutrient requirements are
to be met.
Implications of phase-feeding and compensatory growth for establishing

optimum dietary amino acid levels
Phase-feeding schemes are commonly used to accommodate continuous changes in amino acid
requirements of growing pigs (Figure 5). In phase-feeding programs, periods of feeding pigs
below estimated amino acid requirements, when pig growth performance is compromised, need
to be balanced with periods of feeding pigs above requirements, when costly amino acids are
wasted and contribute to increased N excretion. However, for the development of phase-feeding
programs, the effect of under-feeding on potential compensatory growth during subsequent
periods of over-feeding should be considered. The latter is illustrated in a large-scale pig growth
performance study involving 1,024 pigs that was conducted at the Stotfold Pig Development
Unit in the UK (Table 4; MLC, 2004). In that study, feeding one constant dietary level of protein
(in this case representing protein with a constant and close to optimal amino acid balance for
growing pigs) throughout the growing-finishing phase was compared to blend feeding, whereby
the dietary protein levels were adjusted daily to match previously established requirements, by
changing the contribution a high protein containing diet and a low protein containing diet to
the delivered feed. These two diets were also used to prepare the feed with the constant dietary

Over-feeding: waste of nutrients

Dietary nutrient level (%)
Under-feeding: reduces
animal performance Requirements
3 phase-feeding
20 30 40 50 60 70 80 90 100 110 120

Body weight (kg)
Figure 5. The concept of phase-feeding, illustrating the continuous change in the pigs’ nutrient requirements
(% of diet), the constant dietary nutrient levels in a three phase-feeding program, and the periods where
pigs are fed below or above nutrient requirements. Compensatory growth may occur immediately following
periods where pigs are fed below requirements.
Table 4. Effect of feeding one diet with a constant dietary protein level or blend feeding to meet daily changes
in amino acid requirements on growth performance and carcass characteristics of growing-finishing pigs.1
Feeding one diet Blend feeding Significance
Feed intake (kg/d) 2.06 1.99 P<0.10

Gain (g/d) 886 860 P<0.10
Feed:Gain 2.36 2.36 NS
Carcass weight (kg) 75.7 74.5 NS
Back fat (mm) 11.67 11.53 NS
UK cost of production (US $/kg carcass) 1.73 1.86 +8%
1 Adjusted from MLC (2004). In blend feeding the dietary amino acid levels were adjusted daily to match previously
established requirements by changing the contribution a high protein diet and a low protein containing diet to the
delivered feed. These two diets with extreme protein levels were also used to prepare the diet with the constant
dietary protein level. The cumulative and per pig intake of protein was identical for the two treatments. NS = not
found to be significant (e.g. P>0.10).

protein level, to eliminate confounding of feeding scheme with dietary ingredient composition.
The cumulative and per pig intake of protein was identical for the two treatments.
The observations presented in Table 4 indicate that in this case there is no need to apply phase-
feeding of growing-finishing pigs and that phase-feeding increases production costs, largely
because of extra feeding equipment requirements. These findings are consistent with research
conducted at the University of Guelph where it was shown that the rate and extent of compensatory
growth following a period of amino acid intake restriction varies with pig type (Martinez and
De Lange, 2007). Entire male pigs with high capacity for PD, such as these in the MLC (2004)
study, showed complete compensatory growth following a period of reduced growth induced by
amino acid intake restriction (Table 5; Martínez-Ramírez et al., 2008). In contrast, compensatory
growth was not observed in barrows with medium capacity for PD. These findings indicate that
the capacity of the pig for PD represents a limit to the extent of compensatory growth. It can
be hypothesized that compensatory growth occurs only during the energy intake dependent
phase of PD and is driven by a genetically determined target body composition, which can be
represented by a target ratio between whole body lipid mass and whole body protein mass (L:P;
Moughan et al., 1987; Weis et al., 2004; Martínez-Ramírez and De Lange, 2007). In other words,
during a period of amino acid intake restriction, PD is reduced and body lipid deposition (LD)
is increased, resulting in actual L:P exceeding target L:P. Following a period of amino acid intake
restriction, pigs will have a reduced need for LD and can direct more energy towards PD, simply
because actual L:P exceeds target L:P. Increases in PD, however, will only occur when PD is not
constrained by the capacity of the pig for PD. Therefore, compensatory growth is more likely to
occur during the energy intake dependent phase of PD. There is no evidence to suggest that the
rate of inevitable amino acid catabolism is reduced during compensatory growth (Martínez-
Ramírez and De Lange, 2007). Therefore, it can be assumed that the amino acid requirements
per g of PD are not altered during compensatory growth. These concepts can be represented in
dynamic pig growth models, in which daily changes in the composition of both growth and body
composition are represented (Moughan et al., 1987, 1995; De Lange et al., 2001a). The potential
benefits of capitalizing on compensatory growth are: (1) reduced need for phase-feeding; (2)
reduction in overall feeding costs; and (3) improvements in health of pigs. Feeding costs may
be reduced by restricting amino acid intake during early stages of growth, which is generally
when dietary protein and amino acids are most expensive, and increasing amino acid intake
at later stages of growth, in order to maintain the cumulative per pig intake of amino acids
constant. Health of young pigs may be improved by avoiding high dietary protein intakes when
it may compromise digestive function and the stability of gut microflora (e.g. Nyachoti et al.,
2006; Jeaurond et al., 2008). However, until more information is available care should be taken
when capitalizing on compensatory growth. It requires characterization of pig types in aspects
of nutrient partitioning of growth (Black et al., 1986; Schinckel and De Lange, 1996; Van Milgen
et al., 2008), and in particular in defining up to what body weight pigs are in the energy intake
dependent phase of PD and target L:P for the various types of pigs. The concept of compensatory
growth suggests that it may be cost-effective to feed pigs below their amino acid requirements
during early stages of growth and that previous nutrition should be considered when optimizing
dietary amino acid levels.

Table 5. Growth performance of entire male pigs during (15 to 38 kg body weight) and following (38 kg to
final body weight) a period of amino acid intake restriction.1
Dietary treatment Significance
Control -30% lysine during

restriction phase
Body weight (kg)

Initial 15.5 16.2 NS
End of restriction phase 38.4 37.7 NS
Final 110.9 112.0 NS
Body weight gain (g/d)
Restriction phase 784 650 P<0.01
Following restriction phase 1,077 1,170 P<0.01
Overall 990 995 NS
Gain:Feed
Restriction phase 0.626 0.509 P<0.01
Following restriction phase 0.470 0.512 P<0.01
Overall 0.499 0.512 P=0.03
Body composition at final body weight
% dissection lean 50.3 49.5 NS
% dissected fat 11.8 11.7 NS
1 Adjusted from Martínez-Ramírez et al. (2008) and based on using entire male pigs with high capacity for whole
body protein deposition. Pigs on control were fed non-limiting diets between initial and final body weight, whereas
pigs that were fed lysine limiting diets during the restriction phase were fed non-limiting diets thereafter. NS = not
found to be significant (e.g. P>0.10).
NRC approach to estimating amino acid requirements of growing-finishing pigs
Brief description, including features and limitations
The model described in NRC (2012) for estimating nutrient requirements of growing-finishing
pigs is a refinement of the model from NRC (1998). In NRC (2012) pig performance levels or
potentials are characterized using PD curves, which can be defined either by the model user,
predicted from energy intake, or estimated from observed growth performance and carcass
quality. Rates of LD and body weight gains are predicted based on the partitioning of energy
intake, which can be defined on a digestible, metabolizable, or net energy basis. Amino acid and
N requirements are estimated from PD, body weight, and feed intake, as outlined earlier in this
chapter. In the NRC (2012) model, amino acid requirements can be expressed as SID, apparent
ileal digestible, or total dietary levels; the latter being applicable to corn and soybean meal-based
diets only. Amino acid requirements can be explored on individual days or across days and body

weight ranges. The model includes a data base of feed ingredients and a simple feed formulation
system, which allow for comparison of estimated nutrient requirements with nutrient levels in
formulated diets.
For estimating amino acid requirements, there are two main differences between NRC (1998)
and NRC (2012): the approach to estimating maintenance amino acid requirements and amino
acid requirements per g PD. The approach to estimating maintenance amino acid requirements
is described earlier in this chapter (Tables 2 and 3). In NRC (1998), amino acid requirements per
g PD were assumed constant and independent of both body weight and the pig’s maximum PD,
whereas in NRC (2012) these requirements increased with body weight and slightly decreased with
increases in the maximum PD of the pig. The effect of body weight on amino acid requirements
per g PD are represented by gradually increasing the adjustment for the impact of between-animal
variability on amino acid requirements (see previous sections), from approximately 10% at 20 kg to
approximately 30% at 125 kg body weight. By implementing this varying adjustment for between-
animal variability, the underestimation of estimated lysine requirements of pigs between 80 and
120 kg body weight that was noted in NRC (1998) has been addressed. The relationship between
maximum PD and amino acid requirements per g PD was described in detail earlier in this chapter.
Additional refinements in the NRC (2012) model are the prediction of body weight gain from
PD and LD, and the representation of the effects of level and duration of feeding ractopamine
or immunization of entire males pigs against gonadotrophin releasing factor (GnRF; a means to
control boar tainted meat) on feed intake, growth performance, and amino acid requirements.
Results of model testing are included in NRC (2012) and an example is given in Figure 6. This
figure shows good agreement between model estimated SID lysine requirements and observed
requirements across 24 carefully conducted lysine requirement studies that involved groups of
pigs. In addition, there appears to be no systematic bias associated with dietary lysine level or
body weight (Figure 6; NRC, 2012). This indicates that the NRC (2012) model can be used with
a reasonable amount of confidence for estimating amino acid requirements of different groups
of pigs in a relatively stress and disease-free environment.
As is the case for any model, it is important to recognize its boundaries and limitations. In the
NRC (2012) model, boundaries for model use are controlled largely by the imposed minimum
and maximum values for model inputs. Even though the model includes some routines to predict
feed intake and to represent the impact of environmental conditions on estimated nutrient
requirements, these routines are a highly simplified representation of reality, being largely
intended for educational purposes and should be interpreted with caution when extrapolated to
commercial conditions. The model is fairly easy to use, but some understanding of the underlying
principles is required for effective model application. This applies in particular when choosing
the more complex options to characterize PD curves, or when manipulating feed intake, PD, and
other model inputs to match model predicted with observed growth performance and carcass
quality. When observed performance is similar to model predicted performance, the user can
have increased confidence in the model generated estimates of nutrient requirements. Three
limitations of the NRC (2012) model that relate directly to estimating amino acid requirements
are: (1) effects of nutritional history and compensatory growth on amino acid requirements are
not considered; (2) the model cannot be used to assess the marginal response to varying amino

1.6
Observed
1.4 Estimated
SID lysine requirements
1.2
1.0
(% of diet)
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100 120
Body weight (kg)
Figure 6. Estimated (NRC, 2012) and observed standardized ileal digestible lysine requirements of growing
pigs; see NRC (2012) for further details.
acid intake levels and, therefore, to conduct cost-benefit analyses; and (3) the effect of differences
in between-animal variability among groups of pigs is not considered. The potential implications
of these three limitations were described earlier in this chapter.
Estimates of amino acid requirements – effect of gender and dietary fiber level
Typical amino acid requirements of barrows and gilts at different body weight ranges according to
NRC (2012) are summarized in Table 6, These requirements are higher than the typical requirements
presented in NRC (1998; Table 10-1; SID lysine requirements 0.66% and 0.52% for 50 to 80 and 80
to 120 kg body weight, respectively), and reflect increases in maximum PD and reductions in feed
intake in pigs since 1998, as well as adjustments to the manner in which amino acid requirements
are derived. In a direct comparison at identical levels of feed intake and PD, estimated lysine
requirements according to NRC (2012) are slightly lower at the lower body weight and slightly
higher at higher body weights than NRC (1998; Table 6). The latter is consistent with the higher
estimate of maintenance amino acid requirements and increasing lysine requirements per g of PD
with body weights in NRC (2012). Across the three body weight ranges that are presented in Table
6, the difference in lysine requirements between gilts and barrows is slightly lower according to NRC
(2012) than NRC (1998), which can be attributed the positive relationship between PD and the
efficiency of using amino acids for PD in NRC (2012). These analyses suggest that for typical levels
of growing-finishing pig performance, differences in lysine requirements between NRC (1998) and
NRC (2012) are 8% or less, and largest at the highest body weight.

Table 6. Typical amino acid and nitrogen requirements of ad libitum fed barrows and gilts at various body
weight ranges.1
Body weight range (kg) 50-75 kg 75-100 kg 100-135 kg
Barrows Gilts Barrows Gilts Barrows Gilts
Feed intake + wastage (g/d) 2,323 2,124 2,744 2,524 3,029 2,842
Body weight gain (g/d) 917 866 936 898 879 853
Body protein deposition (g/d) 145 145 139 144 119 126
Standardized ileal digestible requirements (% of diet)
Lysine 0.81 0.87 0.69 0.77 0.58 0.64
Methionine 0.23 0.25 0.20 0.22 0.17 0.18
Methionine + cysteine 0.46 0.49 0.40 0.44 0.34 0.37
Threonine 0.50 0.53 0.44 0.48 0.38 0.42
Tryptophan 0.14 0.15 0.12 0.13 0.10 0.11
Nitrogen 1.76 1.88 1.54 1.69 1.31 1.43
Standardized ileal digestible requirements (% of diet) according to NRC (1998) but at levels of
performance according to NRC (2012)
Lysine 0.82 0.90 0.68 0.76 0.54 0.61
1 Derived from NRC (2012: Table 16-1A) and at typical levels of performance (Figure 1); assumed feed wastage is 5%
and diet NE content is 2,475 kcal/kg.
A novel aspect of NRC (2012) is that effects of dietary energy density and fermentable fiber
level on amino acid requirements are considered. This aspect will become increasingly relevant
as more high-fiber containing feed ingredients are used and dietary nutrient density is varied
to minimize cost of dietary nutrients and energy. The results presented in Table 7, for example,
show that the inclusion of 20% wheat shorts and 30% corn DDGS (6-9% oil) in the diet of 50
to 75 kg pigs has only a minor impact on daily SID lysine requirements, while it increases daily
threonine requirements by about 6%. In this example it is assumed that pigs consume feed to
achieve daily net energy requirements, and that body weight gain and PD are not altered when
dietary energy density is reduced as a result of including co-products in the diet. The increase
in daily threonine requirements can be attributed to increases in endogenous gut protein losses,
which are induced by increased intake in pigs on reduced energy density diet, and increased fiber-
induced fermentative threonine losses. These concepts have been described in detail earlier in this
chapter and were not considered in NRC (1998). This example clearly shows the interactive effect
of feeding level and diet fermentable fiber content on the optimum dietary threonine to lysine
ratio, which should be considered when using co-products in pig diets; it also supports the use
of the NRC (2010) model for estimating amino acid requirements for individual groups of pigs.

Table 7. Effect of using co-products on amino acid requirements of growing-finishing pigs between 50 and
75 kg body weight.1
Control Elevated fiber containing diet
Dietary energy content (kcal/kg)

Digestible energy 3,421 3,327
Metabolizable energy 3,300 3,182
Net energy 2,477 2,295
Feed intake + wastage (kg/d) 2.228 2.395
Body weight gain (g/d) 900 900
Body protein deposition (g/d) 147 147
Standardized ileal digestible requirements
Lysine (% of diet) 0.847 0.789
Lysine (g/d) 17.9 18.0
Lysine (g/Mcal NE) 3.42 3.44
Threonine (% of diet) 0.513 0.507
Threonine (g/d) 10.9 11.6
Threonine (g/Mcal NE) 2.07 2.21
Threonine / lysine × 100 60.6 64.3
1 Based on NRC (2012); the control diet was formulated to contain 72% corn, 25% dehulled, solvent extracted soybean
meal and 3% premix; the elevated fiber containing diet was formulated to contain 32% corn, 30% corn DDGS (6-9%
oil), 20% wheat shorts, 15% dehulled, solvent extracted soybean meal, and 3% premix.
Conclusions and implications
Sufficient knowledge is now available for factorially estimating requirements for key amino
acids of different groups of growing pigs in a relatively stress and disease-free environment
(NRC, 2012). To estimate amino acid requirements at different stages of growth, information is
required about the patterns of feed intake and whole body protein deposition (PD). The latter
can be estimated from energy intake over and above maintenance energy requirements and live
body weight gain. For estimation of threonine requirements, information about dietary levels of
fermentable fiber is also required. When predicting amino acid requirements of groups of pigs,
between-animal variability in growth performance and nutrient utilization should be taken into
account. Nutritional history and compensatory growth should be considered when establishing
optimum dietary amino acid levels for maximum profit. The prediction of the extent and rate of
compensatory growth requires that dynamics of relationships between energy intake and target
body compositions are characterized in the various pig types. Knowledge is lacking for estimating
increases in amino acid requirements of pigs that are exposed to disease causing organisms, and
for estimating nitrogen requirements of growing pigs.

References
Averous, J., A. Bruhat, S. Mordier, and P. Fafournoux. 2003. Recent advances in the understanding of amino
acid regulation of gene expression. J. Nutr. 133:20405-20455.
Baker, D. H. 1986. Critical Review: Problems and pitfalls in animal experiments designed to establish dietary
requirements for essential nutrients. J. Nutr. 116:2339-2349.
Bikker, P., M. W. A. Verstegen, R. G. Campbell, and B. Kemp. 1994. Digestible lysine requirement of gilts with
high genetic potential for lean gain, in relation to the level of energy intake. J. Anim. Sci. 72:1744-1753.
Black, J. L., R. G. Campbell, I. H. Williams, K. J. James, and G. T. Davies. 1986. Simulation of energy and
protein utilization in the pig. Research and Development in Agriculture 3:121-145.
Columbus, D. and C. F. M. De Lange. 2012. Evidence for validity of ileal digestibility coefficients in
monogastrics. Brit J. Nutr. (In press).
De Lange, C. F. M., B. J. Marty, S. H. Birkett, P. Morel, and B. Szkotnicki. 2001a. Application of pig growth
models in commercial pork production. Can. J. Anim. Sci. 81:1-8.
De Lange, C. F. M., A. M. Gillis, and G. J. Simpson. 2001b. Influence of threonine intake on whole-body protein
deposition and threonine utilization in growing pigs fed purified diets. J. Anim. Sci. 79:3087-3095.
De Lange, C. F. M., P. C. H. Morel, and S. H. Birkett. 2003. Modeling chemical and physical body composition
of the growing pig. J. Anim. Sci. 81 (14, Suppl., Electronic Supplement 2):E159-E165.
Dunshea, F.R., 2012. Emerging technologies with the potential to improve feed efficiency in swine. Pages
259-275 in Feed efficiency in swine. J.F. Patience, ed. Wageningen Academic Publishers, Wageningen,
the Netherlands.
Fontaine, J., U. Zimmer, P. J. Moughan, and S. M. Rutherfurd. 2007. Effect of heat damage in an autoclave
on the reactive lysine contents of soy products and corn distillers dried grains with solubles. Use of
the results to check on lysine damage in common qualities of these ingredients. J. Agric. Food Chem.
55:10737-10743.
Jeaurond, E. A., M. Rademacher, J. R. Pluske, C. H. Zhu, and C. F. M. De Lange. 2008. Impact of feeding
fermentable proteins and carbohydrates on growth performance, gut health and gastrointestinal
function of newly-weaned pigs. Can. J. Anim. Sci. 88:271-281.
Le Floc’h, N., L. Lebellego, J. J. Matte, D. Melchior, and B. Sève. 2009. The effect of sanitary status degradation
and dietary tryptophan content on growth rate and tryptophan metabolism in weaning pigs. J. Anim.
Sci. 87:1686-1694.
Libao-Mercado, A. J. O., S. Leeson, S. Langer, B. J. Marty, and C. F. M De Lange. 2006. Efficiency of utilizing
ileal digestible lysine and threonine for whole body protein deposition in growing pigs is reduced when
dietary casein is replaced by wheat shorts. J. Anim. Sci. 84:1362-1374.
Libao-Mercado, A. J. O., C. L. Zhu, J. P. Cant, H. Lapiere, J. N. Thibault, B. Sève, M. F. Fuller, and C. F. M.
De Lange. 2009. Dietary and endogenous amino acids are the main contributors to microbial protein
in the upper gut of normally nourished pigs. J. Nutr. 139:1088-1094.
Martínez-Ramírez, H. R. and C. F. M. De Lange. 2007. Compensatory growth in pigs. Pages 331-353 in
Recent Advances in Animal Nutrition. P.C. Garnsworthy and J. Wiseman, ed. Nottingham University
Press, Nottingham UK.
Martínez-Ramírez, H. R., E. A. Jeaurond, and C. F. M. De Lange. 2008. Dynamics of body protein deposition
and changes in body composition following sudden changes in amino acid intake: II. Entire male pigs.
J. Anim. Sci. 86:2168-2179.

MLC (Meat and Livestock Commission). 2004. Finishing Pigs: Systems Research, Production Trial 2.
Evaluation of phase feeding in two contrasting finishing systems (fully slatted versus straw based
conditions). British Pig Executive, Stoneleigh Park, Kenilworth, Warwickshire, CV8 2TL, UK.
Moehn, S. and C. F. M. De Lange. 1998. The effect of body weight on the upper limit to protein deposition
in a defined population of growing pigs. J. Anim. Sci. 76:124-133.
Moehn, S., A. M. Gillis, P. J. Moughan, and C. F. M. De Lange. 2000. Influence of dietary lysine and energy
intakes on body protein deposition and lysine utilization in the growing pig. J. Anim. Sci. 78:1510-1519.
Moehn, S., R. O. Ball, M. F. Fuller, A. M. Gillis, and C. F. M. De Lange. 2004. Growth potential, but not body
weight or moderate limitation of lysine intake, affects inevitable lysine catabolism in growing pigs. J.
Nutr. 134:2287-2292.
Morel, P. C. H., G. R. Wood, and D. Sirisatien. 2008. Effect of genotype, population size and genotype
variation on optimal diet determination for growing pigs. Acta Horticulturae 802:287-292.
Moughan, P. J., L. H. Jacobson, and P. C. H. Morel. 2006. A genetic upper limit to whole-body protein
deposition in a strain of growing pigs. J. Anim. Sci. 84:3301-3309.
Moughan, P. J., W. C. Smith, and G. Pearson. 1987. Description and validation of a model simulating growth
in the pig (20-90 kg liveweight). New Zealand. J. Agri. Res. 30:481-489.
Moughan, P. J., M. W. A. Verstegen, and M. I. Visser-Reyneveld (eds.), 1995. Modeling Growth in the Pig.
Wageningen Pers, Wageningen, the Netherlands. 238 pp.
Moughan, P. J. 1999. Protein metabolism in the growing pig. Pages 299-332 in A Quantitative biology of the
pig. I. Kyriazakis, ed. CAB International Wallingford, Oxon, UK.
NRC (National Research Council) 1998. Nutrient Requirements of Swine. Tenth Revised Edition. National
Academic Press, Washington, D.C. 20418 USA.
NRC (National Research Council) 2012. Nutrient Requirements of Swine. Eleventh Revised Edition.
National Academic Press, Washington, D.C. 20418 USA.
Nyachoti, C. M., F. O. Omogbenigun, M. Rademacher, and G. Blank. 2006. Performance responses and
indicators of gastrointestinal health in early-weaned pigs fed low-protein amino acid-supplemented
diets. J. Anim. Sci. 84:125-134.
Pomar, C., I. Kyriazakis, G. C. Emmans, and P. W. Knap. 2003. Modeling stochasticity: Dealing with
populations rather than individual pigs. J. Anim Sci. 81:E178-186E.
Rakhshandeh, A. 2011 Impact of immune system stimulation on metabolism and requirements for sulfur
amino acids in growing pigs. Ph.D. Thesis. Department of Animal and Poultry Science, University of
Guelph, Guelph, Ontario, Canada.
Rakhshandeh, A., and C. F. M. De Lange. 2011. Immune system stimulation in the pig: effect on performance
and implications for amino acid nutrition. Pages 31-46 in Manipulating pig production XIII. R.J. Van
Barneveld, ed. Australasian Pig Science Association Inc. Werribee, Victoria, Australia.
Schinckel, A.P., and C.F.M. De Lange. 1996. Characterization of growth parameters needed as inputs for pig
growth models. J. Anim. Sci. 74:2021-2036.
Schinckel, A. P., J. R. Wagner, J. C. Forrest, and M. E. Einstein. 2010. Evaluation of the prediction of
alternative measures of pork carcass composition by three optical probes. J. Anim. Sci. 88:767-794.
Stein, H., M. Fuller, P. J. Moughan, B. Seve, and C. F. M. De Lange. 2007. Amino acid availability and
digestibility in pig feed ingredients: Terminology and application. J. Anim. Sci. 85:172-180.
Van Milgen, J., J. Noblet, A. Valancogne, S. Dubois, and J. Y. Dourmad. 2008. InraPorc: a model and decision
support tool for the nutrition of growing pigs. Anim. Feed Sci. Techn. 143:387-405.

Weis, R. N., S. H. Birkett, P. C. H. Morel, and C. F. M. De Lange. 2004. Independent effects of energy intake
and body weight on physical and chemical body composition in growing entire male pigs. J. Anim. Sci.
82:109-121.
Zhu, C. L., M. Rademacher, and C. F. M. De Lange. 2005. Increasing dietary pectin level reduces utilization
of digestible threonine intake, but not lysine intake, for body protein deposition in growing pigs. J.
Anim. Sci. 83:1044-1053.

5. The influence of dietary energy on feed efficiency in grow-finish
swine
J.F. Patience
Department of Animal Science, Iowa State University, 201 Kildee Hall, Ames, IA 50011-3150,
USA; jfp@iastate.edu
Abstract
It is widely accepted within the field of nutrition that there is a very close relationship between
the concentration of energy in the diet and the feed efficiency achieved by a group of pigs. All
other things being equal, increasing dietary energy concentration will normally improve feed
efficiency. However, in commercial practice, all other things are not equal, so the statistical
correlation between dietary energy and feed efficiency is in reality very small. This is because
so many factors influence feed efficiency outcomes. Some are related to the pig and some are
related to the physical, social and health environment in which the pig lives. And some are
related to dietary energy and how the pig uses this energy for maintenance and for growth.
It is only through a thorough understanding of energy metabolism that we can begin to most
effectively manage feed efficiency on the farm by manipulating dietary energy concentration and
dietary energy supply. This chapter will provide an overview of energy metabolism in the pig:
(1) defining exactly what energy is; (2) describing the dietary sources of energy; (3) explaining
energy metabolism and discussing why it is so much more complicated than the metabolism of
nutrients in the diet; (4) describing the various energy systems; and (5) explaining how energy
is ultimately used by the pig for maintenance and for growth. The chapter will also discuss
the importance of feed intake, and how this will become even more critical for success in the
future when diets are likely to contain less energy than they do today. The chapter will end
by taking the information on energy metabolism presented in this chapter and applying it in
a practical circumstance to assemble very specific recommendations on how to improve feed
efficiency in the pig.
Introduction
It is a widely accepted within the field of nutrition that there is a very close relationship between
the concentration of energy in the diet and the feed efficiency achieved by a group of pigs. This
view is supported by an exhaustive list of papers reporting the results of research that achieved
predictable improvements in feed efficiency when dietary energy was increased (Dove, 1993;
Apple et al., 2004; Beaulieu et al., 2009; Collins et al., 2009; Benz et al., 2011; Hinson et al.,
2011). The results of these and other studies draw the same conclusion, that as dietary energy
concentration increases, feed efficiency is improved. There is no empirical data to refute this
relationship. Indeed, some nutritionists have suggested that a 1% increase in diet metabolizable
energy (ME) will increase feed efficiency by about 1.5%. However, a review of the literature
suggests a very wide range in this ratio, from 0.7:1 to 2.4:1. A ratio of 1:1.5 is an arithmetic mean
of the results of many experiments, but it is not a very predictable average because there is a
wide range in outcomes. This is perhaps the first hint that the relationship between diet energy

J.F. Patience
concentration and feed efficiency is not as close as people might at first glance assume. In other
words, it is one thing to increase dietary energy in a carefully controlled experiment and observe
an increase in feed efficiency, but it is quite another to define a consistent relationship between
the two variables under less orderly conditions. The reality is that many, many factors affect feed
efficiency, and diet energy concentration is only one of them.
The challenge of understanding the relationship between dietary energy and feed efficiency is
further demonstrated by looking at a different set of data in the literature, where diets of diverse
energy density and composition were fed to pigs. In these cases, the relationship between dietary
energy concentration, or even the more definitive daily energy intake, and feed conversion
is surprisingly small. For example, Oresanya et al. (2008) reported no statistically significant
correlation between daily digestible energy (DE) intake and feed efficiency.
If we seek to fully understand the variables related to the pig and its surroundings, as well as those
related to the composition of the diet, we can begin to better grasp these seemingly divergent
outcomes and once understood, will allow us to manage feed efficiency in order to achieve more
predictable productivity and financial outcomes in the future. In order to fully comprehend the
relationship between dietary energy and feed efficiency, there is a need to delve deeply into the
science underlying energy – what it is, how it is measured, for what purposes the pig uses energy
and what steps in metabolism are open to improvements in efficiency. In other words, what can
we control and how can we control it in order to improve the efficiency with which the pig uses
dietary energy. When we are able to do this, we are in a much stronger position to achieve the
lowest possible feed cost while maximizing producer net income.
Feed efficiency can become a false god sought by pork producers to improve profitability. As
the cost of feed – and specifically of feed energy – rises, we can appreciate that maximizing feed
efficiency may not maximize net income. Other chapters in this book will address management
practices, as well as other aspects of the diet – amino acid nutrition, processing, etc. – that can
be implemented to improve feed efficiency. This chapter will focus on the relationship between
dietary energy and feed efficiency, considering the supply of dietary energy and the metabolic
fate of that energy. The chapter ends with specific practical recommendations for improving feed
efficiency as it relates to dietary energy and energy metabolism.
Defining and expressing feed efficiency
Recently, there has been discussion on how to best calculate and represent feed efficiency.
The traditional calculations of gain divided by feed, the primary expression used in scientific
publications and which I will refer to as feed efficiency, or feed divided by gain, the primary
expression used in commercial production and which I will refer to as feed conversion, are
increasingly viewed as too simplistic. Some of the broader challenges with this calculation are
well defined in other chapters in this book, but suffice it to say that using feed conversion as a
benchmark is fraught with problems when initial or final weight differ or if dressing percentage
varies. It is also complicated when comparing diets with differing particle sizes or comparing
mash versus pellets. These rather simple variables can lead to disturbingly large errors. I think it
is fair to say that feed efficiency is something that must be tracked and monitored, but it should

5. The influence of dietary energy on feed efficiency in grow-finish swine
only be used as a management driver if one fully understands exactly what the feed efficiency
values mean.
In reality, efficiency of feed utilization may be better represented by alternative calculations.

Since we are discussing dietary energy, the term ‘Mcal of energy per unit of gain’ is gaining
greater use in commercial practice because it focuses specifically on dietary energy rather than
on the total quantity of feed (Oresanya, 2005). This approach possesses intuitive appeal, and
makes particular sense when one considers that energy is by far the most expensive component
of the diet (Table 1). However, this approach may not provide increased accuracy given that the
actual DE, ME or net energy (NE) content of the diet may differ radically from the estimated or
formulated content (Figure 1). Variation in the energy content of ingredients as well as differences
in digestibility of ingredients due to processing, such as particle size and pelleting, or due to the
age of the pig, are just a few of the factors that can alter the actual digestibility of the diet and
lead to problems in accurately predicting energy values of diets (Wondra et al., 1995; Le Goff and
Noblet, 2001). It would improve the value of the measurement of energy efficiency if one only
compares diets of a similar ingredient composition, or of the same particle size, or only pelleted
diets, or among pigs of the same age. In reality, expressions of efficiency cross all forms of diets,
many ages of pigs and increasingly diverse ingredient composition. The greater the diversity that
exists within the comparison, the greater will be the risk of error. Nonetheless, the approach has
merit, because it focuses in energy per se and because its expanded use will demand that we
improve our capabilities of estimating the energy content of ingredients and of feeds in the future.
Table 1. The relative cost of meeting the energy, energy plus amino acids or all nutrient specifications in a
typical growing pig diet – November, 2011.
Ingredient (%) Ingredient price Energy specifications Energy plus Energy, amino acids,
($/ton) only amino acids minerals and vitamins
Corn 220 54.93 47.65 47.01

Corn DDGS 190 30.00 25.58 27.68
Wheat midds 200 7.60 5.50 -
Soybean meal 300 - 13.50 14.19
Bakery product 230 7.50 7.50 7.50
L-lysine HCl 2,500 - 0.30 0.30
Limestone 50 - - 1.10
Salt 90 - - 0.45
Vitamin premix 1,750 - - 0.15
Trace mineral premix 1000 - - 0.12
Phytase 5,000 - - 0.08
AV-blend 900 - - 1.16
Cost, $/ton 210.24 229.58 244.00
% of total diet cost 86.2% 94.1% 100%

J.F. Patience
3,800
Measured 3,600
3,400
3,200
3,000
2,800
2,800 3,000 3,200 3,400 3,600 3,800
Formulated
B
3,600
3,550
3,500
Measured
3,450
3,400
3,350
3,300
3,250
3,200
3,250 3,300 3,350 3,400 3,450 3,500 3,550 3,600
Formulated
Figure 1. Relationship between formulated and measured digestible energy contents of weanling (A) and
grower (B) diets.
If the trend is to express feed conversion on the basis of energy, perhaps further refinement of
this calculation would be warranted, and subtract the energy utilized for maintenance needs
from the total supply of energy in order to focus specifically on the energy available for growth
(Oresanya, 2005; Oresanya et al., 2008). Of course, this places pressure on our ability to estimate
maintenance energy requirements; even a cursory review of the literature will reveal the difficulty
we have in this respect. However, since about one-third of the total daily supply of energy is
devoted to maintenance (Table 2), it may be advantageous to focus specifically on the energy
available for growth and the efficiency with which it is used for lean and fat gain. If a fixed value
for maintenance energy is subtracted from daily energy intake, using equations presented later in
this chapter, then perhaps the resulting ‘efficiency’ value might best be described as the efficiency
of energy used for growth and extraordinary maintenance needs. It serves the purpose of focusing
on the efficiency with which energy is utilized by the pig for growth, but one must recognize that

Table 2. How daily energy intake is partitioned among maintenance, protein gain and lipid gain in a 70 kg pig.1
Gain Partitioning of ME intake
Mcal/d % of total
Maintenance - 2.52 34
Protein gain 138 g/d (16% of total gain) 1.38 19
Fat gain 294 g/d (34% of total gain) 3.45 47
Total 860 g/d (100% of total gain) 7.35 100
1 This example represents the energy utilization of a 70 kg pig growing 860 g/d and eating 2.2 kg/d of a diet
containing 3.34 Mcal ME/kg. It is assumed that the pig is housed in thermoneutral conditions. ME intake assigned
to maintenance is determined by the equation MEm = 197 × BW0.60 (NRC, 2012). It is further assumed that 16% of
the total gain is represented by protein gain; this results in a protein deposition rate of 138 g/d. The energy required
to deposit protein is determined by assuming that it takes 10.0 kcal ME per g protein gain (NRC. Accounting for
maintenance and protein gain, the remaining ME consumed can be directed to lipid gain which requires 11.7 kcal
ME per g lipid gain.
some of this energy may be apportioned to extraordinary maintenance needs, such as thermal
regulation, behavioral stressors or immunological needs (Knap, 2009).
It must also be recognized that expressing feed conversion using energy in the numerator does not
diminish the errors inherent in the calculation due to differences in initial and final body weight,
which are reflected in the denominator. Nor does it provide any improvement in accuracy when
diets high or low in fiber are compared, resulting in different dressing percentages (Linneen et al.,
2008; Beaulieu et al., 2009; Patience, 2011). Data reported by Beaulieu et al. (2009) is recalculated
to demonstrate what feed conversion might look like when expressed on the basis of total energy
intake or energy intake above basal maintenance (Table 3). This particular study was used for
these calculations because the actual digestible energy content of the diet was determined during
the experiment and reported, so book values were not used.
In reality, the best expression of feed conversion will be determined by the information required
by the pork producer and the purpose for which the information is being collected. There is no
perfect calculation. Provided that the person interpreting the data has a clear understanding of
the limitations of any value, there is less danger of misinterpreting them and making incorrect
management decisions as a result.
Definition of dietary energy
Nutrients are chemical substances that are present in food that are necessary for maintenance
of the body and for growth, lactation or gestation. Thus, nutrients include proteins (amino
acids), fats, carbohydrates, vitamins, minerals and water (Drummond, 1996; Stipanuk, 2006).
Energy is not considered to be a nutrient, because it is not a chemical substance, but rather a

J.F. Patience
Table 3. The influence of increasing dietary energy (DE, Mcal/kg) on pig performance and various estimates
of efficiency (Beaulieu et al., 2009).
Determined diet DE (Mcal/kg) 3.09 3.24 3.34 3.42 3.57 SEM5
Initial weight (kg) 31.2 31.1 31.5 31.2 31.1 0.24

Final weight (kg) 115.1 115.5 115.3 115.0 115.6 0.41
Average daily gain (kg/d)1 1.00 1.02 1.04 1.02 1.03 0.01
Days on test 84.6 83.6 84.2 82.5 81.9 0.04
Average daily feed intake (kg/d)2 2.66 2.62 2.62 2.52 2.44 0.09
Feed efficiency, gain:feed1 0.39 0.40 0.41 0.42 0.44 0.004
Carcass backfat (mm)1 16.8 17.8 18.3 18.6 19.4 0.5
Energy intake (Mcal DE/d)2 8.22 8.49 8.76 8.61 8.71 0.10
Calculation from the above data
Energy efficiency (Mcal DE/kg gain)3 8.22 8.32 8.42 8.44 8.49
Energy efficiency (Mcal DE/kg gain)4 5.57 5.73 6.08 5.82 5.82
1 Linear effect of dietary energy significant, P<0.05.
2 Linear effect of dietary energy significant, P<0.10.
3 Includes total energy consumed divided by total gain.
4 Includes total energy less maintenance energy divided by total gain.
5 SEM = standard error of the means.
characteristic of the diet. Indeed, it is a very complex constituent of the diet because it is supplied
in the diet by many different sources, and the metabolic pathways that transform each source
into useable energy are equally diverse. When a nutritionist studies energy, we need to consider
that energy comes from four different dietary sources – simple carbohydrates (primarily starch),
complex carbohydrates (fiber), fat and protein (Figure 2) – that each source will have different
bioavailabilities and each source will be used with a different level of metabolic efficiency
depending on whether it contributes to maintenance energy or retained energy in tissue (Tables
4 and 5). This is a much more complex circumstance than is the case with nutrients such as amino
acids or minerals. This complexity probably explains, at least in part, why energy utilization by
the pig remains something of a mystery. Even when the basic concepts of energy metabolism
are understood, applying this information under farm conditions can be a truly daunting task;
variation in ad libitum feed intake, the presence of social and environmental stressors, and disease
load all influence the response to dietary energy content.
Antoine Lavoisier (1743-1794) was the first to identify the concept of combustion, which he
defined as the ‘uniting of a substance with oxygen’. Previous to this, burning was viewed as a
mechanical, rather than chemical, process. He is also attributed with naming ‘oxygen’ for the first
time. He applied his concept of combustion to the body and the ‘burning’ of food. He went on
to define a fixed relationship between the quantity of heat produced and the amount of carbon
dioxide formed, irrespective of whether it was released by burning or by metabolism in the animal
– and thus founded the modern science of calorimetry. Indeed, he was also the first to note that

Starch Protein Fat Fiber
Energy
Maintenance Growth
Lean Fat
Figure 2. Energy is supplied by the diet from 4 distinct and unique sources: starch, protein, fat and fiber. In this
way, energy differs substantially from nutrients in the diet.
the quantity of energy released from a given amount of a substance was the same, whether it was
‘burned’ inside the body or outside (Thorbek, 2000). His design of an early ice calorimeter was
able to achieve an impressive degree of accuracy given the state of instrumentation at that time.
As a result of the enormous new concepts he established, the new knowledge that he generated
and his ability to expand the understanding of existing knowledge at that time has resulted in
some calling him the father of modern chemistry. While Priestly, Black, Hales, Cavendish, Scheele
and others might protest, Lavoisier’s achievements were truly impressive (Ihde, 1964). Many
have also called him the father of nutrition (Bender and Bender, 1997). Alas, his extraordinary
scientific accomplishments failed to protect him from the events of the French Revolution and
he was guillotined at the age of 51. It was a distressing end for a man who played such a central
role in the transformation of chemistry from its very primitive state in the mid-18th century to
the modern science as we know it today (Bensaude-Vincent, 1996; Holmes, 1985). According to
a story – difficult to confirm – relating Lavoisier’s final days, the judge who rejected an appeal of
his death sentence is reported to have stated ‘The Republic needs neither scientists nor chemists;
the course of justice cannot be delayed’.
In the more than 200 years after Lavoisier’s discoveries, much has been learned about the physics,
chemistry and biochemistry of oxidation. In nutritional terms, we now know that the caloric
content of ingredients and mixed feeds leads to predictable quantities of energy being either
retained in the body, or utilized by the body, as an essential part of life. The calories ‘burned’
by the body results in either work being performed or heat being generated, or in most cases, a
combination of the two. So, when a pig consumes, for example, 7 Mcal of ME per day, a portion
of that energy will be retained in the body as new tissue, a portion will be used to perform work
(activity) and a portion will be released as heat, the consequence of various chemical reactions
associated with maintenance. More on this later.
The calorie has become the basic unit of measure of energy, where 1 calorie is defined as the
quantity of heat required to raise the temperature of 1 gram of water from 25 °C to 26 °C at
one atmosphere of pressure (Russo and Silver, 2011). In the Système International d’Unités (SI)

J.F. Patience
system, the basic unit of energy is the joule; a joule is equal to 0.239 calories, or 1 calorie equals
4.184 joules.
Two methods are typically employed to measure energy balance in pigs. The first is an indirect
calorimetry method, whereby animals are placed in respiratory chambers and the quantity of O2
consumed and CO2 produced can be quantified, along with nitrogen excretion and other gaseous
emissions, notably methane (Brouwer, 1965):
Heat production, kcal = (3.866 × oxygen consumption, l) + (1.200 × carbon dioxide production, l) –
– (0.518 × methane production, l) – (1.431 × urinary nitrogen production, g)
The use of respiration chambers has become the method of choice for many laboratories because
measurements can be more over shorter periods of time, and multiple times on the same pigs
(Van Milgen and Noblet, 2003). The level of random error in this method is therefore typically
lower than that achieved by the comparative slaughter method, described below. Heat production
determined using indirect calorimetry is typically lower than observed with the serial slaughter
method, probably due to lower pig activity (Noblet et al., 1987). It is also possible that indirect
calorimetry underestimates heat production because the energy requirement for protein
deposition is confounded with the energy used for maintenance (Van Es, 1980; Just, 1982b).
The other method is called the comparative or serial slaughter technique, whereby a group of
animals are slaughtered at the end of the test period and the carcasses are ground and analyzed
for nutrient and energy content. Other animals from the same population are slaughtered at the
beginning of the test period and are called the initial slaughter group. They are also minced and
analyzed. The difference between the nutrients and energy in the carcass of the final group of
pigs compared with the initial slaughter group allows for determination of the rate of nutrient
and energy retention in the body over the course of the experiment (Adeola, 2001; Oresanya
et al., 2008). The comparative slaughter method has the advantage of being able to undertake
studies using animals in circumstances closer to commercial conditions than can be achieved
in respiration chambers, e.g. higher feed intake and normal eating activity. However, the key
weakness of this method is the error introduced by the initial slaughter group, which may – or may
not – represent the body composition of the test pigs at the commencement of the experiment.
Dietary sources of energy
As explained above, energy is supplied in the diet by carbohydrates (simple and complex), by
protein and by fat. While the pig can utilize starch with great efficiency, and degrade starch
granules into constituent glucose molecules for absorption into the body, it is not well bestowed
with enzymes to break down the more complex carbohydrates, or fiber as it is more commonly
known. There is not a lot of research on the subject, but the contribution of fiber to the energy
supply of the growing pig is much smaller than that from other energy sources. Indeed, Noblet
and Le Goff (2001) concluded that when one takes into account the fact that fiber in the diet
reduces the digestibility of other nutrients, and also increases endogenous secretions, it provides
little or no energy to the growing pig. However, there will be a wide range in the quantity of
energy supplied to the growing pig by fiber, because there is a very large range in the digestibility

of fiber (Graham et al., 1986). Acid detergent fiber (ADF) has a very low digestibility (38% in
growing pigs and 49% in sows) and neutral detergent fiber (NDF) has a digestibility of about 60%
in growing pigs and 72% in sows. A great deal more research is required on the subject to be able
to understand the contribution of fiber to the total energy supply in the pig.
The utilization of simple carbohydrates, mainly starch, differs greatly from fiber. Dietary starch,
or at least a large portion of it, and other sugars are digested in the small intestine, releasing
glucose which can then be absorbed by the enterocytes and released into the portal vein for
distribution to the liver and other parts of the body. Some starch, known as ‘resistant starch’
cannot be utilized in the small intestine, and effectively behaves like fiber since it is fermented
with other carbohydrates in the large intestine and cecum.
The utilization of protein is different yet again from carbohydrates, since only amino acids
supplied by the body in excess of that needed for protein synthesis will be directed to catabolic
pathways, leading to removal of the amine side chain for elimination as urea or free ammonia in
the urine. Since ammonia is toxic, the majority of nitrogen will be directed to the urea cycle so
it can be removed from the body in a less toxic form. Patience and Chaplin (1997) reported that
only about 15% of nitrogen is eliminated in the urine in the form of ammonia; due to the normal
pH of the urine, it will actually exist as an ammonium salt. Thus, there is a substantial cost to the
body to use excess amino acids as an energy source, because a large majority of the amine group
must be incorporated into urea, a process that is energetically expensive. The residual carbon
chain of the amino acid is then available for catabolic processes via the TCA cycle to generate ATP
in amounts that will depend on the amino acid in question. Alternatively, the carbon chain can
be utilized for the purposes of gluconeogenesis or lipogenesis; thus, depending on the metabolic
status of the pig, the carbon molecules present in amino acids can be used to produce ATP for
immediate use, or to contribute to the available energy pool for later use if needed by the pig.
Because of the intermediary steps required to direct excess amino acids to the energy pool, they
are a less efficient source of energy than glucose or free fatty acids, irrespective of whether they
are used for gluconeogenesis or lipogenesis (Table 4).
Fat and fatty acids present yet another unique metabolic option to the pig. Fatty acids in the form
of triglycerides can be directed with great efficiency to body fat tissue or can be catabolized to
Table 4. Marginal energetic efficiency of using digestible nutrients for generating ATP or depositing lipid
(Black, 1995, as adapted by Birkett and de Lange, 2001a).
Source Generating ATP Depositing lipid
Efficiency % relative to fatty acids Efficiency % relative to fatty acids
Fatty acids 66 100 90 100

Glucose 68 103 74 82
Amino acids 58 88 53 59
Crude fat 50 76 62 69

J.F. Patience
produce ATP with about the same efficiency as glucose (Table 4). If the pig has an excess supply
of energy, or is genetically motivated to synthesize body fat, it is very much to its advantage to
do so from fatty acids in the diet – representing an efficiency of about 90% – rather than adopt
de novo fat synthesis which has an efficiency of 50 to 75%, depending on the substrate (Table 4).
Table 5 summarizes the basics behind the Dutch CVB system, which starts with the total heat
present in a dietary constituent (enthalpy), adjusts for digestibility and heat increment of those
constituents and ultimately estimates the partial efficiency with which each will be used for energy
retained in the body (Rijnen et al., 2004). The data in Tables 4 and 5 are in general agreement with
each other, although they are reporting slightly different approaches.
It is clear from this very brief discussion that energy metabolism in the pig is a complex subject.
It is not surprising that discussions on the subject among nutritionists often result in animated
debate!
Energy systems
Energy systems fill two critical roles in diet formulation and the development of feeding programs.
First, they must provide a method for assigning accurate relative nutritional and economic values
to all of the feed ingredients that a nutritionist might want to use. In other words, it forms an
essential foundation for trade and commerce in ingredients and mixed feeds. The second role of
energy systems is to support the formulation of diets that result in predictable performance by
the pigs. To satisfy these roles, the value of the ingredients will be dependent on the availability
or digestibility of the energy in each ingredient, and also on the efficiency with which the energy
in the ingredient can be used by the pig for maintenance and productive functions. The system
also requires that energy values will be additive with other ingredients in the diet when used in
linear programming feed formulation procedures.
The development of energy systems has provoked great controversy over the years. The Weende
system, based on crude fiber, crude fat, crude protein and nitrogen free extract pre-dates the 20th
century. Nonetheless, it was the forerunner of modern energy systems. The Germans initiated
research on net energy at Rostock in the 60’s and 70’s (Schiemann et al., 1972), with further
Table 5. Values for enthalpy, net energy, heat increment and partial efficiencies for digestible nutrients (Rijnen
et al., 2004).
Nutrient Enthalpy Net energy Heat increment Partial efficiency

(Mcal/kg) (Mcal/kg) (Mcal/kg) (%)
Crude protein 5.64 2.58 3.06 46

Crude fat 9.46 8.63 0.83 91
Sugars 3.79 2.96 0.83 78
Starch 4.21 3.27 0.84 78
Dietary fiber 4.21 2.29 1.92 54

developments evolving in Denmark (Just, 1982a), the Netherlands (CVB, 1993) and France
(Noblet and Henry, 1993). It is not surprising that the Europeans have shown great leadership in
the development of energy systems, because they have historically experienced the most costly
diets of any major pork producing region. They were uniquely motivated to study energy most
intensively and develop energy systems that met their needs and achieved their primary goal
of feeding pigs at the lowest possible cost. Research on energy is very expensive, and the strong
financial support that European nations provided to national and regional research institutions
was absolutely critical to success. One must ask the question where would our understanding of
energy be today were it not for the financial support of central governments in such countries as
Germany, Denmark, France and the Netherlands?
On the other hand, North America possessed bountiful supplies of relatively inexpensive
ingredients, and the diets used in most parts of North America – western Canada most notably
excepted – were very simply in composition so there was less motivation to invest in research on
energy systems. Furthermore, the respective pork industries on both sides of the Atlantic evolved
in different directions. In Europe, minimizing feed cost and achieving a very, very lean carcass
placed the greatest emphasis on cost of production and less on growth performance, while in
North America, producers placed a much higher premium on growth rate and barn throughput.
North American producers were feeding pigs to increasingly heavier market weights and felt
pressure to achieve carcass size targets in the shortest time possible. Complex and high cost
diets in Europe fueled interest there in net energy systems, originating from outstanding early
research in Germany and Denmark and refined more recently in the Netherlands and France.
In North America, simple diets and rapid growth provided little interest in evolving beyond
the energy systems already in place, namely DE and ME. However, it is interesting to note that
ruminant nutritionists in North America, unlike their monogastric counterparts, appeared to be
much more motivated to study sophisticated energy systems and contributed global leadership
in ruminant net energy systems (Moe, 1981; Fox et al., 1992).
With rapidly increasing feed costs, and increasing dependence on export markets, North America
is re-visiting the concept of more precise energy systems and net energy is certainly on the radar
of many pork producers and feed companies. Indeed, some have already made the switch.
As previously mentioned, dietary energy is a complex subject, and quantifying it in terms of

utilization by the pig is no small task. Different energy systems have been developed and adopted
around the world. It is fair to say that no single system has captured the full attention of all
nutritionists. The various systems that are currently in the most prominent use around the
world are DE, ME and NE. The Danes have developed their own unique feed evaluation system
called potential physiological energy, or PPE (Boisen, 2007). North and South America have
traditionally used the DE and ME systems, although the NE system is gaining increasing support
as diet composition becomes more diverse and feed costs rise. Europe is largely split between
DE and NE, although there is no question that NE has its greatest support in Europe, especially
in countries like France, the Netherlands, Spain and the UK. Australia uses the DE system and
China uses the ME system. And of course, the PPE system is in use in Denmark.

J.F. Patience
Gross energy (GE) measures the total quantity of energy contained in a feed or an ingredient.
It does so through combustion within an enriched oxygen environment, resulting in the release
of heat that is reported in calories or joules. The total sample will be burned to completion, and
the only residual left behind will be inorganic materials, namely minerals. Gross energy has very
little value in animal nutrition, other than as the basis for determining other, more useful values
(Figure 3). Digestible energy corrects gross energy for that portion of energy lost in the feces.
Digestible energy can more accurately be referred to as apparent total tract digestibility (ATTD)
of GE, because it corrects for energy that is not absorbed by the pig as the feed passes along the
gastrointestinal tract, but does not correct for endogenous contributions to fecal energy. Indeed,
the energy excreted in the feces will consist of indigestible diet components, metabolic products
released into the GI tract and microbial materials. DE typically represents about 85% of GE in
highly digestible diets based on such ingredients as wheat, corn and soybean meal (Figure 3).
Producers can control the quantity of energy lost in the feces by feed processing technologies,
mainly reducing particle size and the adoption of pelleting or expander technology. DE will
also be higher in adult animals as compared to growing pigs or weanling pigs, prompting some
nutritionists to adopt different energy values for adult pigs as compared to growing pigs. Even
within the growing period, the digestibility of energy increases as the pig grows (Table 6).
Energy utilization in the weanling pig
100% GE Energy in feces Ingredient factors
85% DE Energy in urine
Energy in gases: CH4 + H2

82% ME
Heat increment
56% NE
Km
27% NE NEg 29%
m
Animal factors
Kl Kp
NEl NEp
Figure 3. Traditional view of energy systems, showing the portion of gross energy represented within each,
using a highly digestible diet typical of that used in North America in the late 20th century (adapted from
Ewan, 2001; Oresanya, 2005).

Table 6. Impact of body weight on the digestibility of energy in growing pigs (Noblet, 2005).
Body weight (kg) Coefficient of energy digestibility (%)
38 82.6
49 83.0
61 83.6
72 84.2
80 84.8
90 85.3
Mean 83.6
Metabolizable energy (ME) represents a modest improvement over DE, because it adjusts, in
theory, for the loss of energy via the urine and gases. I say ‘in theory’ because experimentally,
gaseous emissions are rarely measured; they represent only about 0.5 to 1.0% of GE (Just,
1982a). Furthermore, a significant number of ME values in the literature were actually derived
arithmetically from DE using relationships such as those reported by Le Goff and Noblet (2001)
or Noblet and Perez (1993) and were not measured directly. Metabolizable energy represents a
loss of about 3-4%, relative to DE, so ME averages about 96% of DE and 82% of GE (Figure 3).
Sometimes, ‘Modified ME’ is employed. Modified ME recognizes that high protein ingredients,
such as soybean meal, provide less usable energy to the pig due to its high heat increment (see
below), so its energy value is lowered proportionately. Gradually, Modified ME is being replaced
by NE, which adjusts for differences in heat increment among all ingredients in a more structured
manner.
Net energy (NE) is determined by adjusting ME for heat increment, which is the metabolic cost
of converting ME into forms of energy that can be used by the pig. Heat increment includes the
energy lost through digestive processes as well as nutrient metabolism. Table 5 presents one
estimate of the differing efficiency with which dietary components – fat, starch, protein and fiber
– can be converted from ME to NE. The logic of NE is that it quantifies more accurately than
DE or ME the true amount of usable energy in ingredients and pig diets. By accounting for heat
increment, NE presents the quantity of energy that can be used by the pig for maintenance and
productive purposes. The logic of the NE system is acknowledged when one recognizes that an
energy system should provide accurate relative values of usable energy from ingredients.
Net energy can be calculated from DE and from ME according to the following equations (Noblet
et al., 1994):
NE (kcal/kg DM) = 0.70 × DE + 1.61 × EE + 0.48 × ST – 0.91 × CP – 0.87 × ADF
NE (Kcal/kg DM) = 0.73 × ME + 1.33 × EE + 0.39 × ST – 0.62 × CP – 0.83 × ADF

J.F. Patience
Where ME is expressed in Kcal/kg DM, EE = ether extract, ST = starch, CP = crude protein and ADF = acid detergent
fiber, all expressed as g/kg DM. Note that ether extract (fat) and starch elevate NE relative to DE or ME, while crude
protein and fiber lower it.
Net energy cannot be calculated from ME using a fixed constant, as some have suggested. Because
ingredients differ in their respective contents of protein, starch, fiber and fat, it is logical that the
relationship between ME and NE will differ. Van Milgen et al. (2000) illustrated this very well,
showing that the NE:ME ratio varied from 0.744 to 0.795 among 5 different diets, with a mean
of 0.768.
The significance of the impact of NE on relative pricing is shown in Table 7. According to the DE
system, the 6 ingredients would rank in the following order: choice white grease, soybean meal,
corn DDGS, corn, wheat, wheat middlings and barley. Based on the ME system, the ranking
changes modestly, with some of the higher protein ingredients being discounted slightly: choice
white grease, corn, corn DDGS, soybean meal, wheat, wheat middlings and barley. Applying the
NE system changes the rankings substantially: choice white grease, corn, wheat, corn DDGS,
barley, wheat middlings and soybean meal. Obviously, ingredients that contain a notable amount
of fat will be elevated and ingredients higher in protein and/or fiber will be discounted. As an
example, under the ME system, soybean meal is worth 97% of corn – as a source of energy – but
under the NE system, it is worth only 75% that of corn. On the other hand, choice white grease
rises from 235% of corn to 271%.
Intuitively, the NE system will offer little advantage in simple diets with only 2 or 3 basal
ingredients, but begins to show its value when multiple ingredients are used and when they vary
Table 7. The relative energy content of example feed ingredients using the DE, ME and NE systems (National
Swine Nutrition Guide, 2010).
Ingredient DE ME NE
Mcal/kg Relative Mcal/kg Relative Mcal/kg Relative

to corn to corn to corn
Corn 3.53 100 3.43 100 2.65 100

Wheat1 3.36 95 3.22 94 2.46 93
Corn DDGS3 3.63 103 3.42 100 2.37 89
Barley 3.05 86 2.91 85 2.28 86
Wheat middlings 3.07 87 3.03 88 2.19 83
Soybean meal2 3.64 103 3.34 97 1.97 75
Choice white grease 8.30 235 7.97 232 7.17 271
1 Hard red spring.
2 46.5% crude protein.
3 10% crude fat.

widely in fat, protein and fiber content. Of course, the savings that might accrue from the use
of the NE system will depend on the relative pricing of the various ingredients available to the
pork producer.
The adoption of the NE system in North America has been stalled or delayed by a history of
low cost feed ingredients. More recently, with rapidly increasing energy costs, there is definitely
more interest in NE, but the absence of research using local ingredients conducted under local
production conditions leads to uncertainty. With some nutritionists responsible for feed budgets
in the 10s or even 100s of millions of dollars, uncertainty is not something that can easily be
tolerated – unless there is a substantial cost advantage awaiting those that accept the risk!
Furthermore, unlike schools in Europe where the NE system would be an important part of the
curriculum, NE receives little attention in most North American undergraduate and graduate
programs, so the lack of familiarity with the concepts and practice of NE is another impediment.
Without doubt, there is increasing research interest in NE, and if, as expected, it proves to
deliver in practice what it offers in theory, the pace of adoption in North America will speed up
considerably. Certainly, there are data that at least indirectly suggest that growth rate should not
suffer from the adoption of NE, and feed efficiency may be slightly more predictable than under
the ME system (Patience et al., 2004, 2006; Rijnen et al., 2004).
Perhaps one of the greatest challenges in effectively adopting the NE system is related to the very
precision that it seeks to achieve. As one moves from GE to DE to ME to NE, the ingredient
influences on final dietary responses are reduced (Figure 3). In other words, moving along this
continuum addresses or manages the variation among ingredients that arises from differences
in ingredient digestibility (DE) and metabolizability (ME) and heat increment (NE). As these
variables are addressed, the variables associated with the pig become more prominent. We have
previously discussed the fact that ME and NE are used with differing efficiencies, depending on
whether they are used for maintenance, protein accretion or lipid accretion. Consequently, in
order to maximize the utility of the NE system, we must understand – and be able to quantify
– the fate of energy intake in terms of these functions. In other words, to take advantage of this
enhanced precision, we need to understand the portion of dietary energy that will be used for
maintenance, protein gain and lipid gain. Furthermore, since different components of the diet –
fat, starch, fiber and protein – are used for these functions with differing efficiencies, it becomes
quite clear that by addressing one broad source of variation – that being ingredient characteristics
– we place pressure on ourselves to deal with the other source of variability – the pig! Of course,
we should know this when using the DE and ME systems as well, but in those systems, there is so
much ‘noise’ due to ingredient issues that being able to compartmentalize energy use in the pig
is less crucial. I am left with the current conclusion, then, that adopting the NE system should
advance the financial efficiency of the pork producer, because ingredients are priced appropriately
relative to one another, due to differences in heat increment, but that gains in performance will
be much more modest and more difficult to come by.
Given the challenges any energy system faces in dealing with variability deriving from both
the pig and the diet, it seems logical to anticipate the evolution of arithmetic models to help
quantify the multitude of variables inherent in the pig’s use of dietary energy: differing degrees
of digestibility, differing proportions of energy derived from protein, fat, fiber and starch and

J.F. Patience
associated differences in utilization by the pig for differing metabolic and growth functions and
the variation that exists among pigs in terms of the portion of energy that will be directed to
maintenance, protein gain and lipid gain (De Greef and Verstegen, 1995; Danfer, 2000; Birkett
and de Lange, 2001a,b,c; Van Milgen and Noblet, 2003). This seems to be the only logical next step
in the evolution of energy ‘systems’. I say this recognizing that models are extremely expensive to
create, require constant updating – also expensive – and require a high level of knowledge and
understanding to effectively operate under practical conditions. I therefore acknowledge that an
energy model that can be used on the farm is likely still some distance away. The fact that growth
models are already in use in the pig industry gives me cause for optimism.
The Potential Physiological Energy (PPE) system was introduced in Denmark in 2002. It is a
centralized national system widely adopted by pork producers and feed manufacturers in that
country. It differs from the previously described energy systems, which have been derived
from research on pigs, because it has evolved from in vitro analysis of ingredients and mixed
feeds. Specifically, it is based on the quantity of ATP that would theoretically be released by
the oxidation of the feed components at the cellular level. Starch is employed as the baseline,
and other constituents of the diet are valued relative to their replacement of starch in the diet.
The Danes argue, with some validity, that energy systems based on animal experimentation are
constrained by the conditions under which the research was conducted, and how relevant these
conditions might be to actual commercial practice. Interestingly, the relative energy values from
the PPE system and actual NE values are not that different (Table 8).
Dietary energy used for maintenance and for gain
When considered in totality, energy intake is comprised of energy required for maintenance plus
energy required for protein accretion and energy required for lipid accretion (Kielanowski, 1965).
Thus, the energy used by the pig, or the energy required by the pig for a given level of performance,
can be estimated in a factorial manner by summing that used for: (1) maintenance; (2) protein
accretion; and (3) lipid accretion in the body. However, in order to undertake such a calculation,
one must be able to quantify: (1) maintenance energy requirements; (2) the rate of protein and lipid
accretion; and (3) the efficiency with which dietary energy is used for each (Table 2).
Table 8. Typical complete diet energy content of practical Danish diets, according to ME, NE and PPE levels
(Mcal/kg) (Anonymous, 2008).
Phase ME NE PPE ME NE PPE
Mcal/kg Relative, where weaner diet = 100
Lactation 3.18 2.29 1.89 92 91 91

Gestation 3.01 2.10 1.77 88 84 85
6-9 kg 3.44 2.51 2.08 100 100 100
9-30 kg 3.37 2.49 2.06 98 99 99
30-100 kg 3.20 2.22 1.89 93 88 91

Maintenance (Em) is often described as energy expenditures by the pig not associated with protein
and lipid gain (Kotarbińska and Kielanowski, 1969). Thus, maintenance will include such things as
basal metabolism, normal protein turnover, nutrient digestion and absorption and adaptive functions
such as thermoregulation, immune function and coping with social stressors (Knap, 2009).
Maintenance energy has also been described as daily intake of energy that results in neither
gain nor loss of tissue (Kleiber, 1961). While simple in concept, it is essentially impossible to
achieve experimentally, because energy intake that achieves zero change in weight or zero
change in energy retention may result in concurrent and counterbalancing changes in protein
or lipid tissues. As a result, maintenance is typically measured indirectly. One example is to
feed pigs decreasing quantities of daily energy, measuring energy retention at each, and then
extrapolating to determine the daily energy intake that results in zero energy balance (Figure
4). Alternatively, maintenance can be determined using indirect calorimetry by measuring
fasting heat production (FHP) and adding to this value energy required for activity (van
Milgen and Noblet, 2003). Fasting heat production, in turn, can be measured by estimating
heat production at zero energy intake (Figure 4). This assumes that the determination occurs
within the pig’s thermoneutral zone, so that energy is not being expended to maintain body
temperature. It also assumes that the value for Em is not influenced by energy intake, which
also may not be true (Labussiere et al., 2011).
Activity appears to require some 8 to 15% of total ME consumed by the pig (Le Bellego et al.,
2001; van Milgen et al., 2001), although the exact amount will vary widely depending on the
social and environmental conditions under which the animals are housed. The energy required
to support activity is a difficult value to estimate accurately, because it can only be measured in
Kg
Daily energy retained, Er
Em
Daily energy intake, Em
Km
FHP
Figure 4. Diagrammatic representation of the relationship between energy intake and energy retention in
the growing pig. Em = energy intake at maintenance, which is defined as the level of energy intake at which
energy retention is zero; FHP = fasting heat production; Kg = efficiency with which energy intake is used for
growth; Km = efficiency with which energy intake is used for maintenance.

J.F. Patience
respiration chambers, where activity will be much less than that observed under typical farm
conditions.
It is important to acknowledge that heat production and maintenance are not the same things.
Heat production has three components: fasting heat production, which represents 30 to 60%
of ME intake, the thermic effect of feeding, which represents 9 to 17% of ME intake and heat
produced as a consequence of physical activity (8%). Overall, more than 50% of the metabolizable
energy consumed by the pig is lost as heat (Van Milgen and Noblet, 2000). Oresanya (2005)
reported that only about 30% of total ME consumed is retained in the body as tissue in weanling
pigs. This was lower than the 43% reported by Le Bellego et al. (2001) using growing pigs.
While nutritionists have conventionally – and conveniently – defined maintenance energy

requirements using a single estimate based on body weight, the biological reality is somewhat
different. It is generally recognized that maintenance energy (Em) is to a first approximation a
function of body weight raised to a power value. There has been considerable discussion on what
that function should be, and for many years, BW0.75 appeared to be the preferred value (Close
et al., 1983; Knap, 2009; NRC, 1998; Whittemore, 1993); in their comprehensive review of the
literature at that time, ARC (1981) argued for an exponent of 0.63, but many publications adopted
an exponent of 0.75 (NRC, 1998). However, data over the past decade or more continue to suggest
that an exponent of 0.60 is preferable (Noblet et al., 1999) and was adopted by NRC (2012). When
a single value is required, the NRC (2012) estimate of the maintenance requirement, expressed
in ME units, is 197 × BW0.60.
Sometimes, Em has been expressed on the basis of protein mass in the body rather than total
body weight, recognizing that protein tissue is much more metabolically active than the body as a
whole (Whittemore, 1993). Noblet et al. (1999) pursued this question further and determined that
a combination of body protein and visceral mass could explain a large portion of the variation in
Em among pigs. In fact, on average, viscera represented 3 times the amount of energy required for
maintenance compared to muscle mass. Taking into account differences in demands by protein
mass and viscera, there remained unexplained differences among genotypes – about 10% of the
total – suggesting that there is a genetic line component to Em. This is supported by Fandrejewski
et al. (2000), Knap (2009) and Barea et al. (2010).
In summary, maintenance energy is a concept that is impossible to achieve experimentally. This

does not mean that it does not exist, but rather that scientists must define it indirectly rather than
directly. Understanding maintenance energy is crucial to our ability to characterize and quantify
how the pig uses dietary energy in total, because daily energy intake must equal maintenance
energy plus energy retained as protein and energy retained as fat, all divided by the efficiency
with which dietary energy fulfills these functions (Kielanowski, 1965):
MEintake = MEm + PD/kp + LD/kl
where MEm is the daily ME intake required for maintenance, PD is protein deposition, kp is the fractional efficiency with
which ME intake is used for protein accretion, LD is lipid (or fat) deposition and kl is the fractional efficiency with which
ME intake is used for lipid accretion. See Table 2 for an example of this calculation.

Protein deposition (PD) or protein gain is driven by sufficient supply of energy and amino
acids. PDmax, is the upper limit of protein deposition achieved or achievable by the pig. PDmax
is sometimes referred to as the maximum capacity of the pig for protein gain, and only achieved
if nutrient supply is adequate and negative influences in growth, such as social stressors, disease
and environment are absent (De Lange et al., 2001). PDmax is thought to be constant between
20 and 100 kg (Moughan and Verstegen, 1988). PDmax and observed PD are most likely to be
equal under commercial conditions between 50 and 90 kg bodyweight (De Lange et al., 2001;
Duijvesteijn et al., 2010). From a modeling perspective, the PD determined during this period,
and when a non-limiting diet is fed, can be referred to as the ‘operational’ PDmax (De Lange et
al., 2001). However, Bikker (1994) disputes this conclusion; his data suggest that among pigs with
a high potential for protein accretion, PDmax will not be reached below 80 to 90 kg bodyweight.
Protein gain requires about 10.03 kcal ME per gram of gain, compared to 11.65 kcal ME per gram
of lipid gain (Table 9). Of course, lean gain is much more efficient than lipid gain, because lean
contains so much water. Lean gain can be converted to protein gain by dividing by 2.55 (NRC,
1998). One can estimate the quantity of energy used by a pig for protein gain by determining
protein gain and multiplying this amount by 10.0 kcal/g (Table 2). The portion of body weight
gain that is protein varies widely based on genotype and diet, but growth of weanling pigs will
typically be in the range of 16 to 18%, while that of finishing pigs will be in the range of 15 to 17%.
It is then a relatively simple task to determine the energy available for lipid gain by subtracting
the energy required for maintenance (and for thermal homeostasis and other ‘maintenance’
functions) and the energy required for protein gain from total daily energy intake. Dividing this
energy value by 11.7 kcal/g will estimate the quantity of lipid gain likely achievable by the pig in
Table 9. Estimated energetic efficiency (ME basis) and the energy cost of protein and lipid deposition in
growing pigs (Adapted from Oresanya, 2005).
Reference kp1 kl1 Energy cost of protein Energy cost of lipid

deposition (kcal ME/g)2 deposition (kcal ME/g)2
Close et al., 1983 0.57 0.82 9.72 11.39

ARC, 1981 0.56 0.74 9.89 12.62
Quiniou et al., 1996 0.49 0.81 11.31 11.53
Williams et al., 1997 0.48 0.69 11.54 13.54
Van Milgen and Noblet, 1999 0.51 0.92 10.86 10.15
Noblet et al., 1999 0.62 0.84 8.94 11.12
Danfer, 2000 0.58 0.90 9.55 10.38
Van Milgen and Noblet, 2003 0.60 0.80 9.23 11.68
Barea et al., 2010 0.60 0.75 9.23 12.45
Mean 0.56 0.81 10.03 11.65
1 Partial efficiencies of protein and lipid deposition, respectively.
2 Assuming energy retained in protein is 5.54 kcal/g and in lipid is 9.34 kcal/g (Birkett and De Lange, 2001b). Obviously,
the kp and kl observed with any particular diet will vary, depending on its relative content of starch, protein and fat.

J.F. Patience
question. The pig modeled in Table 2, using this relatively simple method, reveals much about
energy use by the growing pig. About 35% is used for maintenance function and only about 20%
is used for protein gain. Almost half of the energy consumed by this pig will go to fat accretion.
Thus, almost three times as much energy is used for lipid accretion as for protein accretion, a
ratio that agrees with the actual measurements reported by Noblet et al. (1999). Williams et al.
(1997) reported that immune system stimulation did not alter the efficiency with which the pig
uses dietary energy for protein gain or lipid gain, but it did increase maintenance requirements.
Daily energy intake
While the concentration of energy in the diet is important, the true determinant of pig
performance is daily energy intake (Figure 5). Knowing intake requires knowledge of how much
the pig is eating; it is much easier to define the energy concentration in a diet than it is to define
daily energy intake, because very few farms know feed intake during the course of the wean-to-
finish or feeder-to-finish period. This is unfortunate.
It is generally assumed that under farm conditions, feed intake is limiting pig performance, so
increasing the concentration of energy in the diet will often result in increased daily energy
intake and thus pig growth rate. However, this assumption is not universally true. If feed intake is
somehow compromised, then increasing dietary energy concentration will indeed increase daily
energy intake. If intake is not compromised, then increasing dietary energy concentration will
actually result in lower feed intake and daily energy intake will remain the same.
Consider the response curve shown in Figure 6. As the concentration of energy in the diet
increases, the daily supply of energy to the pig will rise, up to a point beyond which further
elevation in dietary energy concentration will achieve no further increase in daily energy intake.
A critical problem for nutritionists is knowing this response curve and understanding it under
Daily energy intake
Dietary energy concentration
Figure 5. Response curving showing the relationship between dietary energy concentration and daily energy
intake in the growing and finishing pig.

80 kg BW
Daily protein deposition

65 kg BW
95 kg BW
Daily ME intake over maintenance
Figure 6. Stylized response curve showing protein deposition rates (Pd) in pigs at three body weights versus
daily energy intake after maintenance requirements are met. Notice that Pd plateaus at a higher level for
pigs at 80 kg than at 65 kg or 95 kg, but the slope is lower for the heavier pigs, indicating lower efficiency of
energy used for protein gain (De Lange et al., 2001; Quiniou et al., 1995).
diverse farm conditions. A review of relevant literature shows that the daily energy intake response
by growing and finishing pigs is highly variable. For example, Apple et al. (2009) studying pigs
from 28 to 114 kg, increased dietary ME content from about 3.34 Mcal/kg to about 3.55 Mcal/kg
and saw no increase in growth rate or feed efficiency, but daily ME intake rose from 7.6 Mcal/d to
about 8.0 Mcal/d. Not surprisingly, the increased energy, supplied by added fat, increased carcass
and intramuscular fat. Collins et al. (2009) saw a different response. They added up to 6% added
fat in 1% increments to finishing pigs weighing from 64 to about 95 kg and reported an increase
in growth rate up to the highest fat level but daily ME intake increased only to the 3% added fat
level and was constant at higher levels of fat. Feed conversion improved in a linear response to
dietary energy. So, unlike Apple et al. (2009), they saw an increase in both growth rate and feed
efficiency, something that is fairly common in commercial practice. Daily ME intake was 8.02
Mcal in the control diet, peaking at 8.38 Mcal at 5% added fat. Stein et al. (1996) increased dietary
energy concentration from 2.7 Mcal ME/kg to 3.5 Mcal ME/kg and like Collins et al. (2009)
observed a curvilinear increase in growth rate and a linear improvement in feed efficiency. Daily
ME peaked at 10.8 Mcal/d. Finally, Beaulieu et al. (2009) reported two experiments investigating
the response of growing and finishing pigs to dietary energy concentration. In the first study,
ME ranged from 2.94 Mcal/kg to 3.39 Mcal/kg. Daily energy intake increased in a curvilinear
fashion, peaking at 8.32 Mcal ME/d, as did growth rate, while feed efficiency improved in a linear
fashion. In a second study conducted in a commercial farm, ME was increased from 2.96 to 3.26
Mcal/kg, resulting in daily energy intake rising from 8.68 Mcal/d to 9.10 Mcal/d. Feed efficiency
improved, especially in the growing period, but average daily gain was constant. Looking at these
and other studies, one can see that increased dietary energy concentration may result in increased
growth rate or may result in improved feed efficiency, or may result in a combination of the two.
Interestingly, these experiments reported daily ME intake peaking at anywhere from 8.5 Mcal/d
to over 9.0 Mcal/d; in some cases, it can rise above 10 Mcal per day. We see energy intake on
some commercial farms in the 6.5 to 8.0 Mcal/d range. These inconsistent results make it very

J.F. Patience
difficult for nutritionists to anticipate the results of changes they might make in dietary energy
concentration. However, our current ‘working hypothesis’ is that pigs with a lower daily energy
intake are more likely to respond to increases in dietary energy concentration. Since energy is such
an expensive component of the diet, our inability to be more precise in our recommendations
creates an unhealthy financial situation for our industry.
Other considerations
Dietary energy may have other influences that are critical to the successful feeding of pigs. For
example, if energy is supplied in the diet using added fats, the selection of the nature and the
composition of that fat can not only affect the energy content of the diet, but also the quantity
and quality of fat in the carcass (Realini et al., 2010; Kellner et al., 2012). Excess polyunsaturated
fatty acids (PUFA) in the carcass fat can lead to processing problems and possibly shelf life issues
as well. These PUFAs added through supplemental fat will have an additive effect with fat that is
present in basal ingredients in the diet, such as corn or corn distillers dried grains. There are data
suggesting that elevations in PUFA in carcass fat can occur in only 3 to 4 weeks after changing
the diet (Apple et al., 2009; Wiseman and Agunbiade, 1998) although Benz et al. (2011) suggest
that changes in carcass unsaturation is time dependent – longer feeding of unsaturated fatty acids
increases the magnitude of the problem.
Practical approaches to improving feed efficiency
As we improve our understanding of energy metabolism in the pig, we gain insight into practical
ways in which we can use this knowledge to improve feed efficiency on individual farms.
Following is a checklist of ideas that derive from the previous discussion in this chapter.
1. Manage the pigs and the barn to minimize maintenance energy requirements.
Maintenance energy is a function of body weight, but it is also influenced by the level of
activity of the pig, by the energy required to maintain the immune and other systems whose
role it is to protect the pig from external insults (disease, stress, etc.) and by the need to
maintain constant body temperature.
2. Increase the portion of the GE of the diet that is digested.
Even in a highly digestible diet consisting of corn and soybean meal, 15% of the gross energy
ends up in the manure. Producers have control over this portion, and can reduce losses due
to low digestibility by reducing particle size, pelleting the diet and using exogenous enzymes.
As corn prices have risen, particle size targets have declined; whereas they used to be 750 µ
(microns) a decade ago, some producers have set their sights on 300 to 350 µ. Goodband et
al. (2002) suggested that every 100 µ reduction in particle size over the range from 1,200 to
400 µ improved feed efficiency by 4 points in finishing pigs. Patience et al. (2011) reported
that reducing the standard deviation of particle size from 2.33 to 1.88 µ increased energy
digestibility from 81 to 85%.
3. Determine actual energy content of diets used on the farm.
There is considerable variation in the quantity of digestible or metabolizable energy in any
diet, due to ingredient variability, differing digestion capabilities of various ages of pigs,
particle size, pellets versus mash, etc. It makes it very difficult for a nutritionist to effectively
manage dietary energy content. The solution is relatively easy and not too expensive. An

ingredient called titanium dioxide can be added at 0.4% to the diet as a digestibility marker
and fed for about 5 days. On the 5th day after introducing the marker, fresh fecal samples can
be collected from representative (~10%) pens in the barn. The fecal samples can be pooled
and submitted along with a sample of the feed for analysis of the marker and energy and from
this, determination of the actual DE content of the diet. ME can be estimated from the DE
value (Noblet and Perez, 1993; Le Goff and Noblet, 2001). There is one challenge with this
method, and that is the problem of coprophagy – pigs consuming feces from the floor of their
pen. This will introduce an error into the calculation and steps should be taken to minimize
it, such as regular scraping of the test pens until the collection is completed.
4. Feeding diets with low heat increment during hot weather to minimize heat stress.
The digestion and metabolism of food generates heat; in normal conditions, this is what helps
to keep pigs warm. But in periods of elevated barn temperatures, this heat is a decided negative
to pig performance. Indeed, the pig is designed physiologically to reduce feed intake during
periods of heat stress; this may be desirable from a health and physiological perspective, but
it also reduces growth rate and elevates feed efficiency. Consequently, it is important to select
diet formulations that will minimize heat increment during the hot summer months – those
lower in protein and fiber and/or higher in fat.
5. Minimize health demands placed on the pig.
We know that activation of the immune system increases the energy required for maintenance.
Williams et al. (1997) reported this increase to be in the range of 13%. Since health problems
lower energy intake because feed intake is reduced and also reduce the digestibility of energy
(Jones and Patience, 2012), the elevation of maintenance energy requirements is another blow
to efficiency. Illness causes a multitude of problems in terms of energy metabolism in the pig,
all resulting in much less efficiency. Therefore, it is clear that keeping disease out of the barn
should be a very high priority for all pork producers, because the presence of disease is very
costly.
6. Selection of pigs that use dietary energy more efficiently for maintenance.
Selection for improved feed efficiency has already advanced to selecting for lower maintenance
energy requirements, and this trend will not only continue, but become even more
sophisticated as tools for selection, and our understanding of energy metabolism, advance in
parallel with each other. Since maintenance represents about one-third of total daily energy
intake, this is a very reasonable target for genetic selection programs in the future – and we
know that variation among genotypes does exist, allowing for selection to occur (Knap, 2009).
7. Put a high priority on feed intake.
As the industry in the United States, and in many other parts of the word, is forced to adopt
lower energy diets in order to reduce feed costs, the benefits accruing to barns with high feed
intakes will be greater than many people probably appreciate. The challenge is revealed in
Table 10. In this case, three scenarios were modeled, using Cargill’s Pork MAX growth model,
reflecting how pigs under different circumstances might respond to reductions in dietary
energy concentration: (1) pigs are able to increase their feed intake in response to lower
energy diets and thus maintain equal daily energy intake; (2) pigs are not able to increase
their feed intake, so daily energy intake goes down, but pigs are still fed to the same final
market weight – at a much older age; and (3) pigs are not able to increase their feed intake, so
daily energy intake goes down, and pigs go to market in the same amount of time as scenario
1 and thus at a smaller weight. The differing outcomes are striking. Market weight dropped

J.F. Patience
Table 10. Impact of lowering diet NE content, in order to reduce feed cost, on pig performance outcomes
under differing pig response scenarios (modeling of animal performance undertaken using Cargill Pork MAX).
Diet NE (Mcal/kg) 2.25 2.31 2.40 2.47

Diet ME, Mcal/kg) 3.15 3.24 3.33 3.42
1. Increasing feed intake; constant market weight1

Initial weight (kg) 22.7 22.7 22.7 22.7
Final weight (kg) 132.3 132.0 132.2 132.0
Days to market 125.2 124.9 125.7 126.1
Gain (g/d) 876 876 871 867
Feed intake (kg/d) 2.61 2.53 2.44 2.36
Feed efficiency 0.336 0.346 0.356 0.366
NE intake (Mcal/d) 5.87 5.84 5.86 5.83
NE conversion (Mcal/kg gain) 6.70 6.67 6.73 6.72
2. Constant feed intake; constant market weight2
Final weight (kg) 132.3 132.0 132.1 132.0
Days to market 138.9 133.9 130.3 126.1
Gain (g/d) 789 817 839 867
Feed intake (kg/d) 2.39 2.38 2.37 2.36
Feed efficiency 0.330 0.344 0.355 0.366
NE intake (Mcal/d) 5.38 5.50 5.69 5.83
NE conversion (Mcal/kg) 6.82 6.73 6.78 6.72
3. Constant feed intake; constant days3
Final weight (kg) 122.6 126.6 128.9 132.0
Days to market 125.8 127.2 125.9 126.1
Gain (g/d) 0.794 0.817 0.844 0.867
Feed intake (kg/d) 2.33 2.35 2.35 2.36
Feed efficiency 0.340 0.348 0.358 0.366
NE intake (Mcal/d) 5.24 5.43 5.64 5.83
NE conversion (Mcal/kg) 6.60 6.64 6.68 6.72
1 As diet NE declines, pigs are able to increase feed intake to maintain constant daily energy intake; all pigs finished
to a constant weight.
2 As diet NE declines, pigs are unable to increase feed intake so daily energy intake declines; all pigs are still finished
to a constant weight but slower growth rate results in increased days to market.
3 As diet NE declines, pigs are unable to increase feed intake so daily energy intake declines; pigs are marketed at
the same time, irrespective of diet, since the barn must turn over according to a fixed schedule, so market weight
declines due to slower growth.

from 132 kg to as low as 123 kg, if pigs were marketed at the same time as in scenario 1, or
days to market increased to as high as 139 days, if the pigs are held to the same final weight
as scenario 1.
There is no doubt that feed intake will become a much higher priority in the future, even
though it has been a priority in the past. Managing a barn in such a manner that maximizes
feed intake will allow pigs to adjust to lower energy diets and thus maintain daily energy
intake. This, in turn, will support less expensive feeding programs with little or no impact on
pig growth rate.
8. Determine feed intake curves for individual barns or production systems.
The importance of knowing feed intake in a given barn or system cannot be overstated. Close-
out data are insufficient, and even if the number is correct, it only represents intake averaged
across the whole growout period – far too imprecise for the purposes of developing optimal
feeding programs. The benefit of having accurate feed intake curves is reflected in the growth
scenarios summarized in point 7 above. A barn manager will only know how his pigs are
responding to lower energy diets, and when that response occurs, if feed intake curves are
known.
Conclusion and implications
Energy metabolism is a complex subject, but our understanding of how the pig utilizes energy
and how we can most effectively supply energy to the pig is growing very rapidly. With this
knowledge in hand, we are well positioned to manage the relationship between dietary energy
concentration and feed efficiency. Concurrent with understanding how the pig uses energy, we
must also understand what the calculation of feed efficiency is telling us, ensuring that we avoid
erroneous management decisions because the information was misapplied. Most importantly,
it is only through understanding energy metabolism that we can improve feed efficiency to our
greatest advantage, reducing costs and using feed resources most effectively.
References
Adeola, O. 2001. Digestion and balance techniques in pigs. Pages 903-916 In Swine Nutrition, A.J. Lewis
and L.L. Southern, eds. CRC Press, Boca Raton, FL.
Anonymous. 2008. Danish pig production: nutrient standards. CVR-no. 31-07-14-37. Danish Pig
Production, Copenhagen, Denmark.
Apple, J. K., C. V. Maxwell, D. C. Brown, K. G. Friesen, R. E. Musser, Z. B. Johnson, and T. A. Armstrong.
2004. Effects of dietary lysine and energy density on performance and characass characteristics of
finishing pigs fed ractopamine. J. Anim. Sci. 82:3277-3287.
Apple, J. K., C. V. Maxwell, D. L. Galloway, S. Hutchison, and C. R. Hamilton. 2009. Interactive effects
of dietary fat source and slaughter weight in growing-finishing swine: I. Growth performance and
longissimus muscle fatty acid composition. J. Anim. Sci. 87:1407-1422.
ARC. 1981. The Nutrient Requirements of Pigs. Commonwealth Agricultural Bureaux, Slough, UK.
Barea, R., S. Dubois, H. Gilbert, P. Sellier, J. van Milgen, and J. Noblet. 2010. Energy utilization in pigs
selected for high and low residual feed intake. J. Anim. Sci. 88:2062-2072.
Beaulieu, A.D., N.H. Williams, and J.F. Patience. 2009. Response to dietary digestible energy concentration
in growing pigs fed cereal-grain based diets. J. Anim. Sci. 87:965-976.

J.F. Patience
Bender, D. A., and A. E. Bender. 1997. Nutrition: A Reference Book. Oxford University Press, Oxford, UK.
Benz, J. M., M. D. Tokach, S. S. Dritz, J. L. Nelssen, J. M. DeRouchey, R. C. Sulabo and R. D. Goodband.
2011. Effects of choice white grease and soybean oil on growth performance, carcass characteristics, and
carcass fat quality of growing-finishing pigs. J. Anim. Sci. 89:404-413.
Bensaude-Vincent, B. 1996. Between history and memory: Centennial and bicentennial images of lavoisier.
Isis 87(3):481-499.
Bikker, P. 1994. Protein and lipid accretion in body components of growing pigs: Effects of body weight and
nutrient intake. Ph.D. Dissertation. Wageningen Agricultural University, Wageningen, the Netherlands.
Birkett, S., and C. F. M. de Lange. 2001a. A computational framework for a nutrient flow representation of
energy utilization by growing monogastric animals. Br. J. Nutr. 86:661-674.
Birkett, S., and C. F. M. de Lange. 2001b. Calibration of the nutrient flow model of energy utilization by
growing pigs. Br. J. Nutr. 86:675-689.
Birkett, S., and C. F. M. de Lange. 2001c. Limitations of conventional models and a conceptual framework
for a nutrient flow representation of energy utilization by animals. Br. J. Nutr. 86:647-659.
Black, J. L. 1995. Modelling energy metabolism in the pig – critical evaluation of a simple reference model.
Pages 87-102 in Modelling Growth in the Pig, EAAP Publication no. 78. P.J. Moughan, M.W.A. Verstegen
and M.I. Visser-Reyneveld, ed. Wageningen Pers, Wageningen, the Netherlands.
Boisen, S. 2007. A new concept for practical feed evaluation systems. Publ. No. 79, Faculty of Agriculture,
Aarhus University, Fulum, Denmark.
Brouwer, E. 1965. Report of sub-committee on constants and factors. Pages 441-443 in Proc. 3rd Symp. Eur
Assoc. Anim. Prod. K.L. Blaxter, ed. Publication No. 11, Academic Press, London, UK.
Close, W. H., F. Berschauer, and R. P. Heavens. 1983. The influence of protein:energy value of the ration and
the level of feed intake on the energy and nitrogen metabolism of the growing pig. Br. J. Nutr. 49:255-
269.
Collins, C. J., A. C. Philpotts, and D. J. Henman. 2009. Improving growth performance of finisher pigs with
high fat diets. Anim. Prod. Sci. 49:262-267.
CVB. 1993. Netto energie van voedermiddelen voor varkens: argumentatie van de niewe Nev-formule. CVB
Report No. 7, Lelystad, the Netherlands.
Danfer, A. 2000. Model simulation of energy metabolism and utilization in growing pigs. Pages 293-296 in
Energy Metabolism in Animals. A. Chwalibog and K. Jakobsen, eds. Wageningen Pers, Wageningen,
the Netherlands.
De Greef, K. H., and M. W. A. Verstegen. 1995. Evaluation of a concept on energy partitioning in growing
pigs. Pages 137-149 in Modelling Growth in the Pig. P. J. Moughan, M. W. A. Verstegen and M. I. Visser-
Reyneveld, eds. Wageningen Pers, Wageningen, the Netherlands.
De Lange, C. F. M., B. J. Marty, S. Birkett, P. Morel, and B. Szkotnicki. 2001. Application of pig growth models
in commercial pork production. Can. J. Anim. Sci. 81:1-8.
Dove, C. R. 1993. The effect of adding copper and various fat sources to the diets of weanling swine on
growth performance and serum fatty acid profiles. J. Anim. Sci. 71:2187-2192.
Drummon, K. E. 1996. Dictionary of Nutrition and Dietetics. Van Nostrand Reinhold, New York, NY.
Duijvesteijn, N., B. A. N Silva, and E. F. Knol. 2010. Modelling protein deposition: defining pig genotypes and
deriving their nutritional requirements. Pages 601-602 in Energy and Protein Metabolism and Nutrition,
EAAP 127. G.M. Crovetto, ed., Wageningen Academic Publishers, Wageningen, the Netherlands.
Ewan, R.C. 2001. Energy utilization in swine nutrition. Pages 85-94. In Swine Nutrition, A. J. Lewis and L.
L. Southern, eds. CRC Press, Boca Raton, FL.

Fandrejewski, H., S. Raj, D. Werenko, L. Buraczewska, and G. Skiba. 2000. Protein and energy metabolism in
pigs of two genotypes. Pages 373-376 in Energy Metabolism in Animals. A. Chwalibog and K. Jakobsen,
eds. Wageningen Pers, Wageningen, the Netherlands.
Fox, D. G. C. J. Sniffen, J. D. O’Connor, J. B. Russell, and P. J. Van Soest. 1992. A net carbohydrate and protein
system for evaluating cattle diets: 3. Cattle requirements and diet adequacy. J. Anim. Sci. 70:3578-3596.
Goodband, R.D., R.D. Tokach, and J.L. Nelssen. 2002. The effects of diet particle size on animal performance.
Extension Bulletin, MF-2050. Kansas State University, Manhattan, KS.
Graham, H., K. Hesselman, and P. Aman. 1986. The influence of wheat and suger-beet pulp on the
digestibility of dietary components in a cereal-based pig diet. J. Nutr. 116:242-251.
Hinson, R. B., B. R. Wiegand, M. J. Ritter, G. L. Allee and S. N. Carr. 2011. Impact of dietary energy level
and ractopamine on growth performance, carcass characteristics and meat quality of finishing pigs. J.
Anim. Sci. 89:3572-3579.
Holmes, F. L. 1985. Lavoisier and the Chemistry of Life. Univ. of Wisconsin Press, Madison, WI.
Ihde, A. J. 1964. The Development of Modern Chemistry. Harper and Row, New York, NY.
Jones, C. K., and J. F. Patience. 2012. Birth weight and transition ADG can serve as tools to manage within-
barn body weight variation. J. Anim. Sci. 90 (E-Suppl. 2):74.
Just, A. 1982a. The net energy value of balanced diets for growing pigs. Livest. Prod. Sci. 8:541-555.
Just, A. 1982b. The net energy value of crude (catabolized) protein for growth in pigs. Livest. Prod. Sci.
9:349-360.
Kellner, T. A., K. J. Prusa, and J. F. Patience. 2012. Carcass iodine values taken from three carcass sites are
affected by dietary fat level and source during the finishing period. J. Anim. Sci. 90 (E-Suppl. 2):57.
Kielanowski, J. 1965. Estimates of the energy cost of protein deposition in growing pigs. Pages 13-20 in Proc.
3rd Symp. Energy Metabolism. K.L. Blaxter, ed. Academic Press, London, UK.
Kleiber, M. 1961. The Fire of Life. John Wiley and Sons, New York, NY, USA.
Knap, P.W. 2009. Allocation of resources to maintenance. Pages 110-129 in Resource Allocation Theory
Applied to Farm Animal Production. W.M. Rauw, ed. CAB Int., Cambridge, MA, USA.
Kotarbińska, M., and J. Kielanowski. 1969. Energy balance studies with growing pigs by the comparative
slaughter technique. Pages 299-310 in Energy Metabolism in Farm Animals. K. L. Blaxter, J. Kielanowski
and G. Thorbek, eds. Oriel Press Ltd., Newcastle Upon Tyne, UK.
Labussière, E., J. van Milgen, C.F.M. de Lange and J. Noblet. 2011. Maintenance energy requirements of
growing pigs and calves are influenced by feeding level. J. Nutr. 141:1855-1861.
Le Bellego, L., J. van Milgen, S. Dubois, and J. Noblet. 2001. Energy utilization of low-protein diets in
growing pigs. J. Anim. Sci. 79:1259-1271.
Le Goff, G., and J. Noblet. 2001. Comparative total tract digestibility of dietary energy and nutrients in
growing pigs and adult sows. J. Anim. Sci. 79:2418-2427.
Linneen, S. K., J. M. DeRouchey, S. S. Drtiz, R. D. Goodband, M. D. Tokach, and J. L. Nelssen. 2008. Effects
of dried distillers grains with solubles on growing and finishing pig performance in a commercial
environment. J. Anim. Sci. 86:1579-1587.
Moe, P. W. 1981. Energy metabolism of dairy cattle. J. Dairy Sci. 64:1120-1139.
Moughan, P. J., and M. W. A. Verstegen. 1988. The modelling of growth in the pig. Neth. J. Agric. Res.
36:145-166.
National Swine Nutrition Guide. 2010. Tables of Nutrient Recommendations, Ingredient Composition and
Use Rates. U.S. Pork Center of Excellence, Ames, IA.
Noblet, J. 2005. Recent advances in energy evaluation of feeds for pigs. pp 1-25 in Recent advances in animal
nutrition. P. C. Garnsworthy and J. Wiseman, eds. Nottingham University Press, Nottingham, UK.

J.F. Patience
Noblet, J., Y. Henry, and S. Dubois. 1987. Effect of protein and lysine levels on the diet and body gain
composition and energy utilization in growing pigs. J. Anim. Sci. 65:717-726.
Noblet, J., and Y. Henry. 1993. Energy evaluation systems for pig diets: a review. Livest. Prod. Sci. 36:121-141.
Noblet, J., and J. M. Perez. 1993. Prediction of digestibility of nutrients and energy values of pig diets from
chemical analysis. J. Anim. Sci. 71:3389-3398.
Noblet, J., H. Fortune, X. S. Shi, and S. Dubois. 1994. Prediction of net energy of feeds for growing pigs. J.
Anim. Sci. 72:344-354.
Noblet, J., C. Karege, S. Dubois, and J. van Milgen. 1999. Metabolic utilization of energy and maintenance
requirements in growing pigs: Effects of sex and genotype. J. Anim. Sci. 77:1208-1216.
Noblet, J., and G. Le Goff. 2001. Effect of dietary fibre on the energy value of feeds for pigs. Anim. Feed Sci.
Tech. 90:35-52.
NRC. 1998. Nutrient Requirements of Swine. National Academy Press, Washington, DC.
NRC. 2012. Nutrient Requirements of Swine. National Academy Press, Washington, DC.
Oresanya, T.F., A.D. Beaulieu, and J.F. Patience. 2008. Investigations of energy metabolism in weanling
barrows: The interaction of dietary energy concentration and daily feed (energy) intake. J. Anim. Sci.
86: 348-363.
Oresanya, T. F. 2005. Energy metabolism in the weanling pig: Effects of energy concentration and intake on
growth, body composition and nutrient accretion in the empty body. PhD Disseration. University of
Saskatchewan, Saskatoon, Canada.
Patience, J. F., and R. K. Chaplin. 1997. The relationship among dietary undetermined anion, acid-base
balance, and nutrient metabolism in swine. J. Anim. Sci. 75:2445-2452.
Patience, J. F., A. D. Beaulieu, R. T. Zijlstra, T. Oresanya, and R. Mohr. 2004. Energy systems for swine: A
critical review of DE, ME and NE. Proc. Midwest Swine Nutr. Conf. Indianapolis, IN.
Patience, J. F., P. Leterme, and A. D. Beaulieu. 2006. Advantages of the net energy system. Proc. Pre-
symposium Worshop ‘Net Energy Systems For Growing And Finishing Pigs.’ 10th Int. Symp. Digestive
Physiology in Pigs, Vejle, Denmark. May 24.
Patience, J. F. 2011. Obvious – and not so obvious – challenges associated with the adoption of new ingredients
in swine feeding programs. Proc. 19th Annual Swine Disease Conference for Swine Practitioners. Ames, IA.
Patience, J. F., A. Chipman, C. K. Jones, and T. Scheer. 2011. Varying corn particle size distribution affects
the digestibility of energy for the growing pig. J. Anim. Sci. 89 (E-Suppl. 1):127.
Quiniou, N., J. Noblet, J. van Milgen, and J. Y. Dourmad. 1995. Effect of energy intake on performance,
nutrient and tissue gain and protein and energy utilization in growing boars. Anim. Sci. 61:133-143.
Quiniou, N., J. Y. Dourmad, and J. Noblet. 1996. Effect of energy intake on performance of different types of
pig from 45 to 100 kg body weight. 1. Protein and lipid deposition. Anim. Sci. 63:277-288.
Realini, C. E., P. Duran-Montgé, R. Lizardo, M. Gispert, M. A. Oliver, and E. Esteve-Garcia. 2010. Effect of
source of dietary fat on pig performance, carcass characteristics and carcass fat content, distribution and
fatty acid composition. Meat Sci. 85:606-612.
Rijnen, M. M. J. A., J. Doorenbos, J. Mallo, and L. A. Den Hartog. 2004. The application of the net energy
system for swine. Proc. 25th Western Nutr. Conf., Saskatoon, SK.
Russo, S., and M. Silver. 2011. Introductory Chemistry. Prentice Hall, Upper Saddle River, New Jersey, USA.
Schiemann, R., K. Nehring, L. W. Hoffman, and A. Chudy. 1972. Energetische Futterbewertung und
Energienormen. VEB Deutscher Landwirtschaftsverlag, Berlin, Germany.
Stein, H. H., J. D. Hahn, and R. A. Easter. 1996. Effects of decreasing dietary energy concentration in finshing
pigs on carcass composition. J. Anim. Sci. 74(Suppl. 1: 65.

Stipanuk, M. N. 2006. Biochemical, Physiological, Molecular Aspects of Human Nutrition 2nd ed. Saunders,
St. Louis, Missouri.
Thorbek, G. 2000. Measurements of energy metabolism. How did it start? Pages 11-15 in Energy Metabolism
in Animals. A. Chwalibog and K. Jakobsen, eds. Wageningen Pers, Wageningen, the Netherlands.
Van Es, A. J. H. 1980. Energy costs of protein deposition. In: Protein Deposition in Animals, Butterworth,
London, 215-224.
Van Milgen, J., and J. Noblet. 1999. Energy partitioning in growing pigs: The use of a multivariate model as
an alternative for the factorial analysis. J. Anim. Sci 77:2154-2162.
Van Milgen, J, and J. Noblet. 2000. Modelling energy expenditure in pigs. Pages 103-114 in Modelling Nutrient
Utilization in Farm Animals. J. P. McNamara, J. France and D. E. Beever, eds. CABI, Wallingford, UK.
Van Milgen, J. Noblet, and S. Dubois. 2000. Energetic efficiency of nutrient utilization in growing pigs.
Pages 329-332 in Energy Metabolism in Animals. A. Chwalibog and K. Jakobsen, eds. Wageningen Pers,
Wageningen, the Netherlands.
Van Milgen, J., J. Noblet, and S. Dubois. 2001. Energetic efficiency of starch, protein and lipid utilization in
growing pigs. J. Nutr. 131:1309-1318.
Van Milgen, J., and J. Noblet. 2003. Partitioning of energy intake to heat, protein and fat in growing pigs. J.
Anim. Sci 81(E. Suppl. 2):E86-E93.
Whittemore, C. 1993. The Science and Practice of Pig Production. Longman Scientific and Technical, Essex,
UK.
Williams, N. H., T. S. Stahly, and D. R. Zimmerman. 1997. Effect of chronic immune system activation on
the rate, efficiency, and composition of growth and lysine needs of pigs fed from 6 to 27 kg. J. Anim.
Sci. 75:2463-2471.
Wiseman, J., and J. A. Agunbiade. 1998. The influence of changes in dietary fats and oils on fatty acid profiles
of carcass fat in finishing pigs. Livestock Prod. Sci. 54:217-227.
Wondra, K. J., J. D. Hancock, K. C. Behnke, R. H. Hines, and C. R. Stark. 1995. Effects of particle and
pelleting on growth performance, nutrient digestibility and stomach morphology in finishing pigs. J.
Anim. Sci. 73:757-763.

6. Feed processing to maximize feed efficiency
C.R. Stark
North Carolina State University, College of Agriculture and Life Sciences, Poultry Science and
Nutrition, 234 Scott Hall, Raleigh, NC 27695, USA; charles_stark@ncsu.edu
Abstract
Swine production operations must undertake a comprehensive review of their feeding program,
which includes the selection of ingredients, feed manufacturing processes, and the physical
form of the feed presented to the pig. The increased demand for cereal grains in the bio-fuels
sector has led to greater competition for the limited supply of grains, which has resulted in a
significant increase in grain prices in recent years, especially corn. Companies that are willing to
embrace new technology and evaluate by-products will have the greatest likelihood of long term
profitability in the swine industry. The manager of a feed mill must be both willing and able to
quickly adjust to changes in alternative ingredient markets and then implement new technology
in order to take advantage of these changes to reduce the cost of feed. The swine producer must
focus on reducing the total cost of the feed used to produce pork, which may require an increase
in the feed manufacturing cost in order to reduce the total cost of production. The feed mill
should be designed to facilitate changes in both the procurement of ingredients and the selection
of manufacturing technology based on the least cost production model of the company. The
purchasing agent, nutritionist, veterinarian, and feed mill manager must develop an integrated
approach to purchase ingredients and formulate, manufacture, and deliver feed to the animal
that will minimize the total cost of animal production. There must be a synergistic relationship
between the live animal production group and the feed manufacturing group or commercial feed
mill. Long term goals and objectives of the company should focus on procurement of quality
ingredients, formulation based on optimal profitability, feed manufacturing based on the least cost
formulas, improved animal feed conversion, and lower feed delivery costs. Feed manufacturing
practices that were effective in the past may no longer be beneficial to the swine producer when
faced with rising ingredient, energy, and transportation costs. Feed mills and animal production
units must continuously evaluate the value of ingredients, particle size, mash versus pelleted feed,
and the cost of delivery in today’s volatile ingredient market. Swine producers must view the feed
mill as a resource that can be utilized to manufacture safe high quality feed in order to meet the
long range goals of their business production plan.
Introduction
Swine producers must recognize and take advantage of the cost savings opportunities that exist
within their feed manufacturing operations in order to lower their production cost and improve
feed conversion. The feeding program (Figure 1) should provide a framework for the development
of the goals and objectives that may be required to reduce the feed cost associated with swine
production. The feed mill must be viewed as an instrument that can reduce the total cost of
the feed and improve animal performance. The feed mill has the potential to reduce feed costs
through their selection and utilization of quality ingredients, effective and efficient operation of
manufacturing equipment, and delivery of feed based on the feeding schedule of the animals.

C.R. Stark
Swine producers must understand the constraints and cost saving opportunities that exist within
their own feed mill or the commercial mill that sells them feed. The most common constraints
within a feed mill are the limited number and volume of ingredient storage bins, which can
constrain the purchase and inclusion of local grains and by-products. The number and volume
of ingredient storage bins at a mill will determine if a sufficient quantity of a new ingredient can
be procured and included in the diet at a level that would justify the procurement, logistics, and
matrix development costs associated with the ingredient.
Ingredient selection and least cost formulation
Nutritionists, purchasing agents, and feed mill managers must recognize and address the potential
challenges associated with the addition of new ingredients to the formulation mix. These may
include the following (Stark, 2009):
• added analytical costs to develop a new ingredient matrix profile;
• additional receiving time at the feed mill and the extra hours required for logistics management
to have the ingredient transported to the feed mill;
• slower scaling/batching times due to the addition of another ingredient that must be weighed
on the scale, which can be compounded if the ingredient has poor flow characteristics and
creates a bridged bin;
• reduced inventory of the primary ingredients, which may result in outages that require the
mill to switch to an alternative non-least cost formula;
Feeding
program
Least coast
Quality assurance Manufacturing
formulation
Ingredients Ingredient matrix Grinding
Formula
Feed Mixing
specification
Feed plan Pelleting
Figure 1. Example of a feeding program.

• reduced pellet mill throughput due to the addition of ingredients such as DDGs or the
reduction in phosphate sources;
• changes in pellet quality, which may increase the need for feeder management at the farm
and feed wastage;
• changes in feed density that will affect sows fed with volumetric feeders;
• changes in feed palatability that may cause a decrease in feed consumption for a short period
of time while the animal adjusts to the new taste and texture of the feed.
Operations that address these potential constraints or limitations of a feed mill prior to purchasing
ingredients will limit ingredient driver wait time, demurrage, formulation changes, and reduced
animal performance. Communication with the feed mill manager, quality assurance laboratory,
and production units prior to purchasing new or alternative ingredients will help minimize the
hidden costs that could be associated with an ingredient change. Nutritionists, purchasing agents,
and feed mill managers must develop an ingredient purchasing strategy that does not negatively
impact the feed mill and animal performance.
Growing conditions, drying, processing, and storage methods can affect the nutrient content of
both cereal grains and ingredients. Replacing cereal grains with alternative ingredients and by-
products has the potential to lower feed cost but increase the variation of nutrients in the finished
feed. The variation that exists in raw materials due to growing conditions or the base raw material
can be further expanded due to differences in process methods between suppliers and plants that
produce by-product ingredients. Therefore, least cost ingredient matrix values should take into
account both supplier and plant variation. Batal and Dale (2006) reported significant nutrient
variation in 17 samples of Dried Distillers Grains (DDGs) obtained from six ethanol plants in the
Midwest (Table 1), the range of nutrient content illustrates the difference that exist between plants.
The energy content of feed fat will vary based on the raw material from which it is derived (Meeker,
2006); in addition to energy content, swine producers should also be aware of differences in the
iodine value of fat, which influences belly quality (Table 2).
In addition to variation in growing conditions and processing methods, analytical methods

should be taken into account when developing matrix values and ingredient specification sheets.
A comparative analysis experiment on the same DDGs sample conducted by the American Feed
Industry Association (AFIA), Renewable Fuels Association (RFA), and National Corn Growers
Association (NCGA) suggested the method of analysis resulted in differences in the reported
percentage of moisture and fat (Table 3) in the sample (AFIA, 2007).
Table 1. Nutrient variability based on ethanol plant.
Crude protein (%) Crude fat (%) Crude fiber (%) Ash (%)
Average 27 8.8 6.6 4.4

SD 2 2.3 0.8 0.4
Range 23-30 2.5-10.6 5.1-8.1 3.9-5.4

C.R. Stark
Table 2. Energy and iodine value of feed fats based on source.
Fat Source Swine ME1 (kcal/kg) Iodine
Yellow grease 8,059 Varies

Poultry fat 8,010 77-80
Choice white grease 7,887 63-65
Tallow 7,594 43-45
1 ME = metabolizable energy.
Table 3. Moisture and fat results of dried distillers grains with solubles samples based on different analytical
testing methods.
Method Description Average moisture (%) SD
AOAC 934.01 loss on drying (vacuum) 10.67 0.25

AOAC 935.29 loss on drying (103 °C/5 h) 10.17 0.15
NFTA 2.2.2.5 loss on drying (105 °C/3 h) 9.87 0.18
AOAC 930.15 loss on drying (135 °C/2 h) 12.69 0.19
AOAC 2001.12 moisture (Karl Fischer) 9.03 0.08
Method Description Average crude fat (%) SD
AOAC 2003.05 crude fat (ethyl ether) 9.22 0.28

AOAC 954.02 fat (acid hydrolysis) 13.03 0.57
AOAC 945.16 crude fat (pet ether) 8.85 0.24
AOAC 2003.06 crude fat (hexane) 9.00 0.19
The moisture content of cereal grains has the potential to change when stored in grain elevators
and on-farm bins; a decrease in moisture of 1.5% can result in a 55 kcal/kg increase in the digestible
energy of corn, which can significantly change the cost of the final feed as well as affect animal
performance. The development of the ingredient matrix nutrient values in a least cost formulation
package must be based on samples collected prior to or during the receiving process. The sampling
process is a critical step in determining the nutrient content of the ingredient and the values that will
ultimately be entered into the formulation package. Improper sampling of ingredients can result in
errors that will affect the nutrient content of the feed and possibly feed conversion.

Feed manufacturing quality assurance
Swine producers should have in place a comprehensive feed quality assurance (QA) program
(Figure 2) that encompasses policies and procedures required for the purchase of ingredients, the
manufacture of feed, and the delivery of the feed to the farm. Producers who purchase their feed
should be aware of the QA programs in place at the feed mill that manufactures their feed. The
amount of detail and content associated with a QA program will be based upon the goals of the
company and be customized to match the equipment and manufacturing processes in the feed
mill. The QA program must address the quality specifications of ingredients, the manufacturing
processes, packaging, finished feed delivery, and critical control points to ensure that the feed
conforms to the company’s nutritional and physical specifications at the time of sale or delivery
(Stark, 2010). In addition to the regulatory requirements, the QA program should address
third party certification program (PQA, HACCP, ISO, and Safe Feed/Safe Food) requirements
that would be in place at the feed mill. A comprehensive QA program will include employee
training, ingredient specifications and traceability, QA manual, standard operating procedures,
critical control points, sampling and analytical schedules, reporting systems, and review
processes (Stark, 2010).
The purchase of quality ingredients is the first step in the production of consistent, safe, high
quality swine feed. Feed mills are not typically designed to segregate ingredients at the time of
receipt at the feed mill; therefore, purchasing from suppliers who provide a consistent ingredient
Quality
assurance
Ingredient Feed Regulatory and

QA Lab
specification manufacturing certification
Nutrients
Receiving Receiving FDA
levels
Rejection Commercial
Analytical Processing
procedures feed law
Matrix ISO, HACCP, Safe

Claims Finished feed
development feed/safe food
Figure 2. Design of a comprehensive quality assurance program.

C.R. Stark
will reduce the nutrient variation in the final feed and ultimately variation in animal performance.
Feed mills do not have schemes for improving poor-quality ingredients, but through the use of
ingredient segregation, grinding equipment, long-term conditioning, and pelleting, the nutrient
utilization of ingredients can be improved (Stark and Jones, 2011). The QA manual should outline
the policies and procedures that must be in place to ensure a safe high quality feed product is
manufactured and delivered to the customer or animals (Stark, 2010). These key items include:
• monitoring the particle size of the ground cereal grains on a routine basis;
• reviewing the daily feed production records to insure the scales and meters are properly
calibrated and that ingredients were added in the correct amounts;
• performing a mixer uniformity test to insure that ingredients are added in the correct
sequence and that the feed is properly mixed;
• monitoring the conditioning temperature during the pelleting process and the final moisture
content of the finished feed after the cooling process;
• evaluating both nutrient content of the finished feed and the physical characteristics of the
feed such as pellet quality as determined by the pellet durability index (ASAE, 1987).
The feed mill’s QA program should use a systematic process for identifying, monitoring, and
controlling possible sources of microbiological, chemical, and physical hazards, starting
with raw materials and ending with product consumption (Fairfield, 2009). The new federal
regulations established by the Food Safety and Modernization Act (FDA, 2011) will require
feed manufacturers to conduct a hazard analysis of their processes, as well as to develop and
implement a written preventive control plan that includes monitoring, verification, corrective
actions, and recordkeeping (Epperson, 2011).
Particle size analysis
The approved method for particle size analysis is ANSI/ASAE S319.3 FEB03 ‘Method of
determining and expressing fineness of feed materials by sieving’ (ASABE, 2007). Sample division
is a critical step prior to analysis and should be done with a sample splitter. The approved method
specifies the sieve sizes, sieving agent, and equipment (Ro-tap™ shaker). The use of sieve dispersing
agents and agitators in the procedure should be noted on the report. The estimated particle size
of a sample will be significantly lower when these modifiers are added to the procedure used to
determine the micron size of the sample. Stark and Chewning (2011) reported that the addition
of a sieve agent to hammermill ground corn reduced the estimated particle size of the sample
from 411 to 332 microns, whereas the addition of the sieve agitators only resulted in a 41 micron
decrease in particle size (392 versus 351). However, the addition of both the sieve agent and agitator
reduced the reported particle size from 443 to 323 microns. Goodband et al. (2006) reported a
similar decrease of 80 microns when a dispersing agent was added to samples. The results of these
experiments suggest the addition of a sieve dispersing agent and agitators will lower the reported
geometric mean diameter (dgw), thus producing a better estimate of the true particle size.
Mixer uniformity test
Mixer uniformity should be tested upon installation and checked a minimum of once a year
but preferably twice a year. Mixer uniformity should be based on a single source nutrient (dry

synthetic amino acids, salt, etc.) that has a low analytical variance (Hermann and Behnke, 1994).
Ten samples should be obtained from the mixer by probing the mixer or collecting samples at
equally spaced time intervals during the discharge process. The samples should be analyzed for
the selected ingredient to determine the uniformity of the mix. The coefficient of variation (CV%)
for the ten samples should be less than 10%.
standard deviation × 100%
CV% =
mean
Herrman and Behnke (1994) provided guidelines for interpretation of mixer uniformity results
and potential corrective actions (Table 4).
Once the mix time (dry and wet mix) has been established it should not be changed without
validating the new mixing times. The mixer should be regularly inspected for material build up
on the ribbons, paddles, and shafts.
Pellet durability test
The effectiveness of the pelleting process is measured by pellet quality as defined by the pellet
durability index (PDI), as well as the percent of fines at the mill or the trough of the feeder on the
farm. A comprehensive pellet quality model should be created for each individual production
system. Feed mills that have a method to estimate PDI and percent fines at the feeder can make
adjustments in the milling process to meet or exceed the pellet quality targets at the feeder trough.
Additionally, the model can provide feedback to the nutritionist and purchasing agent as to the
positive or negative effect of an ingredient or formulation change on PDI and percent fines. Pellet
quality can be determined with the standard Pellet Durability Index (ASAE 5269.3) developed
at Kansas State University (ASAE, 1987). The method may need to be modified through the use
of additional hex nuts to create a model that is representative of a company’s manufacturing and
delivery process (Pacheco and Stark, 2009).
Establishing a specification for percent fines at the feeder will allow the nutritionist, purchasing
agent, and feed mill manager to formulate, purchase, and manufacture feed to the specification
while exploring options to lower the cost of the feed. Stark (1994) reported that feeding pelleted
diets which contained 60% fines resulted in a feed conversion that was similar to feeding a mash
Table 4. Interpretation of mixer uniformity test.
CV Rating Corrective action
<10% Excellent None

10-15% Good Increase mixing time by 25-30%
15-20% Fair Increase mixing time by 50%, look for worn equipment, overfilling, or
sequence of ingredient addition
>20% Poor Possible combination of all the above. Consult extension personnel or
feed equipment manufacturer

C.R. Stark
diet, thus negating the benefit of pelleting. The consistency of pelleted feed is as important as the
actual amount of fines. Inconsistency in the percentage of pellet fines between feed deliveries
requires more feeder management to minimize feed wastage. Animal production units that
specify 100% pellets at the farm can expect higher feed manufacturing costs. These costs may
be significantly higher if the feed mill is attempting to pellet ingredients with poor pelleting
characteristics which require more re-pelleting of the fines, thus decreasing the efficiency of the
pellet mill.
Feed manufacturing
Feed manufacturing is the process whereby individual ingredients with different physical
characteristics and nutrient contents are combined to create a feed that will provide the proper
level of nutrition for animal maintenance and growth. The feed manufacturing process can be
as simple as grinding, weighing, and mixing of ingredients to create a mash/meal feed or as
complex as grinding, weighing, mixing, long term conditioning, pelleting, and post-pellet liquid
application. Feed can be manufactured in a multi-million dollar automated mill or on the farm
with a tractor powered grinder mixer. Regardless of the mill design or size, all equipment must
be properly designed, installed, and maintained to produce a safe high quality feed. The major
advantages of new automated feed mills are their ability to capture and analyze processing data,
efficiently control multiple processes, and provide traceability of ingredients and feed. Modern
feed mills are designed with automation systems that allow them to operate with as few as 3 to
4 employees per shift. Additionally, automation systems provide traceability of ingredients and
feed for every aspect of the manufacturing process. This traceability is documented from point
of receipt of the ingredient to when finished feed is loaded onto the truck and transported to the
farm. During the manufacturing process data can be captured, analyzed, and reported within
the system or shared with business management and accounting systems through the use of
data transfer protocols. The data collected by the automation system can be used to develop
future process parameters, implement control points, and monitor changes in the manufacturing
process. The ability of a computer system to efficiently collect, analyze, and create reports on
processing parameters and the quality of finished feed is essential in order to optimize animal
performance at the lowest possible feed cost.
Feed mill design
The predominate design of feed mills in the United States has been the pre-batch grind design
(Figure 3), in which whole grains are ground and stored above the batch scale. This type of mill
was designed primarily to process a single source of grain, which has historically been corn that
has been plentiful and reasonably priced. In contrast to this design, feed mills in other parts of
the world have been designed to grind the cereal grain, protein source, and by-product ingredient
portion of the formula (Figure 4). The post-batch grind design allows a nutritionist to incorporate
by-product ingredients of various particle sizes into the diet, as well as the addition of multiple
sources of grain without multiple grinding systems. Another design advantage of the system is
that it will produce a diet with a more uniform distribution of particles, which will result in better
pellets and a lower percentage of pellet fines in the feed that is delivered to the farm.

Minerals/vitamins
Protein source
Ground grain
Whole grain
By-products
Scale
Grinder
Mixer
Figure 3. Pre-batch grind of grains design.
Ground ingredient
Minerals/vitamins
Protein source
Whole grain
By-products
Scale
Mixer
Grinder
Figure 4. Post-batch grind of ingredients design.
Multiple bins for each ingredient will be required in the mill if one of the goals of the company
is to include a high percentage of by-products in the diet as by-products tend to bridge and not
flow as readily as ground corn and soybean meal. The number of ingredient bins needed in the
mill should be based on both the type and amount of by-products and grains used in the formula.
The mill should have the ability to rotate, empty, and clean bins as needed in order to prevent
bridging and maintain a fresh inventory of ingredients.

C.R. Stark
The installation of a roller mill or hammermill should be based on the type of grains that will be
ground and the target particle size of the ground grain in the diet. Producers who want to feed
pellets will need to install a boiler and steam system, pellet mill, pellet cooler, and post-pellet
liquid application system.
Feed mills should have at least one finished feed bin for each feed type and/or medication option.
The lack of sufficient finished feed storage will result in an inefficient manufacturing and delivery
process, due to an increased number of feed changeovers in the batching and pelleting process.
The design of a feed mill should allow for future expansion and include the flexibility to install
new equipment due to advances in technology. The inability to quickly make changes within the
feed mill could be a constraint that prevents a producer from quickly taking advantage of cost
saving opportunities that lower total feed costs or improve animal performance.
Particle size reduction
Whole grains are typically ground in the feed mill, whereas ingredients such as soybean meal,
rendered products, distillers dried grains, and wheat by-products are ground by the supplier.
Owens and Heimann (1994) indicated the major reasons for particle size reduction are to:
• expose a greater surface area for digestion;
• improve ease of handling of some ingredients;
• improve mixing characteristics of ingredients;
• improve pelleting efficiency and pellet quality.
Researchers continue to evaluate the optimal particle size for swine based on animal performance,
animal health, and processing costs.
Particle size reduction can be accomplished through the use of a roller mill or hammermill.
Each of type of equipment has its advantages and disadvantages that should be considered when
selecting the appropriate type of grinding equipment for the feed mill. Grinder selection should
be based on the anticipated ingredients used in the diet, target particle size, and the final form of
the feed (meal versus pellet). The producer should also take into consideration the operational
costs (electricity, labor, and maintenance) and capital investment of each grinding system. A
double pair roller mill can be used to efficiently grind non-fibrous material to a particle size of
500 microns. A hammermill can reduce the particle size of ingredients to less than 400 microns
when properly sized and fitted with the correct air assist system.
The cost of electrical energy associated with particle size reduction is small in comparison to the
total cost of personnel, facilities, and operations. The improved feed conversion that is linked to
a smaller particle size of the grain will reduce the amount of feed required to raise the animals
and therefore produce a positive return on investment. The cost of particle size reduction is
dependent on the target particle size, type of equipment (hammermill versus roller mill), and
the ingredients that must be processed. Grinding costs are inversely related to particle size; costs
increase as the target particle size decreases. These additional costs are associated primarily with
electricity, screens, hammers, and re-corrugation of rolls.

Hammermill
Particle size reduction in a hammermill occurs as a result of the impaction between the rapidly
moving hammer and a relatively slow moving particle (Heiman, 2005). The impaction continues
to occur until the velocity of the particle has decreased and is slow enough to exit through the
opening in the screen. Changes in moisture, fiber content, and genetic variety of ingredients
tend to have less of an effect on the particle size when ingredients are ground with a hammermill
as compared to a roller mill. The hammermill will produce a product with a wider variation
in particle diameters, especially when equipped with a large screen. The hammermill requires
minimal maintenance and monitoring throughout the week in order to maintain a consistent
particle size. Factors to consider when setting up the hammermill to achieve a target particle
size include: number of hammers, hammer setting (coarse versus fine), tip speed, and screen
size. Heiman (2005) reported that increasing the tip speed of the hammer will produce a smaller
particle size within a given screen (Table 5).
Feed mill managers who have a target of less than 500 microns need to routinely check the
efficiency of their hammermill. Monitoring the feeder rate above the hammermill and the
corresponding motor amps will provide an indication of grinding efficiency. The hammermill
will operate at its highest efficiency when the hammers and screens are new, as the square edge
of the screen holes and the corner of the hammers become round, the production efficiency of
the hammermill will decrease and electrical energy consumption will increase. Anderson (2010)
estimated that increasing the maintenance cost of the hammermill by 2.5 times would result
in a 24% reduction in total operating cost due to lower energy consumption per ton of ground
material.
Roller mill
The roller mill will produce a granular product by passing the product through a pair of horizontally
mounted rolls. The particle size is controlled by the width of the gap between the rolls, number
of corrugations per inch on the roll, the spiral of the corrugation, and the differential speed of
two rolls within the pair. Groesbeck et al. (2003) reported that roller milled corn samples had a
narrower range of particle diameters (Sgw 1.83 to 2.03) in samples that ranged from 503 to 1,235
microns in particle size (dgw). In contrast, the range of particles in hammermill ground corn was
wider (Sgw 2.12 to 2.52) in the samples, ranging from 390 to 980 microns; the reported variation
Table 5. Effect of hammer tip speed and screen size opening on the particle size (microns) of US #2 corn.
Tip speed (m/min) Screen size
1.6 mm 2.4 mm 3.2 mm 4.0 mm 4.8 mm
5,455 450 600 750 950 1,250

6,317 300 500 650 750 900
7,752 200 400 550 650 750

C.R. Stark
decreased as the particle size of the corn decreased. The one primary advantage of a roller mill as
compared to a hammermill is that it processes 15% to 40% more tonnes/hr to a similar particle
size using the same horsepower (Heiman, 2005).
Effect of particle size reduction on animal performance
The benefit of particle size reduction of cereal grains for swine has been well documented in
several swine studies (Table 6). The positive effects of particle size reduction on feed conversion
Table 6. Effect of cereal grain particle size reduction on swine performance.
Reference Microns ADG (kg)1 F:G2 Research notes
Ohh et al., 1983 624 0.46 1.70 hammermill, corn, starter pigs
877 0.45 1.78 hammermill, corn, starter pigs
822 0.46 1.81 roller mill, corn, starter pigs
1,147 0.47 1.92 roller mill, corn, starter pigs
539 0.44 1.78 hammermill, sorghum, starter pigs
722 0.45 1.79 hammermill, sorghum, starter pigs
885 0.45 1.92 roller mill, sorghum, starter pigs
1,217 0.43 1.94 roller mill, sorghum, starter pigs
Healey et al., 1994 369 0.42 1.44 hammermill, corn, starter pigs
Goodband and 698 0.93 3.39 hammermill, sorghum, finishing pigs
Hines, 1987 714 0.89 3.32 hammermill, barley, finishing pigs
902 0.92 3.56 hammermill, barley, finishing pigs
1,126 0.81 3.65 hammermill, barley, finishing pigs
2,200 0.79 3.72 roller mill, barley, finishing pigs
Wondra et al., 1995 400 0.98/0.99 3.22/3.01 hammermill, corn, mash/pellet, finishing pigs
600 0.95/1.02 3.43/3.13 hammermill, corn, mash/pellet, finishing pigs
Mavromichalis 387 0.43 1.19 hammermill, wheat, starter pigs
et al., 2000 568 0.45 1.16 hammermill, wheat, starter pigs
1,380 0.41 1.34 hammermill, wheat, starter pigs
628 0.89 3.44 hammermill, wheat, finishing pigs
1,288 0.88 3.57 hammermill, wheat, finishing pigs
1 ADG = average daily gain.
2 F:G = Feed:Gain ratio.

must be balanced with any potential adverse effects on animal health, especially in animals with
a genetic predisposition to gastric ulcers. A feed mill equipped with a properly sized hammermill
can routinely achieve an average particle size below 400 microns, which improves pellet quality
(Stark, 1994) and also feed conversion. Ohh et al. (1983) demonstrated fine grinding corn and
sorghum improved feed conversion in starter pigs. Healey et al. (1994) reported improvements
in starter pig performance when corn was reduced from 1,000 to 500 microns; however, Healey
et al. (1994) observed a negative effect on stomach mucosa at 300 microns. Mavromichalis et al.
(2000) reported improved feed conversion in starter pigs when the particle size of wheat was
reduced from 1,380 to 387 microns. Wondra et al. (1995) reported reducing the particle size of
corn from 1,000 to 400 microns resulted in an 8% improvement in feed/gain. Mavromichalis et al.
(2000) did not detect a change in finishing pig performance by reducing the particle size of wheat
from 1,288 to 628 microns; however, there was a 3% improvement in feed conversion when the
particle size of wheat was reduced from 614 to 406 microns. Goodband et al. (2002) developed a
regression equation to predict feed conversion from 400 to 1,200 microns in finishing pigs from
55 to 110 kg (Figure 5).
Fastinger and Mahan (2003) reported decreasing the particle size of soybean meal (SBM) improved
amino acid digestibility with the largest increase occurring when particle size was reduced from
900 to 600 microns. Lawrence et al. (2003) reported no difference in average daily gain (ADG)
or feed conversion by reducing the particle size of solvent extracted or extruder expelled SBM
in starter pigs. Similarly Traylor et al. (2005) reported no improvement in feed conversion or
phosphorus availability by reducing the particle size of meat and bone meal (Table 7). Reducing
the particle size of protein sources does not appear to have the same benefits as particle size
reduction in cereal grains. Nutritionists, veterinarians, and feed mill managers must determine
the optimal particle size for individual production units based on herd health status, genetics,
stage of production (sow, nursery, and grow/finish), electrical cost, capital investment, and diet
form (meal versus pellet).
3.60
3.50
3.40
3.30
3.20
3.10
F/G
3.00
400 500 600 700 800 900 1000 1,100 1,200
Microns
Figure 5. Effect of particle size on finishing pigs from 55 to 110 kg.

C.R. Stark
Table 7. Effect of protein source particle size reduction on swine performance.
Reference Microns ADG (kg)1 F:G2 Research notes
Lawrence et al., 2003 639 0.54 1.64 expeller soybean meal, starter pigs
742 0.54 1.64 expeller soybean meal, starter pigs
965 0.54 1.56 expeller soybean meal, starter pigs
444 0.48 1.54 solvent extracted soybean meal, starter pigs
797 0.49 1.54 solvent extracted soybean meal, starter pigs
1,226 0.48 1.54 solvent extracted soybean meal, starter pigs
Traylor et al., 2005 471 0.74 1.98 meat and bone, grower pigs
535 0.74 1.98 meat and bone, grower pigs
635 0.76 2.01 meat and bone, grower pigs
Batching/weighing
Batching and mixing are critical processing operations in the production of uniform feed that
meets the nutrient requirements of animals. Scale resolution should be selected based on both
ingredient and inclusion level. The least cost formula should be produced/based on the resolution
of the scales used to weigh the ingredients. A least cost formula that requires 357 kg of soybean
meal weighed on a scale with ±10 kg increments will always result in too much or too little
soybean meal added to the diet. A formulator who does not use an ingredient rounding factor
based on the scales in each mill will create scenarios in which animals do not receive the proper
amount of nutrients or a shrink in the ingredient inventory as a result of ingredient overages in
each batch of feed. These scenarios could result in poorer animal performance or higher feed
costs, respectively.
The scales and metering devices in a feed mill should be certified by an external company every
three to six months; this should be in addition to an internal scale check program that is routinely
done at the mill. At a minimum, the scales and metering devices in the feed mill should be
checked at installation and annually afterwards (CFR, 2011).
Mixing
The mixing process combines individual ingredients of various nutrient content, particle sizes,
shapes, and densities to produce a feed that will provide the appropriate level of nutrients to the
animal. The mixer should distribute the individual ingredients to produce a feed in which the
ingredients have a uniform distribution within the batch of feed. The most common mixers used
in the feed industry are vertical, double ribbon, and paddle mixers. The vertical mixer is typically
used in small on-farm mills and portable mixers. Double ribbon and paddle mixers are more
commonly used in commercial and integrated feed mills. High speed batching and mixing can

be accomplished through the use of a double shaft ribbon mixer, which will mix feed in less than
30 seconds as compared to the 2 to 3 minute mix time of a double ribbon single shaft mixer. The
selection of the mixer should be based on the ingredients used in the diets and the desired speed
and capacity of the mixing system. The density of ingredients can significantly affect the amount
of material that can be mixed in one batch. Although mixer size is often reported in kilograms, a
mixer has a specific design volume. Therefore, the density of the diet should be considered when
determining the minimum and maximum batch size of a mixer; under and over filling the mixer
can lead to improper mixing.
Pelleting
The pelleting process agglomerates ingredients that have different particle sizes, densities, and
flowability. There are many benefits of pelleting; nutritionists can formulate least cost diets with
ingredients that have poor flow characteristics, the particle size of the grain can be reduced
to less than 600 microns without affecting the handling characteristics of the finished feed,
and feed conversion is improved. Feed conversion is improved due to enhanced palatability,
reduced wastage, and the potential for improved nutrient utilization due to heat treatment of the
ingredients. Although the observed improvement in feed conversion from feeding pellets may
be confounded with the reduced particle size of the grain, the overall improved performance of
pelleted diets has been well documented (Table 8).
Pellets are produced by combining the mixed feed with steam in a paddle conditioner to increase
the moisture and temperature of the mash. The mash is retained and mixed in the conditioner for
15 to 90 seconds; retention time of the mash depends on the number and design of conditioners
above the pellet mill. The conditioned mash is then pressed through a ring die, which forms the
pellet to a specific diameter.
The pellets are cooled with ambient air to remove the moisture and heat that was added during
steam conditioning, which case hardens the pellet in its cylindrical shape. The goal of the cooling
process should be to reduce the moisture of the pelleted feed to the original or less than the
original moisture of the mash feed prior to conditioning. Residual moisture left in the pellets can
lead to the growth of mold, degradation of pellets during handling, and dilution of the energy
content of the feed. The energy content of a pelleted diet that contains an additional 1% moisture
after cooling will be significantly lower than the calculated energy content of the diet, which will
negatively affect feed conversion.
Feed mills that have pellet screeners can remove the fines after the pellets have been cooled and
then re-pellet the fines, thus creating a product with minimal fines when it leaves the feed mill.
This additional cost associated with re-pelleting fines and the lower daily production rate has
steered the industry to produce and ship feed that contains both fines and pellets. This process can
result in as many as 30 to 40% fines in the trough of the feeder at the farm. Feed mill managers,
customers, and animal production groups must set a realistic target for the maximum amount of
pellet fines allowed at the feeder based on the formula and feed manufacturing equipment used
to produce the pellets.

C.R. Stark
Table 8. Effect of pellets on swine performance.
Reference Diet form ADG (kg)1 F:G2 Comments
Skoch et al., 1983 mash 0.61 2.21 nursery

pellet 0.67 2.03 nursery
Harris et al., 1979 mash 0.61 3.85 grow-finish
poor pellets 0.66 3.80 grow-finish
good pellets 0.66 3.30 grow-finish
Hanarahan, 1984 PDI3 – 69% 0.49 3.99 grow-finish
PDI – 62% 0.49 3.94 grow-finish
Stark, 1994 mash 0.47 1.70 nursery
pellet – 0% fines 0.49 1.50 nursery
mash 0.43 1.95 nursery
Stark, 1994 mash 0.93 2.77 grow-finish
pellet – 0% fines 0.96 2.65 grow-finish
Wondra et al., 1995 mash 0.96 3.35 grow-finish
pellet 1.00 3.16 grow-finish
Schell and Van Heugten, pellet – 2.5% fines 1.92 grower
1998 pellet – 13% fines 1.94 grower
pellet – 25% fines 1.98 grower
Steidinger et al., 2000 meal 1.39 nursery
meal 5% SDAP4 1.37 nursery
pellet 60c 1.37 nursery
1 AVG = average daily gain.
3 PDI = pellet durability index.
4 SDAP = spray-dried animal protein.

Factors affecting pellet quality
Feed manufacturers may have different pelleting parameters and pellet quality standards based
on regulations, customer demands, and company philosophy. The pelleting process is influenced
by many variables that can be monitored and controlled during the process in order to achieve
maximum throughput while producing quality pellets. Equipment design and set-up, conditioning
time and temperature, formulation, ingredient particle size, cooling, and die specification (Figure
6) will affect the pelleting process and pellet quality (Behnke, 1994).
The manager of a feed mill can typically control the amount of fat added to the mixer, particle size,
and the conditioning temperature of the pelleting process. The addition of fat to the mixer prior to
the pelleting process will lower the electrical energy consumption, but also decrease pellet quality
(Stark, 1994, Pacheco and Stark, 2009). In addition to the improved feed conversion observed
with a smaller particle size, there is also an improvement in pellet quality. Stark and Wyatt
(2009) reported that pellet quality improved (46 to 62% PDI) as the conditioning temperature
increased due to the addition of more steam to the mash. Feed manufacturers should develop
pelleting parameters and targets for conditioning temperature and time, die specification, and
the maximum inclusion level of fat in the mixer. These parameters will determine the amount of
pellet fines observed at the pellet mill, bulk load-out, and the feeder trough at the farm. Pelleting
parameters, especially the conditioning temperature should be recorded for each pellet run and
Die specification,
15%
Cooling, 5%
Formulation, 40%
Particle size, 20%
Conditioning, 20%
Figure 6. Factors that affect pellet quality.

C.R. Stark
the corresponding pellet quality in order to determine if the pelleting parameters have produced
the quality of pellets that meet the customer’s or grower’s expectations.
Effect of pelleting on animal performance
The effects of pelleting on animal performance are presented in Table 8. Skoch et al. (1983) reported
an 8% improvement in ADG and F:G in nursery pigs. Stark (1994) reported similar benefits to
feeding a pellet diet to nursery pigs, with a tendency for better feed conversion when pigs were
fed screened pellets versus 25% fines. There was also a tendency for poorer feed conversion when
finishing pigs were fed a greater percentage of pellet fines. Pigs fed a pelleted finisher diet which
contained 60% fines resulted in a feed conversion similar to feeding mash diets, thus negating the
benefit of pelleting. Research conducted by Hanrahan (1984) did not show an improvement in
feed conversion; however, Harris et al. (1979) reported an improvement by feeding quality pellets
versus poor quality pellets in growing-finishing pigs. Schell and van Heugten (1998) reported
poorer feed conversion in grower pigs as the level of fines in the diets increased from 3% to 40%.
Inconsistency in the level of fines delivered to the farm will create additional feeder adjustment
as farm personnel attempt to prevent feed wastage. Continually adjusting feeders due to varying
levels of fines is often a complaint of animal production specialists and growers.
Post pellet liquid application
Feed mills utilizing post pellet liquid application (PPLA) to apply fat or liquid enzymes should
routinely test the accuracy of the dry flow equipment and liquid metering device. The system
should be tested at both low and high application rates to determine if the equipment can
consistently and accurately apply liquids under both conditions. Feed mills that apply liquid
enzymes through a PPLA system tend to have a larger degree of variation in analytical results and
thus should increase the frequency of their equipment calibrations, implement a liquid inventory
reconciliation process, and develop a sampling and monitor program that validates uniform and
accurate addition of the liquid enzymes (Froetschner, 2007).
Feed ordering and delivery
The feed ordering system used by a swine production company can impact the cost of feed
delivered to the animal. Feed ordering systems that rely on employees to order the feed
have a greater potential for errors as well as increased costs, lower animal performance, and
compromised medication strategies. The substitution of a feed type and overfeeding of a feed type
are the most frequent errors when employees are allowed to make independent decisions without
management approval. Manufacturing problems, ingredient delivery issues, and emergency farm
orders will always occur and create the need to substitute a feed type; however, written policies
and procedures for managing these issues will minimize the feed cost associated with substituting
the wrong feed.
Production companies can reduce their feed delivery cost by reducing the weight of the delivery
unit, insuring that the truck is filled to the legal limit, driving direct delivery routes, and looking
for backhaul opportunities (Stark, 2009). The tractor and feed delivery trailer should be selected

based on the condition of the roads and distance to farms. The purchase of lighter weight tractors
and trailers will increase the efficiency of the delivery fleet. Increasing the amount of feed on the
delivery unit from 24 to 27 tons will result in a significant reduction in delivery cost over the life
of the delivery unit. Small but simple changes to the units within a delivery fleet can also reduce
the weight of the units. Relatively inexpensive changes such as reducing the weight of fuel and
hydraulic oil on the delivery unit, replacing metal trailer lid covers with rolling tarps, and using
super singles tires on both the tractor and trailer will reduce the total weight of the delivery unit,
which will allow for the delivery of more feed per trip.
Conclusion
Swine producers must develop a business model based on the long term goals and objectives of
the company. These goals and objectives should be the focal point of all conversations related
to the purchase and manufacture of feed. The feed mill manager, nutritionist, veterinarian, and
animal production specialist must use an integrated approach that focuses on purchasing quality
ingredients, formulating diets based on optimal profitability, manufacturing feed based on the
least cost formula, improving feed conversion, and lowering the cost of delivering feed. The
incorporation of by-product ingredients into swine diets will continue to increase as more grains
are diverted to the bio-fuels industry. Feed manufactures and swine producers who can quickly
adapt their purchasing, formulating, and feed manufacturing processes to take advantages of
price changes in the ingredient market and shifts in supply will reduce the cost of their feed.
Swine producers must recognize how a single change in one part of their feeding program will
affect their feed costs and the profitability of their production system. Swine production groups
must regularly communicate with the feed mill to ensure that the feed manufacturing process is
focused on optimizing feed efficiency within their production system.
References
AFIA. 2007. Evaluation of Analytical Methods for Analysis of Dried Distillers Grains with Solubles.
American Feed Industry Association, Arlington VA.
ASABE. 2007. Method of determining and expressing fineness of feed by sieving. Pages 646-649 in ASABE
Standards 2007. American Society of Agricultural and Biological Engineers, St. Joseph, MI.
ASAE. 1987. Wafers, pellets, and crumbles-definitions and methods for determine density, durability, and
moisture content. Yearbook of Standards. P.325. American Society of Agricultural Engineers.
Anderson, S. 2010. Optimizing Hammermill Efficiency. Feed Mill Managers Seminar. US Poultry & Egg
Association. Nashville, TN.
Batal, A. B. and N. M. Dale. 2006. True metabolizable energy and amino acid digestibility of distillers dried
grains with soluble. J. Appl. Poult. Res. 15:89-93.
Behnke, K. C. 1994. Factors affecting pellet quality. Proc. Maryland Nutrition Conference, 20-25 March
1994. Department of Poultry Science, College of Agriculture, University of Maryland, College Park.
CFR, 2011. Current Good Manufacturing Practices for Medicated Feeds. FDA 21 CFR Part 225 Section 30.
FDA. 2011. Federal Food, Drug, and Cosmetic Act with respect to the safety of the food supply. Jan. 4,
2011 – [H.R. 2751].
Fairfield, D. 2009. The HACCP Approach to Feed Product Safety. National Grain and Feed Association,
Washington, DC.

C.R. Stark
Fastinger, N. and D. Mahan. 2003. Effect of soybean meal particle size on amino acid and energy digestibility
in grower-finisher swine. J. Anim. Sci. 81:697-704.
Froetschner, J. R. 2007. Micro-ingredient application and equipment: issues and advances. Animal Feed
Manufacturer Assn. Forum Proceedings. Sun City, South Africa.
Goodband, R. D. and R. H. Hines. 1987. The effect of barley particle size on starter and finishing pig
performance. J. Anim. Sci. 65(Suppl. 1):317.
Goodband, R. D., R. D. Tokach, and J. L. Nelssen. 2002. The effects of diet particle size on animal performance.
Kansas State University Extension Bulletin, MF-2050.
Goodband, R. D., W. Diederich, S. S. Dritz, M. D. Tokach, J. M. DeRouchey, and J. L. Nelssen. 2006.
Comparison of particle size analysis of ground grain with, or without, the use of a flow agent. Kansas
State University Swine Day Report 2006. Agric. Exp. Station. Manhattan, KS.
Groesbeck, C. N., R. D. Goodband, M. D. Tokach, J. L. Nelssen, S. S. Dritz, K. R. Lawrence, and C. W.
Hastad. 2003. Particle size, mill type, and added fat influence flowability of ground corn. Kansas State
Univ. Swine Day Report, p. 203.
Hanrahan, T. J. 1984. Effect of pellet size and quality on pig performance. Anim. Feed Sci. Technol. 10:277.
Harris, D. D., L. F. Tribble, and D. E. Orr Jr., 1979. The effect of meal versus different size pelleted forms of
sorghum-soybean meal diets for finishing swine. Proceedings of the 27th Annual Swine Short Course.
Texas Tech Univ., Lubbock, TX.
Healy, B. J., J. D. Hancock, G. A. Kennedy, P. J. Bramel-Cox, K. C. Behnke, and R. H. Hines. 1994. Optimum
particle size of corn and hard and soft sorghum for nursery pigs. J. Anim. Sci. 72:2227-2236.
Heiman, M. 2005. Particle size reduction. Feed Technology V. E. Schofield, ed. American Feed Industry
Assn. Arlington, VA, USA. pp. 108-126.
Herrman, T. and K. C. Behnke. 1994. Testing Mixer Performance. Kansas State University Extension
Bulletin, MF-1172.
Lawrence, K. R., C. W. Hastad, R. D. Goodband, M. D. Tokach, S. S. Dritz, J. L. Nelssen, J. M. DeRouchey,
and M. J. Webster., 2003. Effects of soybean meal particle size on growth performance of nursery pigs.
J. Anim. Sci. 2003. 81:2118-2122.
Mavromichalis, I., J. D. Hancock, B. W. Senne, T. L. Gugle, G. A. Kennedy, R. H. Hines and C. L. Wyatt. 2000
Enzyme supplementation and particle size of wheat in diets for finishing pigs. J. Anim. Sci. 78:3086-3095.
Meeker, D. L. and C. R. Hamilton. 2006. An overview of the rendering industry. Essential Rendering. D.L.
Meeker, ed. National Renderers Assn. Arlington, VA, USA. pp 1-16.
Ohh, S. J., G. Allee, K. C. Behnke, and C. W. Deyoe. 1983. Effect of particle size of corn and sorghum grain
on performance and digestibility of nutrients for weaned pigs. J. Anim. Sci. 57(Suppl. 1):260 (Abstr.).
Owens, J. M. and M. Heimann. 1994. Material processing cost center. Feed Technology IV. R.R. McElhinney,
ed. American Feed Industry Assn. Arlington, VA, USA. p. 81.
Pacheco, W. J. and C. R. Stark. 2009. Effect of feed sample weight on pellet durability index. Abstr. 348P.
Poult. Sci. Assn. 98th Annual Meeting.
Schell, T. C. and E. Van Heugten. 1998. The effect of pellet quality on growth performance of grower pigs.
J. Anim. Sci. 76(Suppl. 1):185.
Skoch, E. R., S. F. Binder, C. W. Deyoe, G. L. Allee, and K. C. Behnke. 1983. Effects of pelleting conditions
on performance of pigs fed a corn-soybean meal diet. J. Anim. Sci. 57:922.
Stark, C. R. 1994. Pellet quality I. Pellet quality and its effects on swine performance. PhD Dissertation,
Kansas State University, Manhattan, KS.
Stark, C. R. 2009. Feed Manufacturing To Lower Costs. National Hog Farmer Blueprints Focus. Apr. 15,
2009.

Stark, C. R. and C. Wyatt. 2009. Evaluation of a heat stable xylanase enzyme under typical feed industry
manufacturing parameters. Abstr. 191. Poult. Sci. Assn. 98th Annual Meeting.
Stark, C. R. 2010. Development of a Quality Assurance Program. Extension publication. North Carolina
Cooperative Extension, Raleigh, NC.
Stark, C. R. and F. T. Jones. 2011. Quality assurance programs in feed manufacturing. Feedstuff 2012
Reference Issue and Buying Guide.
Stark, C. R. and C. G. Chewning. 2011. The effect of sieve agitators and dispersing agent on the method
of determining and expressing fineness of feed materials by sieving. Anim. Prod. Sci. online 22
December 2011.
Steidinger, M. D., R. D. Goodband, M. D. Tokach, S. S. Dritz, J. L. Nelssen, L. J. McKinney, B. S. Borg,
and J. M. Campbell. 2000. Effects of pelleting and pellet conditioning temperatures on weanling pig
performance. J. Anim. Sci. 2000. 78:3014-3018.
Traylor, S. L., G. L. Cromwell, and M. D. Lindermann. 2005. Effects of particle size, ash content, and
processing pressure on the bioavailability of phosphorus in meat and bone meal for swine. J. Anim.
Sci. 83:2554-2563.
Wondra, K. J., J. D. Hancock, K. C. Behnke, R. H. Hines, and C. R. Stark. 1995. Effect of particle size and
pelleting on growth performance, nutrient digestibility, and stomach morphology in finishing pigs. J.
Anim. Sci. 73:757-763.

7. The genetic and biological basis of residual feed intake as a
measure of feed efficiency
Animal Science, Iowa State University, 239 Kildee Hall, Ames, IA 50011-3150, USA;
jdekkers@iastate.edu
Abstract
Feed efficiency traditionally has been evaluated either using a feed to gain ratio or a gain to feed
ratio. Starting in 1963 with Koch, a new manner of looking at feed efficiency came into play and
has become increasingly important. Koch adjusted feed consumed for gain and mid-weight in
order to evaluate residual feed intake (RFI) of individual animals. Animals with a more negative
residual feed intake are more efficient. Many studies have been performed in order to get a better
understanding of the biological basis behind RFI. Studies have been conducted to compare
animals differing in RFI to evaluate differences in traits such as total feed intake, growth and
other performance traits, meat quality, behavior, and digestibility. At Iowa State University, two
selection lines of pigs have been developed which differ in RFI as a resource population to study
the biological and physiological basis of feed intake and efficiency. The purpose of this chapter
is to summarize the main findings from this selection experiment in terms of the genetic and
biological basis of RFI in growing pigs. RFI was found to be moderately heritable (0.29±0.07) and
responded well to selection. Selection for decreased RFI resulted in pigs that ate less, are leaner,
grow slower, eat faster, have lower maintenance requirements, and have better gut integrity. No
detrimental effects were found for meat quality, litter size, litter performance, or response to
PRRS infection.
Introduction
Feed efficiency has been evaluated by looking at either the feed to gain ratio or the ratio of gain
to feed. However, neither of these measures account for the inherent differences in feed efficiency
between animals of varying sizes and growth rates. Therefore, Koch et al. (1963) proposed to
adjust feed consumption of growing animals for body weight gain and mid-weight. By measuring
feed consumption in this manner, Koch et al. (1963) evaluated the feed consumption between
different animals under constant gain and mid-weight, or animals with equal expected energy
requirements for gain and maintenance. Animals with lower adjusted feed consumption are more
efficient because they consume less feed based on their gain and mid-weight. This has become
known as residual feed intake (RFI), which is defined as the difference between observed feed
intake and expected feed intake based on expected requirements for production (i.e. growth, milk
production, piglet production, etc.) and maintenance. Expected requirements can be based on
NRC requirements but are typically derived empirically by regressing feed intake on production
performance (e.g. growth, backfat) and metabolic body weight (Kennedy et al. 1993). Measures
of RFI can differ based on what production traits are used to adjust daily feed intake for expected
requirements.

RFI has been extensively used as a measure to study the genetic and physiological basis of feed
efficiency in cattle and poultry. Research on RFI in pigs has been scarcer but has been augmented
in recent years based on two selection experiments for RFI, one at Iowa State University (ISU)
and one at INRA, France. The purpose of this review is to summarize research on the genetic and
physiological basis of RFI in cattle and poultry and field studies in pigs, followed by a review of
findings in the ISU selection experiment, with reference to results from the parallel experiment at
INRA, and implications for the physiological basis of differences in feed efficiency in growing pigs.
The genetic basis of residual feed intake
Koch et al. (1963) estimated the heritability of RFI in growing beef cattle to be 0.28, which
suggests a sizeable genetic component to RFI and that selection for RFI can be effective.
Later studies estimated the heritability of RFI at 0.22 in growing dairy heifers (Korver et al.,
1991), 0.19 in lactating dairy heifers (Van Arendonk et al., 1991), 0.016 in lactating dairy cows
(Ngwerume and Mao, 1992), and 0.275 in young beef bulls (Jensen et al., 1992). Kennedy et al.
(1993) evaluated the implications of selection on RFI and its component traits and concluded
that the heritability of and response to selection for RFI are a direct function of the genetic and
phenotypic parameters of its component traits, i.e. feed intake and the production traits that are
used to adjust feed intake.
Physiological basis of residual feed intake
Many factors can contribute to differences in RFI including body composition, physical activity,
maintenance requirements, digestibility, energetic efficiency, tissue turnover rates, and immune
response, along with measurement errors. Since RFI is by definition phenotypically uncorrelated
with the production traits that are included in its derivation, comparison of RFI between animals
which differ in level of production is allowed. Because RFI is independent of production at
the phenotypic level, RFI, in part, represents inherent variation in the basal metabolic state of
animals, which is supported by research by Herd and Bishop (2000), who found that genetic
variation in maintenance energy requirements is strongly associated with genetic variation in
RFI. As feed intake increases, the amount of energy required to digest the feed (heat increment
of feeding) increases, mainly due to the increased size and increased energy expenditure of the
digestive organs (Herd and Arthur, 2009). Since animals that differ in RFI also differ in feed
intake, animals with lower RFI are expected to have a lower heat increment of feeding. This is
supported by research in sheep by Webster et al. (1975), who found that feed intake accounted
for approximately 40% of total heat increment of feeding.
Variation in heat production, and therefore variation in energy available for maintenance and
growth, also results from differences in energy expenditure through differences in activity (Herd
and Arthur, 2009). Activity levels differed between hens which differed in RFI with low efficiency,
or high RFI, hens spending more time food-pecking, walking, pacing, and showing aggressive
behaviors than did high efficiency, or low RFI, hens (Braastad and Katle, 1989). Although there
was no difference in grooming between the two lines, the more efficient hens preened more often
in a sitting position while the less efficient hens were more apt to be standing while preening
(Braastad and Katle, 1989). Richardson et al. (1999) used pedometers to measure activity in

7. The genetic and biological basis of residual feed intake as a measure of feed efficiency
cattle and found the genetic correlation of RFI with pedometer count to be 0.32 which shows that
animals with decreased RFI were less active.
The rate of feed intake and the duration of a meal have also been shown to affect the energy cost
of eating in cattle, with more efficient cattle eating faster and spending less time feeding (Adam
et al., 1984). This is supported by studies that showed a decrease in feeding time (Robinson and
Oddy, 2004; Nkrumah et al., 2006, 2007; Lancaster et al., 2009; Durunna et al., 2011) and an
increase in eating rate (Robinson and Oddy, 2004) in cattle with lower RFI.
Differences in digestibility have been found in cattle that differ in RFI, with more efficient animals
being more efficient at digesting dry matter (Richardson et al., 1996). Differences in digestibility
account for 14% of the difference in feed intake (Richardson et al., 1996) and 19% of the variation
in RFI (Richardson and Herd, 2004). However, differences in digestibility between animals that
differ in RFI have not been observed in pigs (De Haer et al., 1993).
Body composition also appears to be related to differences in RFI. Steers with low RFI parents
had less whole-body chemical fat and greater whole-body chemical protein than steers with
high RFI parents (Richardson et al., 2001). Baker et al. (2006) found that steers with low RFI had
greater moisture content and lower fat content than steers with high RFI. Arthur et al. (2001)
found that subcutaneous fat depth measured over the 12th and 13th ribs and rump to have positive
genetic correlations with RFI of 0.17 and 0.06 in beef weanling bulls and heifers. When evaluating
yearling bulls of different beef breeds, Schenkel et al. (2004) found the genetic correlation of
backfat thickness with RFI to be 0.16, which is similar to the result found by Arthur et al. (2001).
Lancaster et al. (2009) found that heifers with low RFI had larger loins, less intramuscular fat, and
deposited less backfat than heifers with high RFI. These results suggest that selection for decreased
RFI would result in animals that are leaner. Although RFI is phenotypically uncorrelated with
backfat, there are other fat depots that can change and animals with lower RFI will have less fat
in these areas. Also, although adjustment for backfat forces the phenotypic correlation between
RFI and backfat to be zero, there is still the potential for a non-zero genetic correlation between
RFI and backfat (Kennedy et al., 1993).
The energetic efficiency of mitochondria function has also been shown to be associated with
differences in feed efficiency in cattle (Herd and Arthur, 2009). Steers with low RFI had a greater
rate of mitochondrial respiration and steers with high RFI had an impaired electron flux through
the electron transport chain (Kolath et al., 2006).
Variables in the blood plasma have been shown to be associated with differences in RFI. In beef
steers, Richardson et al. (2004) showed evidence that RFI is associated with plasma cortisol and
several red and white blood cell variables, suggesting that high RFI animals are more susceptible
to stress. Concentration of IGF-I in plasma at a young age has been shown to be associated with
growth, carcass traits, and feed efficiency in pigs (Bunter et al., 2005; Hoque et al., 2007) and in
beef cattle (Moore et al., 2005). Moore et al. (2005) found the genetic correlation of IGF-I plasma
concentration with RFI to be 0.57 while Bunter et al. (2005) found the genetic correlation of IGF-I
plasma concentration with feed conversion ratio to be 0.65. However, the correlation between
IGF-I concentration and RFI found by Hoque et al. (2007) was much lower at 0.16.

Differences in reproduction were also found between animals differing in RFI, both in cattle
(Shaffer et al., 2011) and in chickens (Morisson et al., 1997). Shaffer et al. (2011) found that low
RFI heifers reached puberty at an older age than did high RFI heifers. However, no differences
in pregnancy rate, conception rate, estrus detection rate, and inseminations per pregnancy
were found between heifers that differed in RFI (Shaffer et al., 2011). This differs from what
was found in chickens. Morisson et al. (1997) found that low RFI chickens had fewer eggs that
were not fertilized and a greater number of eggs that were hatched than did high RFI chickens.
This was due mainly to a decrease in early embryo mortality in the low RFI line since there were
no differences in late embryo mortality between lines (Morisson et al., 1997). Morisson et al.
(1997) also evaluated the semen from the two lines of chickens that differed in RFI. There was
no difference in the volume of semen between the two lines; however, the low RFI line had a
greater number of spermatozoa resulting in a higher spermatozoa concentration than the high
RFI line (Morisson et al., 1997). Motility was greater in the low RFI while the proportion of dead
spermatozoa was less in the low RFI line than in the high RFI line (Morisson et al., 1997). The
difference in spermatozoa count may account for part of the difference in fertilized eggs between
the two lines.
Selection experiment in Yorkshire pigs to create lines divergent in residual

feed intake
In order to enable study of the genetic and physiological basis of feed efficiency in pigs, a selection
experiment in Yorkshire pigs was started at Iowa State University, with the aim to develop lines
that differ in RFI. A similar selection experiment was initiated at INRA (France) around the same
time in Large White pigs (Barea et al., 2010; Lefaucheur et al., 2011).
Experimental protocols for this study were approved by the Iowa State University Institutional
Animal Care and Use Committee.
Experimental design
Using purebred Yorkshire pigs, a selection line for low RFI (LRFI line) and a randomly selected
control line were initiated in 2001. In generation 5, selection for increased RFI was initiated in
the randomly selected control line and will be referred to as the high RFI (HRFI) line in the
remainder of this chapter. Starting with the random allocation of littermates to the two lines in
generation 0 (the base population), ~90 boars from first parity sows and ~90 gilts from second
parity sows from the LRFI line were evaluated for RFI each generation. After evaluation of first
parity boars based on estimated breeding value (EBV) for RFI (see below), ~12 boars and 70 gilts
from the LRFI line were selected to produce ~50 litters for the next generation. After selection,
full- or half-sisters of the selected boars, produced from the second parity of their dams, were
evaluated for RFI to provide additional data for the next generation. Initially, in the HRFI line,
~10 boars and 40 gilts were selected at random to produce ~30 litters. Starting with generation
5, evaluation and selection protocols were the same in the HRFI line as in the LRFI line but with
selection for increased RFI.

For feed intake recording, pigs were put in pens of 16 pigs from the LRFI line at ~90 d of age, each
of which had a single-space electronic feeder (Feed Intake Recording Equipment, FIRE, Osborne
Industries Inc., Osborne, KS, USA). Starting with gilts from the second parity of generation 4, 8
pigs from the LRFI line and 8 pigs from the HRFI line were placed in each pen. Pigs were given
~1 wk to acclimate to the FIRE feeders before being weighed on test in groups by on-test date
based on age and body weight (typically two age groups per generation and parity). In general,
pigs were taken off test on an individual basis when they reached ~115 kg body weight, but were
removed at a lighter weight if few pigs remained in the pen, in which case all remaining pigs were
taken off test. Prior to generation 6, pigs with an off-test weight less than 102 kg were not scanned
and evaluated for RFI. Average daily feed intake (ADFI) during the test period was derived using
procedures described by Casey et al. (2005).
Each generation, selection of boars and gilts used to produce the next generation of pigs was
based on EBV for RFI, with some consideration of avoiding selection of full-sib brothers. Using
data from all generations up to that point, including the base population (generation 0) and the
generation that produced the base population (generation -1), EBV for RFI were obtained from
a single-trait animal model analysis of ADFI, with fixed effects of on-test group and sex, random
effects of litter and pen within on-test group, and linear covariates for interactions of on- and off-
test body weights, on-test age, average daily gain, and backfat with generation and line. Starting
with generation 5, metabolic mid-body weight was also implemented as a covariate. Further
details are provided in Cai et al. (2008), Bunter et al. (2010), and Young et al. (2011).
Genetic parameters and responses to selection
Cai et al. (2008) showed that 34% of the phenotypic variation in ADFI was accounted for by
RFI with the other 66% of variation being accounted for by average daily gain and backfat.
The heritability of RFI was estimated at 0.20±0.06 when including all data up to the first parity
of generation 8. As expected, selection for RFI resulted in decreased RFI and ADFI. After 8
generations of selection, the LRFI line had 241 g/d less RFI and 376 g/d less ADFI than the HRFI
line (Figure 1). Selection for RFI has resulted in an improvement in feed conversion ratio, with the
LRFI line requiring 0.22 g feed less per g weight gain than the HRFI line in generation 8. Other
performance traits that have changed as a result of selection for RFI are average daily gain and
ultrasonic backfat and loin eye area with LRFI pigs growing 79 g/d slower, having 2.5 mm less
backfat, and 1.5 cm2 larger loins than HRFI pigs, on average in generation 8. Similar results were
observed at INRA with their low RFI line having lower RFI, average daily feed intake, average
daily gain, and backfat thickness and a greater loin muscle area (Dekkers and Gilbert, 2010).
Cai et al. (2011) evaluated differences in feed intake and growth curves between the two lines.
Compared with the HRFI line, the LRFI line had a lower feed intake curve with line differences
being small at the beginning of the growing period but increasing later in the growing period
(Cai et al., 2011). The growth curve was similar between lines up to about 150 d of age at which
point the LRFI line had a lower body weight curve compared to the HRFI line (Cai et al., 2011). In
conclusion, selection for decreased RFI has resulted in lower intake and growth curves, especially
later in the growing period.

1.5
Control/High RFI line
RFI
1.0
FI
ADG
0.5
Response (genetic sd)
LEA +1.5 cm2

BF
0.0 LEA
Line
differences
-0.5 BF -2.5 mm
-1.0 ADG -79 g/d

FCR -0.22 g/g
-1.5
-2.0 FI -376 g/d

Low RFI line RFI -241 g/d
-2.5
0 1 2 3 4 5 6 7 8 9 10
Generation
Figure 1. Response to selection for residual feed intake. Line differences are LRFI – HRFI.
BF = back fat thickness; ADG = average daily gain; FCR = feed conversion ratio; LEA = loin eye area; FI = feed intake,
RFI = residual feed intake.
Impact of selection for RFI on meat quality
In a detailed study of differences in body composition and meat quality between the RFI lines,
Smith et al. (2011) also found that the LRFI line had larger loins and less backfat than the
HRFI line. Although not significantly different (P=0.15), pigs from the LRFI line had greater
hot carcass weights than pigs from the HRFI line (Smith et al., 2011). The greater hot carcass
weights, combined with decreased backfat, resulted in pigs from the LRFI line having a higher
percentage lean (P<0.05) than pigs from the HRFI line (Smith et al., 2011). Percent lipid was
lower and percent moisture was higher in the LRFI line than in the HRFI line (P<0.01; Smith
et al., 2011). Also, loins from the LRFI line tended to have less centrifugation loss (P=0.07)
and a lower Hunter b value (P=0.08) (Smith et al., 2011). There were no significant differences
in other objective measures of meat quality, including pH at 48 h, drip loss, purge loss, cook
loss, Hunter L, and Hunter a (P>0.13; Smith et al., 2011). There were no significant differences
(P>0.15) between lines for sensory panel traits of juiciness, tenderness, chewiness, pork flavor,
and off flavor. However, residual correlations of RFI with tenderness (0.24, P=0.002) and purge
loss (-0.16, P=0.04) suggested that continued selection for decreased RFI could decrease fresh
pork sensory traits since tenderness and purge loss are key players in meat quality. The INRA
selection experiment also found increased loin size, decreased backfat, and decreased IMF in the
line selected for low RFI (Lefaucheur et al., 2011). However, unlike Smith et al. (2011), Lefaucheur
et al. (2011) found that muscle from low RFI pigs had a lower pH 24 h postmortem compared
to the muscles from high RFI pigs. In the INRA lines, low RFI pigs had significantly greater L*

values than the high RFI pigs, which is associated with a greater drip loss value assessed at 2 d
postmortem (Lefaucheur et al., 2011). Smith et al. (2011) found no differences between lines
(P>0.35) for muscle fiber type but Lefaucheur et al. (2011) found that low RFI pigs had lower
proportion of myosin IIBR and greater proportion of myosin IIBW. The results from both Smith
et al. (2011) and Lefaucheur et al. (2011) suggest that selection for decreased RFI will result in
leaner carcasses and an undesirable change in some meat quality traits.
Physiological sources of differences in residual feed intake
In a study by Boddicker et al. (2011), pigs from each line were placed on either an ad libitum or a
weight stasis feeding regime. Although there was no difference in final body weight between lines
within treatment (Figure 2), the LRFI line consumed 9% less feed under the ad libitum treatment
and 20% less feed under the weight stasis treatment (Boddicker et al., 2011). The latter provides
indirect evidence that the LRFI line has lower maintenance requirements than the HRFI line.
This is supported by Barea et al. (2010), who estimated the maintenance energy requirements to
be 841 and 920 kJ × kg BW-0.60 × d-1 for LRFI and HRFI lines, respectively. They also found that
the LRFI line had 8% less heat production than the HRFI line.
Similar to Smith et al. (2011) who found that the LRFI line had a greater lean percentage,
Boddicker et al. (2011) found that the LRFI line had less fat than the HRFI line when evaluated
as a whole carcass percentage (Figure 3). There was no difference between lines within a treatment
for ash or protein content; however, the LRFI line had less fat and greater water carcass content
than the HRFI line under both treatments and a large proportion of the line differences in RFI
could be explained by differences in body composition (Boddicker et al., 2011).
Differences in digestibility have not yet been evaluated in the ISU RFI lines. However, Barea
et al. (2010) evaluated several measures of digestibility between the INRA RFI lines, including
HRFI LRFI
120
Ad lib
3.0 9%
Feed Intake (kg/d)
110
Body weight (kg)
Ad lib
2.5
100
2.0
90 1.5
Weight stasis Weight stasis
80 1.0
20%
70 0.5
0 1 2 3 4 5 6 1 2 3 4 5 6
Week Week
Figure 2. Body weights and feed intake over a 6 week period under ad libitum and weight stasis treatments
(Boddicker et al., 2011).

100%
90% 17 17 19 19
80% Ash
70% 25 17 19
30
60% Protein
50% Fat**
40%
30% 60 59 Water**
55 51
20%
10%
0%
LRFI HRFI LRFI HRFI
Ad libitum Weight stasis
Figure 3. Carcass composition under ad libitum and weight stasis treatments (Boddicker et al., 2011). ** =
significant at P<0.05.
the digestibility coefficients of dry matter, organic matter, nitrogen, energy, and phosphorous.
Digestibility coefficients did not differ between the two lines (P>0.44; Barea et al., 2010). Nitrogen
absorbed and retained was greater in the HRFI line than in the LRFI line but this was due to an
increase in nitrogen intake, rather than a difference in digestibility (Barea et al., 2010).
Using gilts from generation 8, N.K. Gabler (unpublished data), did show that LRFI pigs had less
intestinal inflammation than HRFI pigs, as evidenced by LRFI pigs having less haploglobin and
less ileum myeloperoxidase activity (N.K. Gabler, unpublished data). Barrier function did not
appear to differ between lines, with lines having similar levels of FITC-Dextran (N.K. Gabler,
unpublished data). Low RFI pigs did have more robust detoxification and epithelial defense
mechanisms in the intestine but not in the liver, based on increased lysozyme activity within the
ileum in the LRFI line over the HRFI line and no difference in lysozyme activity between the lines
within the liver (N.K. Gabler, unpublished data).
Differences in behavior
Differences in behavior can also attribute to differences in RFI. Young et al. (2011) showed that
differences in feeding behavior exist between lines while Sadler et al. (2011) demonstrated line
differences in pen behavior. Using pigs from generations 4 and 5 of the selection experiment,
Young et al. (2011) showed that pigs from the LRFI line consumed less feed than pigs from the
HRFI line (Young et al., 2011), as expected and shown in Figure 1. After adjusting for these
differences in intake, LRFI pigs were shown to eat faster, spend less time in the feeder, and visit
the feeder fewer times per day than HRFI pigs (Young et al., 2011). Feed intake per 2 h block
did not differ between lines after adjusting for daily feed intake, indicating that the timing of
feed consumption during the day does not differ between the two lines (Young et al., 2011).
However, LRFI pigs spent less time in the feeder during each 2 h block, after adjusting for daily
feed intake (Young et al., 2011). These results are similar to those found in the INRA lines, where
the LRFI line consumed feed faster and had fewer meals per day than the HRFI line (Dekkers and

Gilbert, 2010). Sadler et al. (2011) evaluated pen behaviors of walking, standing, sitting, lying,
and drinking in gilts of generation 5. Behavior assessments did not include the time pigs spent
in the FIRE feeders. Pigs from the LRFI line spent less time standing and more time sitting as a
percentage of the day than the HRFI line (Sadler et al., 2011). They also tended to spend more
time lying than the HRFI line which resulted in the LRFI line being less active than the HRFI line
(Sadler et al., 2011). Thus, it appears that selection for RFI has affected behavior with LRFI pigs
eating faster, spending less time in the feeder, and being less active than pigs from the HRFI line.
Impact of selection for grower/finishing residual feed intake on reproduction
Selection for decreased RFI during the grower/finishing phase of production has resulted in pigs
that consume less feed during that same period. It is important to evaluate whether or not this
decreased feed consumption continues through lactation and the possible effects on reproduction
of selection for decreased RFI. Young et al. (2010) showed that selection for decreased RFI during
the growing period appears to have had no detrimental effect on sow reproductive performance
and, in fact, has resulted in larger litter sizes, piglet birth weights, and pre-weaning growth.
However, this greater piglet performance was made possible by a greater loss of body condition
for sows from the LRFI line (Young et al., 2010). Sows from the LRFI line consumed less feed
during lactation than sows from the HRFI line starting in generation 6, with no difference in sow
feed intake during lactation between lines prior to generation 6 (Young et al., 2010). As a result of
having increased pre-weaning growth of piglets and decreased feed intake, sows from the LRFI
line were more efficient at converting energy from feed intake and body tissue mobilization into
piglet growth (Young et al., 2010). At INRA, LRFI sows from the INRA lines also had greater litter
size and litter birth weights than HRFI sows (Gilbert et al., 2011). Sows from the INRA LRFI line
also consumed less feed, had a lower residual feed intake, and lost more weight and backfat depth
during lactation than sows from the HRFI line (Gilbert et al., 2011). No differences (P>0.10) were
observed between the ISU LRFI and HRFI lines in milk fat, protein, lactose and gross energy
contents and estimated milk production (Rakhshandeh et al., 2012). However, milk saturated
fatty acid content was lower in the LRFI line compared to the HRFI line, while unsaturated fatty
acid content tended to be higher in the LRFI line compared to the HRFI line (Rakhshandeh et al.,
2012). Relative to HRFI sows, LRFI sows’ milk had higher conjugated linoleic acid contents while
alpha linolenic acid content of the milk did not differ between lines (Rakhshandeh et al., 2012).
Impact of selection for residual feed intake on response to stress and disease
The RFI selection experiment was conducted under high health conditions, typical of a nucleus
breeding farm. Thus, an important question is how pigs selected for RFI under high health
conditions perform in the field. To evaluate the impact of RFI selection on the ability to resist
disease, two groups of pigs consisting of ~50 pigs per line per group were sent to Kansas State
University and challenged with the PRRS virus, using procedures described by Boddicker et al.
(2012). Group 1 was euthanized at 40 d post-inoculation and Group 2 at 28 d post-inoculation.
Mortality did not differ between the lines across groups. There was also no difference between
lines in growth performance under the PRRS challenge with the LRFI line growing slightly faster.
In conclusion, selection for RFI does not appear to have affected the response to a PRRS challenge.
Current research is evaluating the differences and similarities between lines for response to

ACTH challenge and E. coli challenge in order to get a better understanding of line differences
in response to disease.
Indirect measures of residual feed intake
Evaluating RFI, or any measure of feed efficiency, requires individual feed intake recording
which is expensive. Thus, indirect measures of RFI would be very beneficial, in particular if
they can be measured at an early age. Bunter et al. (2005) showed that serum concentration
of juvenile insulin-like growth factor-I (IGF-I) was genetically correlated with economically
important performance traits, including feed efficiency. To evaluate the impact of selection on
RFI on juvenile IGF-I, all piglets from both lines were bled between 35 and 42 d of age in multiple
generations. As shown in Figure 4, selection for decreased RFI also resulted in decreased juvenile
IGF-I serum concentrations (Bunter et al., 2010). Juvenile serum IGF-I concentration was found
to have significant genetic correlations with RFI (0.63±0.15), feed conversion ratio (0.78±0.14),
loin muscle area (-0.35±0.12), and backfat depth (0.52±0.11) (Bunter et al., 2010).
Genetic markers associated with feed efficiency would also be useful for early selection for
feed efficiency. The molecular basis of RFI is not fully understood. A missense mutation in the
melanocortin-4 receptor (MC4R) gene was found to be associated with feed intake, growth, and
feed efficiency (Kim et al., 2000). In a genome-wide association study using the ISU RFI lines,
RFI appears to be very polygenic but there are some regions of the genome which show strong
associations with RFI (Gorbach et al., 2010). Lkhagvadorj et al. (2010) evaluated gene expression
1.5
1.0
Response (genetic sd)
0.5
0.0
IGF-I Control
-0.5
-1.0
IGF-I low RFI
-1.5
-2.0
-2.5
0 1 2 3 4 5
Generation
Figure 4. Correlated response in juvenile IGF-I serum concentration in genetic standard deviations (Bunter
et al., 2010).

differences in adipose tissue between the LRFI and HRFI lines and found that genes related to
carbohydrate metabolic process, regulation of gene expression, potassium ion transport, response
to stress, gene expression, and cellular carbohydrate metabolic process were up-regulated in the
adipose tissue of LRFI pigs. Genes that were down-regulated in the adipose tissue of LRFI pigs
were related to metabolic, homeostatic, developmental processes, respiratory chain complex
IV assembly, protein targeting, ion transport, generation of precursor metabolites and energy,
endocytosis, membrane invagination, DNA repair, membrane organization and biogenesis, and
centrosome cycle (Lkhagvadorj et al., 2010).
Conclusions
In summary, RFI is moderately heritable and responds well to selection. Selection for decreased
RFI has resulted in pigs that eat less but are leaner and grow slower, changes in body composition
(particularly decreased fat content), and no impact on meat quality. Selection for decreased RFI
has resulted in behavioral changes with low RFI pigs eating faster and spending more of their time
in inactive postures. Low RFI pigs appear to have reduced maintenance requirements and tissue
turnover rates than high RFI pigs. Low RFI pigs also have improved gut integrity compared to high
RFI pigs. No detrimental effects of selection for decreased RFI on litter size, litter performance,
or response to PRRS challenge were found. Juvenile IGF-I serum concentration has the potential
to be a selection tool for RFI.
Acknowledgements
The authors would like to acknowledge everyone who has contributed to the RFI selection
experiment at Iowa State University including faculty (Tom Baas, Rohan Fernando, Dorian
Garrick, Max Rothschild, Chris Tuggle, Elisabeth Lonergan, Steven Lonergan, Nick Gabler, John
Patience, Mike Spurlock, Lloyd Anderson, Anna Johnson, Peng Liu, Dan Nettleton, and Vasant
Honavar), graduate students (David Casey, Weiguo Cai, Nick Boddicker, Dinesh Thekkoot,
Emily Waide, Andrew Hess, Danielle Gorbach, Oliver Couture, Rachel Smith, Kyle Grubbs,
Shannon Cruzen, Emily Arkfeld, Venkatesh Mani, Amanda Harris, Jessica Jenkins, Sender
Lkhagvadorj, Larry Saddler, and Long Qu), post-docs and research associates (Suneel Onteru,
Anoosh Rakhshandeh, and Ed Steadham), and the staff at the Lauren Christian Swine Breeding
Research Center. We would also like to acknowledge our collaborators: Bob Rowland’s research
group at Kansas State University; Joan Lunney, Tom Weber, and Brian Kerr from USDA-ARS;
Kim Bunter and Frank Dunshea from Australia; and Rob Bergsma and Egbert Knol from IPG
& Wageningen University. The authors would like to thank PIC/Genus and Newsham Choice
Genetics for donating FIRE© feeders. We would also like to thank funding from USDA-CSREES
NRI Grants #2010-65206-20670 and #2011-68004-30336, National Pork Producers, Iowa Pork
Producers Association, ISU Center for Integrated Animal Genomics, Iowa State and Hatch
Funds, Pfizer Animal Health, and USDA Swine Genome Coordinator.

References
Adam, I., B. A. Young, A. M. Nicol, and A. A. Degan. 1984. Energy cost of eating in cattle given diets of
different form. Anim. Prod. 38:53-56.
Arthur, P. F., J. A. Archer, D. J. Johnston, R. M. Herd, E. C. Richardson, and P. F. Parnell. 2001. Genetic and
phenotypic variance and covariance components for feed intake, feed efficiency, and other postweaning
traits in Angus cattle. J. Anim. Sci. 79:2805-2811.
Baker, S. D., J. I. Szasz, T. A. Klein, P. S. Kuber, C. W. Hunt, J. B. Glaze, Jr., D. Falk, R. Richard, J. C. Miller, R.
A. Battaglia, and R. A. Hill. 2006. Residual feed intake of purebred Angus steers: Effects on meat quality
and palatability. J. Anim. Sci. 84:938-945.
Barea, R., S. Dubois, H. Gilbert, P. Sellier, J. van Milgen, and J. Noblet. 2010. Energy utilization in pigs
selected for high and low residual feed intake. J. Anim. Sci. 88:2062-2072.
Boddicker, N., N. K. Gabler, M. E. Spurlock, D. Nettleton, and J. C. M. Dekkers. 2011. Effects of ad libitum
and restricted feed intake on growth performance and body composition of Yorkshire pigs selected for
reduced residual feed intake. J. Anim. Sci. 89:40-51.
Boddicker, N. J., D. J. Garrick, J. M. Reecy, R. R. R. Rowland, J. K. Lunney, and J. C. M. Dekkers. 2012.
Genetic architecture of response to experimental porcine reproductive and respiratory syndrome virus
infection. Midwest American Society of Animal Science-American Dairy Science Association Joint
Meetings, Abstract #58.
Braastad, B. O., and J. Katle. 1989. Behavioural differences between laying hen populations selected for high
and low efficiency of food utilisation. Br. Poult. Sci. 30:533-544.
Bunter, K. L., S. Hermesch, B. G. Luxford, H.-U. Graser, and R. E. Crump. 2005. Insulin-like growth factor-I
in juvenile pigs is genetically correlated with economically important performance traits. Aust. J. Exp.
Agric. 45:783-792.
Bunter, K. L., W. Cai, D. J. Johnston, and J. C. M. Dekkers. 2010. Selection to reduce residual feed intake
in pigs produces a correlated response in juvenile insulin-like growth factor-I concentration. J. Anim.
Sci. 88:1973-1981.
Cai, W., D. S. Casey, and J. C. M. Dekkers. 2008. Selection response and genetic parameters for residual feed
intake in Yorkshire swine. J. Anim. Sci. 86:287-298.
Cai, W., M. S. Kaiser, and J. C. M. Dekkers. 2009. Genetic analysis of longitudinal measurements of
performance traits in selection lines for residual feed intake in Yorkshire swine. J. Anim. Sci. 89:1270-
1280.
Casey, D. S., H. S. Stern, and J. C. M. Dekkers. 2005. Identifying errors and factors associated with errors in
data from electronic swine feeders. J. Anim. Sci. 83:969-982.
Dekkers, J. C. M., and H. Gilbert. 2010. Genetic and biological aspect of residual feed intake in pigs. Proc.
9th WCGALP. Paper # 287.
De Haer, L. C. M., P. Luiting, and H. L. M. Aarts. 1993. Relations among individual (residual) feed intake,
growth performance and feed intake pattern of growing pigs in group housing. Livest. Prod. Sci. 36:233-
253.
Durunna, O. N., Z. Wang, J. A. Basarab, E. K. Okine, and S. S. Moore. 2011. Phenotypic and genetic
relationships among feeding behavior traits, feed intake, and residual feed intake in steers fed grower
and finisher diets. J. Anim. Sci. 89:3401-3409.
Gilbert, H., J.-P. Bidanel, Y. Billon, H. Lagant, P. Guillouet, P. Sellier, J. Noblet, and S. Hermesch. 2011.
Correlated responses in sow appetite, residual feed intake, body composition and reproduction after
divergent selection for residual feed intake in the growing pig. J. Anim. Sci. Online publication.

Gorbach, D., W. Cai, J. Dekkers, J. Young, D. Garrick, R. Fernando, and M. Rothschild. 2010. Large-scale
SNP association analyses of residual feed intake and its component traits in pigs. Proc. 9th WCGALP.
Paper # 265.
Herd, R. M., and S. C. Bishop. 2000. Genetic variation in residual feed intake and its association with other
production traits in British Hereford cattle. Livest. Prod. Sci. 63:111-119.
Herd, R. M., and P. F. Arthur. 2009. Physiological basis for residual feed intake. J. Anim. Sci. 87:E64-E71.
Hoque, M. A., H. Kadowaki, T. Shibata, T. Oikawa, and K. Suzuki. 2007. Genetic parameters for measures
of the efficiency of gain of boars and the genetic relationships with its component traits in Duroc pigs.
J. Anim. Sci. 85:1873-1879.
Jensen, J., I. L. Mao, B. B. Andersen, and P. Madsen. 1992. Phenotypic and genetic relationships between
residual energy intake and growth, feed intake, and carcass trait of young bulls. J. Anim. Sci. 70:386-395.
Kennedy, B. W., J. H. van der Werf, and T. H. Meuwissen. 1993. Genetic and statistical properties of residual
feed intake. J. Anim. Sci. 71:3239-3250.
Kim, K. S., N. Larsen, T. Short, G. Plastow, and M. F. Rothschild. 2000. A missense variant of the porcine
melanocortin-4 receptor (MC4R) gene is associated with fatness, growth, and feed intake traits.
Mammal. Genome 11:131-135.
Koch, R. M., L. A. Swiger, D. Chambers, and K. E. Gregory. 1963. Efficiency of feed use in beef cattle. J.
Anim. Sci. 22:486-494.
Kolath, W. H., M. S. Kerley, J. W. Golden, and D. H. Kiesler. 2006. The relationship between mitochondrial
function and residual feed intake in Angus steers. J. Anim. Sci. 84:861-865.
Korver, S., E. A. M. van Eekelen, H. Vos, G. J. Nieuwhof, and J. A. M. van Arendonk. 1991. Genetic parameters
for feed intake and feed efficiency in growing dairy heifers. Livest. Prod. Sci. 29:49-59.
Lancaster, P. A., G. E. Carstens, D. H. Crews, Jr., T. H. Welsh, Jr., T. D. A. Forbes, D. W. Forrest, L. O. Tedeschi,
R. D. Randel, and F. M. Rouquette. 2009. Phenotypic and genetic relationships of residual feed intake
with performance and ultrasound carcass traits in Brangus heifers. J. Anim. Sci. 87:3887-3896.
Lefaucheur, L., B. Lebret, P. Ecolan, I. Louveau, M. Damon, A. Prunier, Y. Billon, P. Sellier, and H. Gilbert.
2011. Muscle characteristics and meat quality traits are affected by divergent selection on residual feed
intake in pigs. J. Anim. Sci. 89:996-1010.
Lkhagvadorj, S., L. Qu, W. Cai, O. P. Couture, C. R. Barb, G. J. Hausman, D. Nettleton, L. L. Anderson, J. C.
M. Dekkers, and C. K. Tuggle. 2010. Gene expression profiling of the short-term adaptive response to
acute caloric restriction in liver and adipose tissues of pigs differing in feed efficiency. Am. J. Physiol.
Regul. Integr. Comp. Physiol. 298:R494-R507.
Moore, K. L., D. J. Johnston, H. U. Graser, and R. M. Herd. 2005. Genetic and phenotypic relationships
between insulin-like growth factor-I (IGF-I) and net feed intake, fat and growth traits in Angus beef
cattle. Aust. J. Exp. Agric. 56:211-218.
Morisson, M., A. Bordas, J. M. Petit, C. Jayat-Vignoles, R. Julien, and F. Minvielle. 1997. Associated effects
of divergent selection for residual feed consumption on reproduction, sperm characteristics, and
mitochondria of spermatozoa. Poult. Sci. 76:425-431.
Ngwerume, F., and I. L. Mao. 1992. Estimation of residual energy intake for lactating cows using an animal
model. J. Dairy Sci. 75:2283-2287.
Nkrumah, J. D., D. H. Crews Jr., J. A. Basarab, M. A. Price, E. K. Okine, Z. Wang, C. Li, and S. S. Moore.
2007. Genetic and phenotypic relationships of feeding behavior and temperament with performance,
feed efficiency, ultrasound, and carcass merit of beef cattle. J. Anim. Sci. 85:2382-2390.

Nkrumah, J. D., E. K. Okine, G. W. Mathison, K. Schmid, C. Li, J. A. Basarab, M. A. Price, Z. Wang, and S.
S. Moore. 2006. Relationships of feedlot efficiency, performance, and feeding behavior with metabolic
rate, methane production, and energy partitioning in beef cattle. J. Anim. Sci. 84:145-153.
Rakhshandeh, A., B. M. Adamic, J. M. Young, D. M. Thekkoot, T. E. Weber, M. A. McGuire, J. C. M.
Dekkers, and N. K. Gabler. 2012. Effect of selection for residual feed intake on sow performance II.
Milk composition. Midwest American Society of Animal Science-American Dairy Science Association
Joint Meetings, Abstract #44.
Richardson, E. C., R. M. Herd, P. F. Arthur, J. Wright, G. Xu, K. Dibley, and V. H. Oddy. 1996. Possible
physiological indicators for net feed conversion efficiency in beef cattle. Proc. Aust. Soc. Anim. Prod.
21:103-106.
Richardson, E. C., R. J. Kilgour, J. A. Archer, and R. M. Herd. 1999. Pedometers measure differences in
activity in bulls selected for high or low net feed efficiency. Proc Aust. Soc. Study Anim. Behav. 26:16
(Abstr.).
Richardson, E. C., R. M. Herd, V. H. Oddy, J. M. Thompson, J. A. Archer, and P. F. Arthur. 2001. Body
composition and implications for heat production of Angus steer progeny of parents selected for and
against residual feed intake. Aust. J. Exp. Agric. 41:1065-1072.
Richardson, E. C., and R. M. Herd. 2004. Biological basis for variation in residual feed intake in beef cattle.
2. Synthesis of results following divergent selection. Aust. J. Exp. Agric. 44:431-440.
Richardson, E. C., R. M. Herd, J. A. Archer, and P. F. Arthur. 2004. Metabolic differences in Angus steers
divergently selected for residual feed intake. Aust. J. Exp. Agric. 44:441-452.
Robinson, D. L., and V. H. Oddy. 2004. Genetic parameters for feed efficiency, fatness, muscle area and
feeding behaviour of feedlot finished beef cattle. Livest. Prod. Sci. 90:255-270.
Sadler, L. J., A. K. Johnson, S. M. Lonergan, D. Nettleton, and J. C. M. Dekkers. 2011. The effect of selection
for residual feed intake on general behavioral activity and the occurrence of lesions in Yorkshire gilts.
J. Anim. Sci. 89:258-266.
Schenkel, F. S., S. P. Miller, and J. W. Wilton. 2004. Genetic parameters and breed differences for feed
efficiency, growth and body composition traits of young beef bulls. Can. J. Anim. Sci. 84:177-185.
Shaffer, K. S., P. Turk, W. R. Wagner, and E. E. D. Felton. 2011. Residual feed intake, body composition, and
fertility in yearling beef heifers. J. Anim. Sci. 89:1028-1034.
Smith, R. M., N. K. Gabler, J. M. Young, W. Cai, N. J. Boddicker, M. J. Anderson, E. Huff-Lonergan, J. C. M.
Dekkers, and S. M. Lonergan. 2011. Effect of selection for decreased residual feed intake on composition
and quality of fresh pork. J. Anim. Sci. 89:192-200.
Van Arendonk, J. A. M., G. J. Nieuwhof, H. Vos, and S. Korver. 1991. Genetic aspects of feed intake and feed
efficiency in lactating dairy heifers. Livest. Prod. Sci. 29:263-275.
Webster, A. J. F., P. O. Osuji, F. White, and J. F. Ingram. 1975. The influence of food intake on portal blood
flow and heat production in the digestive tract of the sheep. Br. J. Nutr. 34:125-139.
Young, J. M., R. Bergsma, E. F. Knol, J. F. Patience, and J. C. M. Dekkers. 2010. Effect of selection for residual
feed intake on sow reproductive performance and lactation efficiency. Proc. 9th WCGALP. Paper #223.
Young, J. M., W. Cai, and J. C. M. Dekkers. 2011. Effect of selection for residual feed intake on feeding
behavior and daily feed intake patterns in Yorkshire swine. J. Anim. Sci. 89:639-647.

8. Pig breeding for improved feed efficiency
PIC International Group, Ratsteich 31, 24837 Schleswig, Germany; pieter.knap@genusplc.com
Abstract
The feed efficiency of growing pigs has been a matter of serious commercial and scientific interest
since at least 1970, but early recording technology made it difficult to produce accurate feed
intake data at the individual level. Since electronic feeders were introduced, the pig breeding
industry has been making good genetic improvement in feed conversion ratio (FCR) but this has
been mainly due to genetic improvement of growth and body composition traits. More than one
third of the variation in feed intake is due to processes that are independent of growth and body
composition, mainly body maintenance processes such as basal metabolism, protein turnover,
thermoregulation, physical activity, immune and other coping functions, nutrient digestion and
absorption efficiency. We give an example of how genetic variation in basal metabolism may be
generated by electron leakage through the mitochondrial membrane. This considerable (and up
to now insufficiently exploited) variation can be accessed through the trait residual feed intake
(RFI: feed intake, statistically adjusted for growth and body composition). In routine breeding
value estimation systems, this is catered for by including feed intake (rather than FCR) in the
breeding goal and in the multi-trait BLUP evaluation. We give examples of how selection for
growth and body composition traits and RFI leads to genetic change in feed intake and from
there in FCR, in four real-life breeding populations, and show that genetic improvement of FCR
is a function of genetic improvement of those underlying traits. Improving the efficiency of any
system often leads to a higher sensitivity to extraneous challenges; this also holds for the growing
pig. An important element of a breeding program that focuses on genetic improvement of feed
efficiency is therefore the proper monitoring and control of side effects in other traits, most
notably robustness and quality traits. And because many of the body maintenance processes are
strongly influenced by the production environment, the data used for breeding value estimation
of RFI should be recorded in commercial conditions.
Feed efficiency: past developments
More than thirty years before this book was published, Sellers (1981) argued that ‘feed efficiency
is a complex trait with large economic impact for the pork producer. One method for the pork
producer to increase feed efficiency in his operation is through the seedstock he purchases. As
seedstock suppliers, we must strive to better understand feed efficiency and through intense
selection programs, make improvement available’.
The current chapter describes how this has been happening in commercial pig breeding since
that time. Because feed efficiency is functionally linked to feed intake, see also the book chapter
‘Voluntary feed intake and pig breeding’ (Knap, 2009a).
Hal Sellers’s conference paper was preceded by several academic studies. A landmark one is by
Bernard and Fahmy (1970), who selected three Canadian Yorkshire lines during ten generations

for (1) a low feed conversion ratio (FCR: kg feed per kg body weight gain); (2) a high ‘carcass
score’ (CS: a proxy for leanness based on carcass length, backfat depth, loin muscle area, and belly
grade); or (3) a combination of these two traits. The low FCR line and the high CS line showed
the same selection response in FCR. These results were widely interpreted as evidence that FCR
in pigs can be genetically improved just as easily indirectly through selection for leanness (which
is cheap) as through direct selection (which is expensive), and selection strategies all over the
western world have been dominated by this notion for almost two decades.
The abovementioned selection responses are consistent with the realized selection intensities
and genetic parameters estimated in these lines (heritabilities at 0.11 for FCR and 0.35 for CS,
with a genetic correlation at -0.52). These heritability estimates are very low, particularly the
one for FCR; the breeding values of the young males that were actually selected were estimated
from records on four full sibs (mostly castrated males) that were housed together in another pen
than the selection candidate. For CS these were individual records, but for FCR, this was the pen
average. This makes for a poorly informative data structure. The true heritability of FCR works
out as:
1 + (n – 1) e2
h2 =
n r2G 1 – n
+
4 r2 2
where n is group size (4 in this case), rG is the genetic correlation between the trait as measured on castrates versus entire
males, e2 is the common pen environmental component, and r2 is the reported observed heritability (0.11 in this case).
Conservatively assuming rG=0.95, and using e2=0.42 from Bergsma et al. (2008), this gives
h2=0.34. In other words, two thirds of the potential direct selection response in FCR was lost
(0.11 versus 0.34) due to inadequate recording of that trait, and this notion puts the direct versus
indirect selection strategies in a very different perspective.
This 0.34 heritability value is much more consistent with the subsequent literature than the 0.11
value; see Figure 1, which also shows that the -0.52 genetic correlation between CS and FCR from
this experiment is uncomfortably large.
Later on, large-scale recording of individual feed intake (and FCR) in group-housed pigs became
feasible with the electronic feeding equipment that was widely installed by breeding organizations
since the early 1990s. In combination with improved data processing through BLUP, this has
led to substantial industry-wide genetic improvement in feed efficiency, particularly after
implementation of error detection and data clean-up routines such as developed by Eissen et al.
(1998, 1999) and Casey et al. (2005).
The power of such routines can be illustrated by the fact that David Casey’s algorithms (Casey
and Wang, 2006) have lifted the heritability of FCR as recorded in four PIC lines to above 0.5
(the highest entries in Figure 1a). At the same time, the rG with backfat depth in those four lines
is now close to zero (the lowest realistic entries in Figure 1b) which makes multi-trait breeding
value estimation much easier.

a b
0.1 0.2 0.3 0.4 0.5 0.6 0.0 0.2 0.4 0.6 0.8
h of FCR
2
rG of FCR with BFD
Figure 1. Frequency distributions of literature values for the heritability of feed conversion ratio (FCR) in
grower-finisher pigs (a; n=43), and for its genetic correlation with backfat depth (BFD) (b; n=32).
The abovementioned ‘substantial industry-wide genetic improvement in feed efficiency’ is

illustrated in Figure 2. These are results of public Commercial Product Evaluation trials – each
data point in the graph represents the mean FCR of a group of grower-finisher pigs from a
particular commercially available terminal cross. These data points are disconnected over time;
most of these crosses appear in such a test only once, because the breeding companies that produce
them often enter a different cross next time they participate, or they may participate only once.
In addition, we have not adjusted these data for any possible differences in diet composition,
housing system, end weight, or gender. In spite of these shortcomings, the 35-year time trend is
very clear: average commercial FCR has come down from 3.3 to 2.6, with a quite considerable
bandwidth among products which does not show any signs of narrowing over time. See Figure 2
in Knap (2009a) for the associated trends of feed intake and lean tissue growth rate.
Feed efficiency: biological backgrounds
An easy way to improve gross feed efficiency of growing animals is to increase their leanness.
As described in the previous section, this indirect selection strategy has played an important
role in pig breeding for decades. But a considerable part of the variation in feed intake (and
feed efficiency) is independent of body growth and body composition and can therefore not
be exploited that way. For maximum effect, we need a combination of the two elements. In
Chapter 7 in this book, Young and Dekkers (2012) introduce this production-independent part
as residual feed intake (RFI) and suggest that it covers roughly one third of the total variation
in feed intake. From simulation results, Knap (2009b) derived an even higher proportion, up to
half. Most of RFI is due to the energy requirements for body maintenance, comprising processes
such as basal metabolic rate, thermoregulation, protein turnover, physical activity, immunity, and
other coping functions, and processes that govern the cost of nutrient digestion and absorption.
From a production economy point of view, these are overhead functions; from the production
biology point of view, they determine net feed efficiency. All these processes show variation

3.6
3.4
3.2
3.0
2.8
2.6
2.4
2.2
75
80
85
90
95
00
05
10
19
19
19
19
19
20
20
20
Figure 2. Time trends of feed conversion ratio in grower-finisher pigs of 103 terminal crosses, as recorded in
public Commercial Product Evaluation trials in Denmark, France, Germany, the Netherlands, UK and USA.
Unadjusted phenotypic population means, data from 18 literature and internet sources. The trendline is a
spline interpolation plot through the data, with its 95% confidence limits.
among animals, and a considerable part of that variation is of a genetic nature. We give here one
example of how such a basic function can cause genetic variation in feed efficiency.
Mitochondrial respiratory efficiency
Mitochondria are animal cell organelles; they are small (about 0.005 mm long, on average) but
there are many of them (roughly 500 in every body cell) so that they make up about 20% of
mammalian body weight. They drive about 90% of oxygen consumption. Mitochondria contain a
large area of tightly packed and folded membrane that separates a mitochondrial matrix from an
intermembrane compartment. Glycolysis, the Krebs cycle and related processes turn glucose and
fatty acids into malate, pyruvate, and succinate, and in the matrix these compounds pass electrons
to a series of four membrane-embedded enzymes that transport them within the membrane back
into the matrix where the fourth enzyme of the series links them to oxygen. Oxygen acts as the
driving (or rather, pulling) force behind this electron transport chain, which generates a negative
current that must be accompanied by a positive one; this is the proton pump, transporting
protons through those same enzymes from the matrix into the intermembrane compartment
and back into the matrix through a fifth membrane-embedded enzyme known as ATP synthase.
The proton flow through this enzyme generates rotational energy that is used to couple adenosine

diphosphate with phosphorus to form adenosine triphosphate (ATP), the universal carrier of
energy in the body.
This mechanism, known as oxidative phosphorylation, is very much a part of basal metabolism,
but it is not a perfect system. Electrons may leak out of the electron transport chain into either
the matrix or the intermembrane compartment, where they may interact with oxygen to form
aggressive reactive oxygen species such as superoxide, peroxide, and hydroxyl radicals. When this
happens in the matrix, it uses up oxygen which goes at the expense of the abovementioned pulling
force so that ATP production is compromised – the first source of inefficiency. At both sides of the
membrane, the reactive oxygen species cause oxidation of proteins (leading to increased protein
turnover – the second source of inefficiency) and peroxidation of lipids, damaging the membrane
phospholipid layer which leads to even more electron leakage (a positive feedback loop). Leaked
electrons can be neutralized by active proton leakage, but those protons are then not available
to drive ATP synthase – the third source of inefficiency. Finally, the reactive oxygen species may
cause oxidation of DNA, which leads to mutations and therefore to genetic variation. The text
above derives from Lehninger et al. (1993), Dittrich et al. (2003), Bottje et al. (2009), Bottje and
Carstens (2009), and Jastroch et al. (2010), where much more detail can be found.
Bottje et al. (2002) and Ojano-Dirain et al. (2004) studied this system in broiler chickens that
had been high-low sampled for FCR, and reported significant correlations between FCR and
mitochondrial respiratory activity in red and white muscles (Figure 3), and between FCR and
mitochondrial hydrogen peroxide production. Such studies have been repeated in cattle (Kolath
et al., 2006), mice (McDonald et al., 2009), and pigs (Lefaucheur et al., 2011), with different
mitochondrial protein per minute
Respiratory activity units per mg
M. pectoralis superficialis,
150 r = –0.56
M.quadriceps femoris,
130 r = –0.68
110
90
70
1.1 1.2 1.3 1.4 1.5 1.6 1.7

Feed conversion ratio
Figure 3. Mitochondrial respiratory activity in red and white muscles of broiler chickens, in relation to the feed
conversion ratio of the bird. Data from Bottje et al. (2002). The horizontal variable in this high-low sampling
experiment has a range of 5 standard deviations. Within the low or high subgroups, this range is 1.2 standard
deviations, clearly not wide enough to allow for the detection of any meaningful relationship.

results that seem to suggest mainly that the experimental protocols of such studies need more fine
tuning – not surprising for an attempt to link gross feed efficiency to such nano-scale phenomena.
This is an example of a basic life-supporting system with self-enhancing flaws that cause metabolic
inefficiency and associated genetic variation.
Variation in maintenance requirements
Figure 4 summarizes literature estimates for the coefficient of variation and heritability of the
energy requirements for body maintenance (MEm) and of RFI in several mammalian species,
to illustrate that MEm varies considerably among animals, partly because of genetic factors. See
Knap (2009b) for more detail.
Figure 5 is a counterpart of the FCR trend plot of Figure 2, this time for RFI. Comparison of
these two graphs shows that the industry-wide reduction in FCR has been much stronger than in
RFI; it follows that the FCR trend is mainly a result of genetic improvement of growth and body
composition – entirely in line with the above reasoning.
So, here we have a neatly defined source of genetic variation in net feed efficiency, covering more
than a third of the total variation in gross feed efficiency, but largely unexploited by pig breeding
up to now. The obvious question then is how to get this variation under genetic control. Before
addressing this question, it is useful to illustrate the interdependencies of production traits, feed
intake, RFI, and FCR as they are evolving in a real-life pig breeding program.
0.35 1.0
0.30
0.8
Coefficient of variaton
0.25
0.6
Heritability
0.20
0.15 0.4
0.10
0.2
0.05
0.00 0.0
P M IC MC P M IC MC
Figure 4. Literature values (from 50 sources) for the phenotypic coefficient of variation (left) and the heritability
(right) of the energy requirements for body maintenance (open circles) and residual feed intake (solid circles)
in grower-finisher pigs (P), mice (M), and immature (IC) and mature (MC) cattle. Updated from Knap (2009b).

0.4
0.3
0.2
Residual feed intake (kg/day)
0.1
0.0
-0.1
-0.2
-0.3
-0.4
75
80
85
90
95
00
05
10
19
19
19
19
19
20
20
20
Figure 5. Time trends of residual feed intake in grower-finisher pigs. See Figure 2 for background and design
information.
Genetic change in production traits and feed efficiency
Figure 6 gives realized genetic trends from 2001 to mid-2011 for the production traits average
daily gain (ADG), backfat depth (BFD), loin muscle depth (LMD), and daily feed intake (DFI),
and for the feed efficiency traits feed conversion ratio (FCR) and residual feed intake (RFI) in
four PIC pig lines, ordered here from left to right by increasing genetic trend for FCR. During
this decade, the breeding goal of each line was revised a few times. This involves changes in the
relative focus on the traits shown here and also on a range of reproductive traits, meat quality
traits, and robustness traits. Because the trends in those other traits are not shown here, the
trends in Figure 6 do not always appear advantageous (or logical) – but that is not the purpose
of this Figure. Instead, recall that changes in FCR are by definition due to changes in DFI (due
to changes in ADG, BFD, LMD, and RFI) and in ADG. It is illustrative to inspect these trends in
that order, as below.
Because its breeding goal focuses on other traits, dam line D1 has changed relatively little in terms
of the traits shown here, apart from a steady increase in LMD. Up to late 2004, ADG, BFD, and
RFI decrease, leading to a reduction in DFI. Because ADG and DFI go down at the same rate, FCR
shows zero change. From there to 2007, ADG shows zero change whereas BFD and RFI increase,
leading to an increase in DFI. Because ADG does not change and DFI goes up, FCR goes up as

LMD BFD FCR DFI ADG RFI
3.2 D1 3.2
Genetic standard deviation units
D2 S1 S2
2.8 2.8
2.4 2.4
2.0 2.0
1.6 1.6
1.2 1.2
0.8 0.8
0.4 0.4
0.0 0.0
01
03
05
07
09
20 11
01
03
05
07
09
20 11
01
03
05
07
09
20 11
01
03
05
07
09
11
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Figure 6. Realized ten-year genetic trends of performance traits and feed efficiency traits in four PIC pig lines.
D1 and D2 are dam lines, S1 and S2 are sire lines. Horizontal axes: year of birth. All traits were standardized
by their genetic standard deviation to make the trends comparable across traits and across lines. The vertical
starting point of each trendplot is arbitrary. ADG: average daily gain; BFD: backfat depth; DFI: daily feed
intake; FCR: feed conversion ratio; LMD: loin muscle depth; FRI: residual feed intake. The DFI, FCR, and RFI
trends are based on independently-estimated breeding values. Note that the breeding goal of each of these
lines comprises many more traits than these six, and those trends are not shown here.
well. From 2007, ADG increases whereas BFD and RFI decrease, leading to zero change in DFI.
This causes FCR to go down.
The breeding goal of dam line D2 places much more focus on growth rate than the line D1
breeding goal. This is reflected in a steady increase in ADG. LMD shows relatively little change
throughout the whole period. Up to mid-2004, ADG increases slowly, BFD decreases and RFI
shows hardly any change, leading to a zero change in DFI. This causes FCR to go down. From there
to 2008, ADG increases strongly whereas BFD and RFI increase slowly, leading to an increase in
DFI that is just as strong as the increase in ADG. FCR shows a zero change as a result. From 2008,
ADG continues to increase, BFD shows a zero change and RFI decreases slightly. This leads to
a diminished upward trend in DFI. As a consequence, the reduction in FCR becomes stronger.
Sire line S1 shows the most dynamic breeding goal of these four lines, with slight revisions every
two to three years. LMD shows a steady increase throughout the whole period. Up to late 2003,
ADG increases strongly whereas BFD and RFI decrease. This leads to an increase in DFI, but
less so than in ADG. As a result, FCR shows a strong reduction. From there to mid-2005, ADG
shows zero change and BFD and RFI continue as before. This leads to a reduction in DFI, which
is mirrored in a reduction in FCR. From there to 2007, ADG increases somewhat again whereas
BFD and RFI show a zero change. This leads to a very slight increase in DFI and therefore a
hardly perceptible reduction of FCR. From 2007, ADG increases some more, BFD decreases

relatively strongly and RFI shows a zero change. This leads to a zero change in DFI and therefore
a reduction in FCR at the same rate as ADG goes up.
The breeding goal of sire line S2 places much focus on carcass composition traits. This is reflected
in steady changes in BFD and LMD.Up to mid-2004, ADG increases at the same fast rate as BFD
goes down and LMD goes up, and RFI is strongly reduced. This leads to a slight increase in DFI
and therefore a strong reduction of FCR. From there to early 2007, the increase in ADG is much
diminished although still positive, whereas BFD, LMD and RFI continue as before. The result is
a slight reduction of DFI. This leads to a diminished downward trend in FCR. From early 2007,
ADG does not change whereas BFD, LMD and RFI continue as before. This leads to reductions
in DFI and FCR at the same fast rate.
So far these genetic trend patterns. In summary, genetic changes in FCR can be fully explained
from genetic changes in its underlying traits. It follows that genetic improvement of gross feed
efficiency is a simple matter of judicious genetic improvement of its component traits. RFI is one
of these: gross feed efficiency comprises net feed efficiency plus the effects of size, growth rate and
body composition. The logical setup for achieving genetic improvement of gross feed efficiency is
then to calculate estimated breeding values (EBVs) for net feed efficiency (such as RFI), growth
(such as ADG) and body composition (such as BFD and LMD). These can then be combined
into a selection index that aims at the maximization of gross feed efficiency (FCR), as part of a
breeding goal that also caters for the required amounts of improvement of growth rate, carcass
composition, reproduction, meat quality and robustness traits.
The trendlines in Figure 6 represent the population average for each trait, but give no information
on the available variation among individual animals. This aspect is illustrated in Figure 7, where
the FCR trend plots for the two most extreme lines (D1 and S2) are repeated together with their
individual scatter. There are two interesting details here. First, the genetic improvement of FCR
in line S2 has caused a situation where individual levels that are just about average at the end of
the reporting period did not occur at all ten years earlier, and vice versa – see the white horizontal
reference lines in the graph. Second, in spite of this considerable change in genetic level in line
S2, its individual bandwidth hardly diminishes over time: balanced selection does not take away
genetic variation. Of course the latter aspect holds for line D1 too, but there it would be expected
because of the (deliberate) lack of genetic trend in FCR.
Breeding for improved feed efficiency
The feasible way to quantify net feed efficiency is through RFI; this requires recording of DFI plus
some statistical adjustment for production traits to obtain the residual term. For routine breeding
purposes, Luiting et al. (1992) and Kennedy et al. (1993) have shown that this adjustment is
equivalent to what happens in a multi-trait BLUP system for breeding value estimation, and that
a set of multi-trait EBVs for RFI and production traits holds the same information as such a set
for DFI and those same production traits. Selection for reduced RFI is therefore equivalent to
selection for reduced DFI while keeping the genetic change of the production traits at zero. Of
course, the latter aspect is impractical because those production traits are breeding goal traits in
their own right, but that does not affect the principle. The logical consequence is a breeding goal

3.2 3.2
D1 S2
2.4 2.4
Genetic standard deviation units
1.6 1.6
0.8 0.8
0.0 0.0
-0.8 -0.8
-1.6 -1.6
-2.4 -2.4
01
03
05
07
09
11
01
03
05
07
09
11
20
20
20
20
20
20
20
20
20
20
20
20
Figure 7. Time trends of estimated breeding values for feed conversion ratio of grower-finisher pigs of two
PIC lines (D1, S2). The white trendlines represent the genetic trends, the same as the more strongly smoothed
FCR plots in Figure 6.
comprising DFI and production traits (and reproduction, meat quality and robustness traits),
taking care of the statistical adjustments implicitly in multi-trait BLUP. FCR and RFI can still be
calculated phenotypically, and included in the routine BLUP evaluation to monitor their genetic
trends (that is how the FCR and RFI trendplots in Figure 6 were created) but there is no need to
include their EBVs in the selection index – that would even over-parameterize the system.
The EBV of DFI can be weighted into the selection index (relative to the EBVs of the other traits)
with its marginal economic value, which is closely related to the one for FCR. A selection index
is a linear combination of traits (of the form I = b1 × trait1 + b2 × trait2 + ...) and having FCR
(i.e. DFI/ADG) in such an index together with ADG has the disadvantage that the relationship
between these traits is by definition non-linear and therefore does not fit into a linear system. See
Gunsett (1984, 1987) for more detail. But the most important reason to focus on DFI (RFI) rather
than FCR in the breeding goal is that it brings every relevant element of the system under control,
because it separates FCR into its functional component traits so that these can be manipulated
independently of each other.

Side effects: correlated responses in other traits
Any system that is made more efficient has a strong tendency to become more sensitive as a side
effect, because there are fewer idle resources available for coping with challenges from outside.
The growing pig is no exception: with genetic improvement of feed efficiency, serious control
of unintended correlated responses becomes necessary. Two large-scale selection experiments
for RFI in pigs were initiated in the 2000s in France and the USA (see Dekkers and Gilbert,
2010), and these provide much useful information on such side effects. This includes correlated
responses of behavior traits (Gilbert et al., 2009; Meunier-Salaun et al., 2011; Sadler et al., 2011;
Young et al., 2011), muscle physiology and related meat quality traits (Fan et al., 2010; Faure et
al., 2011; Lefaucheur et al., 2011; Smith et al., 2011), and subsequent reproductive performance
(Young et al., 2010; Gilbert et al., 2012).
Control of such side effects requires (1) routine recording of the relevant data (in the relevant
production conditions); (2) routine estimation of breeding values from these data; (3) monitoring
genetic change; and (4) including those trait EBVs into the selection index to counter any detected
undesirable correlated responses. Examples of real-life cases are in Figure 8, where the EBVs for
growing pig mortality and loin muscle pH in two of the lines of Figure 6 are shown in relation
to the EBVs for RFI. Remedial action is clearly not required here, and that is very useful process
control information – quite apart from the fact that growing pig mortality and loin muscle pH
are established breeding goal traits in these lines.
Grower-finisher mortality
Loin muscle pH
Residual feed intake
Figure 8. Estimated breeding values (EBVs) for grower-finisher pig mortality (left) and loin muscle pH24 (right),
both in relation to the EBV for residual feed intake, in two PIC lines (S1, S2). All these EBVs reflect data recorded
on crossbred pigs in commercial conditions. The white trendlines are spline interpolation plots through the
data. Note that the correlation coefficients (r) are correlations between independently-calculated EBVs, not
genetic correlations; those would be larger, dependent on the EBV accuracies.

As always in a sensible breeding program, all those other traits must be recorded, monitored, and
kept under control. Overfocus on a single hype trait (e.g. litter size in the 1990s, mortality rates in
the early 2000s, feed efficiency right now, possibly boar taint or group behaviour in the late 2010s,
and who knows what else in 2020) carries the strong risk of letting the rest slip out of control.
Unprofessional breeding programs have consistently produced unbalanced non-sustainable
outcomes that have had to be remedied by emergency selection at a later stage.
Environment-dependent maintenance components
Any serious pig breeding program should define its breeding goals on the commercial level,
particularly with regard to body maintenance processes that are under strong environmental
influence: thermoregulation, immunity, coping, robustness, activity, nutrient digestion, and
absorption. Consequently, when net feed efficiency plays an important role in the breeding goal,
selection should be based on feed intake data recorded in commercial conditions. This calls for
designs such as RRS (reciprocal recurrent selection; Biswas et al., 1971) and CCPS (combined
crossbred and purebred selection; Wei and Van der Werf, 1994), where crossbred halfsibs of
purebred nucleus selection candidates are recorded in commercial conditions, and these data are
used for breeding value estimation at the nucleus level (see Casey et al., 2006; Newman et al., 2010).
The importance of such a setup for feed intake recording is illustrated in Figure 9. This shows
the EBVs for four traits as calculated from commercial crossbred records, in relation to the same
animals’ EBVs from purebred nucleus performance, for two of the lines of Figure 6. These EBV
correlations are reasonably high (around 0.8) for average daily gain and backfat depth, but feed
intake is clearly much more environmentally sensitive with correlations around 0.5. This can only
mean that RFI must show much lower correlations than that, which is indeed the case. It follows that
feed intake as it is recorded in nucleus conditions is not very useful for breeding value estimation of
net feed efficiency in a system that aims at producing commercially viable end products.
Implications
Coming back to our earlier quote from Sellers (1981) who argued that ‘one method for the pork
producer to increase feed efficiency in his operation is through the seedstock he purchases; as
seedstock suppliers we must strive to better understand feed efficiency and through intense
selection programs make improvement available’, we can now list the main action points for a
pork producer to optimize his herd’s genetic potential for feed efficiency, as follows:
1. Find a genetics supplier who can document, for their relevant lines, (1) consistent long-term
genetic improvement of feed efficiency; and (2) absence of unfavorable correlated responses
in fitness traits.
2. Check that their breeding goals center around feed intake, not around FCR (so that they are
more in control of possible side effects).
3. Make sure they supply animals with favorable multi-trait BLUP breeding value estimates for
feed intake in commercial conditions.
4. Influence their breeding goals: keep communicating all of the above to them.
5. Support the pigs of point 3 with good animal management – protein cannot be made out of
thin air, certainly not efficiently.

Crossbred commercialperformance
Purebred nucleus performance
Figure 9. Estimated breeding values (EBVs) for average daily gain (ADG), backfat depth (BFD), daily feed
intake (DFI), and residual feed intake (RFI) of grower-finisher pigs of two PIC lines (S1, S2). Horizontal axis:
EBVs reflect data recorded on purebred nucleus pigs. Vertical axis: EBVs reflect data recorded on crossbred
halfsibs of nucleus pigs, housed in commercial conditions. Note that the correlation coefficients (r) are
correlations between EBVs, not genetic correlations; those would be larger, dependent on the EBV accuracies.
References
Bergsma, R., E. Kanis, E. F. Knol, and P. Bijma. 2008. The contribution of social effects to heritable variation
in finishing traits of domestic pigs (Sus scrofa). Genetics 178:1559-1570.
Bernard, C., and M. H. Fahmy. 1970. Effect of selection on feed utilization and carcass score in swine. Can.
J. Anim. Sci. 50:575-584.
Biswas, D. K., A. B. Chapman, N. L. First, and H. L. Self. 1971. Intrapopulation versus reciprocal recurrent
selection in swine. J. Anim. Sci. 32:840-848.
Bottje, W., Z. X. Tang, M. Iqbal, D. Cawthon, R. Okimoto, T. Wing, and M. Cooper. 2002. Association of
mitochondrial function with feed efficiency within a single genetic line of male broilers. Poult. Sci.
81:546-555.
Bottje, W. G., and G. E. Carstens. 2009. Association of mitochondrial function and feed efficiency in poultry
and livestock species. J. Anim. Sci. 87:E48-E63.
Bottje, W., M. D. Brand, C. Ojano-Dirain, K. Lassiter, M. Toyomizu, and T. Wing. 2009. Mitochondrial
proton leak kinetics and relationship with feed efficiency within a single genetic line of male broilers.
Poult. Sci. 88:1683-1693.
Casey, D. S., H. S. Stern, and J. C. M. Dekkers. 2005. Identification of errors and factors associated errors in
data from electronic swine feeders. J. Anim. Sci. 83:969-982.
Casey, D. S., and L. Wang. 2006. Methods of editing errors in data from electronic swine feeders impact
heritability estimates of average daily feed intake. J. Anim. Sci. 82(Suppl. 1):120 (Abstr.).

Casey, D., M. Perez, D. McLaren, and T. Short. 2006. Crossbred breeding values: selecting for commercial
performance in pigs. Communication 06-26 in Proc. 8th World Congr. Genet. Appl. Livest. Prod., Belo
Horizonte, Brazil.
Dekkers, J. C. M., and H. Gilbert. 2010. Genetic and biological aspect of residual feed intake in pigs.
Communication 0287 in Proc. 9th World Congr. Genet. Appl. Livest. Prod., Leipzig, Germany.
Dittrich, M., S. Hayashi, and K. Schulten. 2003. On the mechanism of ATP hydrolysis in F1-ATPase.
Biophys. J. 85:2253-2266.
Eissen, J. J., E. Kanis, and J. W. M. Merks. 1998. Algorithms for identifying errors in individual feed intake
data of growing pigs in group housing. Appl. Eng. Agr. 14:667-673.
Eissen, J. J., A. G. de Haan, and E. Kanis. 1999. Effect of missing data on the estimate of average daily feed
intake of growing pigs. J. Anim. Sci. 77:1372-1378.
Fan, B., S. Lkhagvadorj, W. Cai, J. Young, R. M. Smith, J. C. M. Dekkers, E. Huff-Lonergan, S. M. Lonergan,
and M. F. Rothschild. 2010. Identification of genetic markers associated with residual feed intake and
meat quality traits in the pig. Meat Sci. 4:645-650.
Faure, J., L. Lefaucheur, N. Bonhomme, L. Brossard, H. Gilbert, and B. Lebret. 2011. Pork quality differences
between lines divergently selected for residual feed intake. Communication P012 in Proc. Int. Congr.
Meat Sci. Tech., Ghent, Belgium.
Gilbert, H., S. al Aïn, J. P. Bidanel, H. Lagant, Y. Billon, P. Guillouet, J. Noblet, and P. Sellier. 2009. Relations
génétiques entre efficacité alimentaire et cinétiques de croissance et d’ingestion chez le porc Large White.
Communication G01 in Proc. 41st Journées de la Recherche Porcine, Paris, France.
Gilbert, H., J. P. Bidanel, Y. Billon, H. Lagant, P. Guillouet, P. Sellier, J. Noblet, and S. Hermesch. 2012.
Correlated responses in sow appetite, residual feed intake, body composition and reproduction after
divergent selection for residual feed intake in the growing pig. J. Anim. Sci. 90:1097-1108.
Gunsett, F. C. 1984. Linear index selection to improve traits defined as ratios. J. Anim. Sci. 59:1185-1193.
Gunsett, F. C. 1987. Merit of utilizing the heritability of a ratio to predict the genetic change of a ratio. J.
Anim. Sci. 65:936-942.
Jastroch, M., A. S. Divakaruni, S. Mookerjee, J. R. Treberg, and M. D. Brand. 2010. Mitochondrial proton
and electron leaks. Essays Biochem. 47:53-67.
Kennedy, B. W., J. H. van der Werf, and T. H. Meuwissen. 1993. Genetic and statistical properties of residual
feed intake. J. Anim. Sci. 71:3239-3250.
Knap P. W. 2009a. Voluntary feed intake and pig breeding. Pages 11-33 in Voluntary feed intake in pigs.
D. Torrallardona and E. Roura, eds. Wageningen Academic Publishers, Wageningen, the Netherlands.
Knap P. W. 2009b. Allocation of resources to maintenance. Pages 110-129 in Resource allocation theory
applied to farm animal production. W.M. Rauw, ed. CAB International, Wallingford, UK.
Kolath, W. H., M. S. Kerley, J. W. Golden, and D. H. Keisler. 2006. The relationship between mitochondrial
function and residual feed intake in Angus steers. J. Anim. Sci. 84:861-865.
Lefaucheur, L., B. Lebret, P. Ecolan, I. Louveau, M. Damon, A. Prunier, Y. Billon, P. Sellier, and H. Gilbert.
2011. Muscle characteristics and meat quality traits are affected by divergent selection on residual feed
intake in pigs. J. Anim. Sci. 89:996-1010.
Lehninger, A. L., D. L. Nelson, and M. M. Cox. 1993. Principles of biochemistry. 2nd ed. Worth Publishers,
New York NY, USA.
Luiting, P., J. H. J. van der Werf, and T. H. E. Meuwissen. 1992. Proof of equivalence of selection indices
containing traits adjusted for each other. Page 146 in Proc. 43rd Ann. Meet. Eur. Assoc. Anim. Prod.,
Madrid, Spain.

McDonald, J. M., J. J. Ramsey, J. L. Miner, and M. K. Nielsen. 2009. Differences in mitochondrial efficiency
between lines of mice divergently selected for heat loss. J. Anim. Sci. 87:3105-3113.
Meunier-Salaün, M. C., C. Guérin, Y. Billon, A. Priet, P. Sellier, and H. Gilbert. 2011. Sélection divergente
sur la consommation moyenne journalière résiduelle chez le porc en croissance: caractéristiques
phénotypiques de l’activité physique et comportementale des porcs en fonction de la lignée et du sexe.
Communication BE02 in Proc. 43rd Journées de la Recherche Porcine, Paris, France.
Newman S., L. Wang, J. Anderson, and D. Casey. 2010. Utilizing crossbred records to increase accuracy of
breeding values in pigs. Communication 0632 in Proc. 9th World Congr. Genet. Appl. Livest. Prod.,
Leipzig, Germany.
Ojano-Dirain, C. P., M. Iqbal, D. Cawthon, S. Swonger, T. Wing, M. Cooper, and W. Bottje. 2004.
Determination of mitochondrial function and site-specific defects in electron transport in duodenal
mitochondria in broilers with low and high feed efficiency. Poult. Sci. 83:1394-1403.
Sadler, L. J., A. K. Johnson, S. M. Lonergan, D. Nettleton, and J. C. M. Dekkers. 2011. The effect of selection
for residual feed intake on general behavioral activity and the occurrence of lesions in Yorkshire gilts.
J. Anim. Sci. 89:258-266.
Sellers, H. I. 1981. Selection for feed efficiency. Pages 55-60 in Proc. Conf. Nat. Swine Impr. Fed., Des Moines
IA, USA.
Smith, R. M., N. K. Gabler, J. M. Young, W. Cai, N. J. Boddicker, M. J. Anderson, E. Huff-Lonergan, J. C.
Dekkers, and S. M. Lonergan. 2011. Effects of selection for decreased residual feed intake on composition
and quality of fresh pork. J. Anim. Sci. 89:192-200.
Wei, M., and J. H. J van der Werf. 1994. Maximizing genetic response in crossbreds using both purebred and
crossbred information. Anim. Prod. 59:401-413.
Young, J.M. and J.C.M. Dekkers, 2012. The genetic and biological basis of residual feed intake as a measure
of feed efficiency. Pages 153-166 in Feed efficiency in swine. J.F. Patience, ed. Wageningen Academic
Publishers, Wageningen, the Netherlands.
Young, J. M., R. Bergsma, E. F. Knol, J. F. Patience, and J. C. M. Dekkers. 2010. Effect of selection for residual
feed intake on sow reproductive performance and lactation efficiency. Communication 0223 in Proc.
9th World Congr. Genet. Appl. Livest. Prod., Leipzig, Germany.
Young J. M., W. Cai, and J. C. M. Dekkers. 2011. Effect of selection for residual feed intake on feeding
behavior and daily feed intake patterns in Yorkshire swine. J. Anim. Sci. 89:639-647.

9. Effect of climatic environment on feed efficiency in swine
D. Renaudeau1, H. Gilbert2, and J. Noblet3
1INRA, UR143, Animal Production Unit, F-97170 Petit Bourg, France
2INRA, UMR 444, Cell Genetics Laboratory, F-31326 Toulouse, France
3INRA, UMR 1348, Physiology, Environment and Genetics for the Animal and Livestock Systems,
F-35590 Rennes, France
Abstract
Since feed costs comprise about 60 to 70% of the total costs of production, efforts to reduce
feed costs is then a major preoccupation for increasing competitiveness of the pig industry.
Understanding the factors that affect feed conversion in pigs can enable producer to more
effectively combine various inputs in order to achieve a low feeding cost. There are many factors
involved in reaching good feed efficiency including genetic, diet, feed, management, housing and
environment. Temperature is the single most important environmental factor affecting the global
farm feed efficiency. In ad libitum fed animals, changes in metabolic heat production are essential
mechanisms to maintain body temperature within a physiologically safe range under cold or
heat stress. These adjustments have direct consequences on energy intake and/or maintenance
requirements which in turn could reduce energy efficiency as conversion of feed to tissue or
other products. However, the low level of performance related to a thermal challenge can also be
attributed to a direct effect of ambient temperature (independent of feed intake) on reproductive
physiology, health, and energy metabolism. Finally, thermal stress could also be a major cause
of pig mortality at birth, during the nursing period or thereafter, and could predispose pigs to
mortality or morbidity by other causes (starvation, diseases, etc.). High mortality rates in nursing
piglets or in finishing pigs have a significant impact on overall farm feed efficiency. The chapter
describes how climatic environment impacts feed efficiency in pigs and reviews solutions that can
be used in order to attenuate the effect of environmental temperature on feed efficiency.
Introduction
In pig production, 50 to 70% of the total production costs are due to feed expenses. The reduction
of feeding costs is then a major preoccupation for increasing competitiveness of the pig industry,
especially when cost of feedstuffs shows dramatic augmentations and (or) fluctuations. Thus, feed
efficiency is an important determinant of profitability for swine producers and small increments/
deteriorations in this trait can have a major economic impact. The overall feed efficiency of a
pig farm is a complex trait which combines the sow’s feed utilization to produce piglets and the
pig’s feed utilization to produce meat during the growing-finishing phase. As a consequence,
feed efficiency depends upon a lot of factors such as (1) nutritional factors (the quantity, the
composition, and the digestibility of feed), (2) animal factors (breed, sex, age, body weight), and
(3) environmental factors (management, health and climate). Climatic factors can be considered
as one of the main limiting environmental factors of feed efficiency both in reproductive sows and
in growing pigs. The increasing concerns on production losses due to thermal cold or heat stress
is justified not only for the coolest and the hottest periods of the year in temperate countries but

also in tropical areas where pig production has dramatically increased over the past 2 decades
(Renaudeau et al., 2008).
After some general considerations, this chapter will describe the effects of thermal stress on feed
efficiency and the main strategies available for reducing the impact of climatic factors on feed
efficiency.
General aspects
Feed efficiency
In farrow-to-finish farms, the overall feed efficiency is based on the total amount of purchased
feeds required for producing one kilogram of marketed carcass. This calculation must consider
the feed efficiency in the sow herd by measuring the sow feed per year or the sow feed per
kilogram of litter weight produced at weaning and the feed efficiency in growing-finishing
phase by measuring the amount of feed provided to the pig between weaning and slaughter per
kilogram of carcass weight produced.
The total amount of feed required for a reproductive sow for producing piglets mainly depends on
the feeding level during the gestation and lactation period and the duration of the non productive
periods, i.e. days when a sow is neither pregnant nor lactating. These days include the interval
between the weaning to effective service between successive litters, the time from when the sow
enters the herd until she becomes pregnant, and the time when the sow is weaned at her last litter
until the day she is culled and removed from the herd. On the other hand, the amount of litter BW
produced at weaning is directly connected to the litter size at birth, to the piglet mortality rate and
the litter growth rate during the lactation period. These parameters are generally related to sow
prolificacy and her ability to produce milk. Sow and piglet performance is affected by numerous
factors related to the sow’s characteristics (breed, parity, sanitary status, body condition) and/or
environmental factors (feed composition, housing system, climatic environment). Among these
latter factors, there is now clear evidence that thermal environment (cold/heat) is one of the main
factor influencing a sow’s productivity.
As a large proportion of feed input (about 75 to 80% of total feed purchased) is utilized during the
growing-finishing phase in a traditional farrow-to-finish pig production system, feed efficiency
during this period is an important determinant of profitability for swine producers. Feed efficiency
in growing finishing pigs can be evaluated by measuring the feed conversion ratio (FCR) which
defines the feed requirement in kg per kg body weight (BW) gain. From a practical point of view,
FCR can be expressed with the following relationship:
FCR = (MEi/[ME])/(RE/[EADG])
where MEi, RE, [ME] and [EADG] correspond to daily ME intake, daily energy gain, diet energy content (Mcal ME/kg)
and energy content of BW gain (Mcal/kg), respectively.

As the amount of RE depends both on the amount of ME above maintenance (MEi-MEm) and
the partial efficiency of its utilization for growth (kg), FCR can also be expressed as follows (Henry
and Noblet, 1986; Noblet et al., 1994b):
FCR = (1/[ME]) × (1/kg) × [E]ADG × (FL/[FL – 1])
where FL is the feeding level (MEi as a multiple of maintenance requirement).
This equation shows the well-known effects of energy density and energy content of BW gain
(or body composition) on FCR. But it also indicates that FCR is reduced with increased FL and
according to a curvilinear relationship with a sharp increase of FCR at energy intakes close to
the maintenance level and a small decrease at very high feed intakes. A reduced FCR can then
be achieved by raising the ME intake and/or by reducing the maintenance requirement (MEm).
However, according to the curvilinear relationship between FL and [EADG], the reduction of
FCR with the increase of FL is attenuated, especially when MEi increases above ME required
for maximal protein deposition. Under ad libitum conditions and if adequate feed is used, it is
assumed that climatic factors, especially hot climatic conditions, have small effects on intake (kg)
(D. Renaudeau, unpublished results).Therefore, the effects of climatic conditions on FCR should
be envisaged according to both their effects on body composition and feed intake relative to the
maintenance energy requirements.
Thermoregulation and energy utilization
Pigs are homeothermic animals as they can maintain body temperature within narrow limits
under varying environmental thermal conditions by balancing heat loss and heat production
(HP).The relationships between ambient temperature (Ta) and the balance between heat loss and
heat production is schematically illustrated in Figure 1. According to Mount et al. (1974), the
thermoneutral zone is defined as the Ta range above which, at a fixed level of energy intake, heat
production is minimal and constant. The lower and upper limits of the thermoneutral zone are called
the lower (LCT) and upper (UCT) critical temperatures, respectively. In thermoneutral conditions,
heat production is related to the utilization of metabolisable energy (ME) for maintenance and
productive processes. At maintenance, total ME intake is converted into heat as the energy retention
is nil. The contribution of maintenance HP to total body HP varies with the level of production
averaging 70 to 72% in the growing pig (Noblet et al., 1999), 90 to 92% in the pregnant sow (Noblet
et al., 1997), and 65% in the lactating sow (Noblet and Etienne, 1987). Heat associated with the
productive processes results from the synthesis of new tissues (muscle and fat tissues, fetal tissues,
milk) from ME above maintenance. The amount of heat generated for tissue deposition depends on
the partition of ME into protein and lipids synthesis and on the biochemical origin of nutrients used
for meeting the requirements. In fact, the energetic efficiency (net energy to ME ratio) for protein
deposition is lower than for lipid deposition (60 vs. 80%; Van Milgen et al., 2000) and the efficiency
of utilization of ME for NE (kg) varies with dietary characteristics: the kg value is higher for starch
(82%) and fat (90%) than for crude protein (CP; 60%) or dietary fiber (60%) (Noblet et al., 1994a).
In other words, the amount of HP that will be dissipated in the environment will be higher in high
CP or fiber diets than in high starch or fat diets. These variations in thermic effect of diet (TEF) are
used to formulate diets adapted to different Ta situations (see below).

Thermoneutral zone
Heat production or losses Body temperature
Heat production
Latent heat loss
Sensible heat loss
Th LCT ECT UCT
Ambient temperature
Figure 1. Diagrammatic representation of the relation between heat production or heat losses and body
temperature as affected by ambient temperature (adapted from Mount et al., 1974). Th, LCT, ECT, and UCT are
threshold temperatures for hypothermia, lower, evaporative and upper critical temperatures, respectively.
Ambient temperature affects both total heat loss and the partition between latent (evaporative)
and sensible heat loss. In thermoneutral conditions, body temperature is maintained constant
through some adjustments in heat losses. Between the LCT and the evaporative critical
temperature (ECT), sensible heat loss is predominant. This loss mainly depends on the thermal
gradient between the animal surface and the contacting surface (surrounding air for radiation
and convection, and the floor or other animals for conduction). In both cases, the body surface
area that is in contact with the surrounding environment also plays a crucial role in heat loss
process. Below LCT, the sensible heat losses increase as a consequence of the increased thermal
gradient between animal surface and surrounding environment, and the only way for the animal
to maintain homeothermy is to increase the rate of heat production. The maintenance of a high
metabolic rate below LCT is closely dependent on both availability of energy substrate (body
reserves, energy intake) and the ability of the animal to utilize these substrates as an energy source
(Le Dividich et al., 1998). When Ta is below the hypothermia threshold, the summit metabolic rate
is reached and the animal cannot compensate for the high rate of heat loss, so body temperature
falls and sooner or later, the body temperature decreases and the animal dies (Curtis, 1983).
Above the ECT, the increase in Ta makes sensible heat transfer less effective because of the
reduction of the required minimal thermal gradient between the skin and the surrounding
environment. Due to the fact that the pig has relatively few functional sweat glands, the major
component of the increase in evaporative heat loss is through increased respiratory rate. Skin
evaporation can also occur by passive processes, especially when the pig can have access to a
wallow. The efficiency of both respiratory and skin evaporative heat loss depends on the vapor
pressure gradient between the animal and the environment. As a consequence, in tropical humid
climates, the evaporative heat losses are less efficient, which accentuates the effect of heat stress
on pig performance. When Ta increases above the UCT, the pig can no longer control its body
temperature. Therefore, the increase in body temperature causes an increase in the metabolic rate

(due to the Van’t Hoff effect) which causes an increase in HP which in turn results in a further
increase in body temperature until the lethal temperature is reached (Curtis, 1983). For instance,
the lethal rectal temperature for a 50 kg BW pigs is about 42 °C on average, but it is variable
between animals (Figure 2).
For the purpose of animal production, the determination of the lower and upper limits of the
thermoneutral zone is of major importance because this zone represents the temperature range
over which the efficiency of energy utilization is maximum and the energy available for tissue
deposition is optimal. Little information exists on the thermoneutral zone in pigs, especially for
UCT values (Table 1). In contrast, studies on LCT are well documented. It depends on many
factors, including animal factors (breed, body weight, physiological stage, thermal insulation),
management factors (feeding level, group size), and environmental factors (relative humidity, air
movement, radiant heat, floor type, sanitary status) (Close and Clark, 1981; Noblet et al., 2001).
In general, LCT decreases as the body weight and the thermal insulation increase or when the
feeding level is increased.
In the following parts of this chapter, the term ‘thermal stress’ will refer to thermal conditions for
which Ta is lower than LCT or higher than UCT.
Consequences of thermal stress on feed efficiency
Physiological and metabolic adjustments resulting from the thermoregulatory responses to a

thermal stress have negative consequences on pig productivity and health. In ad libitum fed
animals, changes in metabolic heat production (HP) are essential mechanisms to maintain
body temperature within a physiologically safe range under cold or heat stress. These changes
have direct consequences on energy intake and/or maintenance requirements which in turn
could reduce energy efficiency as conversion of feed to tissue or product. However, the low
level of performance related to a thermal challenge can also be attributed to a direct effect of Ta
44
Rectal temperature (°C)
43
42
41
40
39
38
20 22 24 26 28 30 32 34 36 38
Ambient temperature (°C)
Figure 2. Change in body temperature from thermoneutrality to lethal ambient temperature in 50 kg BW

Large White pigs housed in individual crates (D. Renaudeau, unpublished results).

Table 1. Values of critical temperature according to the physiological stage and the housing conditions.
Class of pigs Housing Age/BW Critical temperature (°C)
Th LCT ECT UCT
Newborn piglets individually 1d 131 341 - -

individually 2d 101 301 31-332
group 2d - 263
Nursing piglets group 3-4 kg - 203
Weaned piglets individually 28 d - 304
group 28 d - 275
Growing/finishing pigs individually 20 kg - 196 287 306
group 20 kg - 176
individually 60 kg - 186 26-278,9 296
group 60 kg - 166 2510
individually 100 kg - 176 2411 296
group 100 kg - 156
Pregnant sows individually - - 2112,13 302
group - - 1413 302
Lactating sows individually - - 1214 2515 2715
Boars individually - - 2016 - -
1 Berthon et al. (1994), 2 Holmes and Close (1977), 3 Kovacs and Rafai (1973), 4 Rinaldo and Le Dividich (1991),
5 Le Dividich et al. (1980), 6 Adapted from Holmes and Close (1977) for an average ME intake equal to × 2.5 ME
for maintenance, 7 Ingram (1964), 8 Brown-Brandl et al. (2001), 9 Renaudeau et al. (2007), 10 Huynh et al. (2005), 11
Comberg et al. (1972), 12 Noblet et al. (1989), 13 Geuyen et al. (1984), 14 Black et al. (1993), 15 Quiniou and Noblet
(1999), 16 Kemp et al. (1989).
(independent of feed intake) on reproductive physiology, health, and energy metabolism. Finally,
thermal stress could also be a major cause of pig mortality at birth, during the nursing period
or thereafter, and could predispose pigs to mortality or morbidity by other causes (starvation,
diseases, etc.). High mortality rates in nursing piglets or in finishing pigs have a significant impact
on overall farm feed efficiency.
Young pigs
It is well established that pre-weaning mortality has serious economic impact on a sow’s feed
efficiency. Currently, up to 20% of live born piglets do not survive from the onset of farrowing to
weaning at 28-d age and this proportion tends to increase with hyper prolific sows (Le Dividich,
2006). Piglet mortality comprises stillbirths and live-born piglets that die before weaning.
Stillborn piglets represent 5 to 7% of total born and die during farrowing or immediately after
farrowing as a result of asphyxiation or soon after birth since they have been weakened by the
birth process (Herpin et al., 1996). During the last two decades, the number of stillborn piglets

has increased with selection for litter size, limiting the effectiveness of selection for this trait
(Canario et al., 2006). From a survey conducted in Belgium, when Ta in the farrowing room
exceeds 22 °C at the time of parturition, the risk of stillbirth increases significantly (Figure 3).
Similar results were reported in tropical areas (Yang et al., 1996; Renaudeauet al., 2003b). The
related prolonged farrowing duration at high Ta would explain the increased number of stillbirths.
However, appropriate supervision of births and provision of assistance to weaker piglets could
help to save many pigs (Herpin et al., 1996).
Many reports have been conducted to examine the causes of pre-weaning mortality (Le Dividich,
2006). Crushing by the sow, starvation, infection or abnormalities are immediate causes of
mortality with, in most surveys, crushing by the sow and starvation being more frequent (Varley,
1995). Pre-weaning mortality increases with litter size, lighter pigs being at a particular risk
(Herpin et al., 2002). At birth, the piglet experiences a dramatic change in the Ta of its new
surroundings. In comparison with the homeostatic temperature in the sow uterus (38 to 40
°C), the piglets are usually born into a much cooler environment (20 to 22 °C) (Kammersgaard
et al., 2011). In addition, the ability of the piglet to conserve heat is very limited because the
piglet is virtually hairless, devoid of subcutaneous fat, and wet with fetal fluids. Although a warm
environment is usually provided in the farrowing pen using various heat systems, the piglets are
relatively unable to detect it during the first day. In practice, as the Ta of the farrowing room is
much lower than its LCT and close to the temperature for which the metabolic rate is maximal,
the newborn piglet experiences a dramatic period of cold stress until it displays an efficient
thermoregulatory response consisting essentially in the ability to locate itself in the heated area
or close to the udder, to huddle, and to consume sufficient amounts of colostrum and milk
(Le Dividich et al., 1998). The maintenance of a high metabolic rate during this period of cold
stress is closely dependent on both the availability of energy substrate and the ability of the
newborn pig to utilize these energy sources. The requirement for energy is met by mobilization
of energy body reserves (glycogen and fat) and by colostrum ingestion. As the fat content of
newborn piglet is very low when compared to most other mammalian newborns (Mellor and
12
Stillborn piglets (%)
10
2
16 18 20 22 24 26
Setting of farrowing unit temperature at parturition (°C)
Figure 3. Effect of farrowing unit ambient temperature at parturition on the mean percentage of stillborn
piglets (adapted from Vanderhaeghe et al., 2010).

Cockburn, 1986), most of the energy from the body reserves is derived from glycogen stored in
the liver and skeletal muscles. As the depletion rate of the glycogen reserves is rapid after birth,
colostrum rapidly becomes the main source of energy for thermoregulation. The thermogenic
importance of colostrum is illustrated by Noblet and Le Dividich (1981) who reported a positive
relationship between metabolic heat production and the amount of colostrum intake by piglets
kept at 18 °C during the first day postpartum. The colostrum intake is 27% lower at 18 to 20 °C
compared to that at 30 to 31 °C (Le Dividich and Noblet, 1981) with detrimental consequences
on the acquisition of passive immunity and energy supply for sustaining a high metabolic rate.
However, factors related to the piglet such as low birth weight and poor vitality also influence the
consumption of colostrum and therefore they are crucial for piglet survival. Noblet et al. (2001)
summarized the complex etiology of piglet mortality and the central role of cold stress on this
analysis in Figure 4.
From 2 days of age to weaning, the improvement of piglet’s thermal insulation, related to the
increase in peripheral fat accretion, reduces the LCT and decreases its susceptibility to cold stress.
In addition, in the presence of a supplemental heat source in the farrowing pen, the piglet fully
expresses its thermoregulatory behavior with a reduced metabolic extra-thermoregulatory heat
demand (Noblet et al., 2001). During the nursing period, piglet BW gain is mainly related to sow
milk production and, to a lesser extent, on the amount of additional feed (solid/liquid) intake
(Pluske et al., 1995).
Cold stress Low body weight
– –
+
Colostrum intake
–
–
Heat production Immunoglobulins
Energy intake
intake
– –
Hypothermia Passive protection

+
Infectious diseases
Morbidity and (or)
death
Figure 4. Representation of the possible effects of cold stress and low birth weight on the health of the
neonatal pig (Noblet et al., 2001).

Post-weaning, and growing-finishing pigs
The period following weaning between 3 and 4 weeks of age is characterized by rapid changes
in environmental components of the piglets in relation to changes in feed intake and feed
composition, metabolism, and tissue thermal insulation (Le Dividich and Herpin, 1994).
This period is an important time in the life of the pig, one which has significant impact on
future performance (Tokach et al., 1992). Under most commercial situations, weaning results
in simultaneous stress including separation from the sow and mixing of the litters, changes in
housing and climatic environment and in diets. This feeding transition is generally associated with
a critical period of underfeeding leading to reduced feed efficiency and growth rate. The related
low feeding level elevates the newly-weaned pig’s thermal requirements (Le Dividich, 1999). On
average, the LCT is about 26 to 28 °C during the first week post weaning and it decreases down to
23-24 °C in the second week post weaning (Close, 1987). This means that after the critical period
corresponding to about the first two weeks after weaning, Ta can be reduced by 2 to 3 °C/week
until the temperature to be maintained in the finishing house is reached. During this period,
the pig is to some extent able to compensate for a suboptimum environment by increasing its
voluntary feed intake.
In growing-finishing pigs fed nutritionally adequate diets, feed efficiency mainly depends on the
level of feed or energy intake, the maintenance requirement, and the efficiency with which ME
above maintenance is used for growth. Changing feed intake is an efficient strategy used by the
animal to cope with changes in Ta. These variations in energy intake have direct and indirect
effects on energy deposition.
In ad libitum feeding conditions, pigs adjust their energy intake to compensate for the effect
of changes in Ta. In practice, changes in energy intake vary according to the duration and the
intensity of the thermal challenge. For example, when pigs are challenged with a chronic thermal
stress, changes in feed intake occur within the 3 to 6 days after the onset of the cold or the hot
thermal challenge (Verhagen et al., 1988; Renaudeau et al., 2010). According to a quantitative
analysis of data published in the literature, our results show a curvilinear decrease in feed
intake with an increase in Ta (Figure 5). For a 50 kg BW pig, feed intake decreases at a rate of
8 g/d/°C from 16 to 24 °C and then at a rate of 46 g/d/°C from 24 to 32 °C. Over the same Ta
range, feed intake is reduced by 30 and 70 g/d/°C, respectively for a 75 kg BW pig. In addition,
the threshold temperature at which feed consumption starts to decrease is reduced when BW
increases (Renaudeau et al., 2011b). This larger susceptibility in heavier pigs is mainly related to
a lower ability to dissipate heat (lower surface area-to-mass ratio, high backfat thickness). In our
meta-analysis approach, we pointed out that a large variability existed among studies regarding
the effect of Ta on feed intake. As suggested by Le Dividich et al. (1998), this high variability is
explained by numerous factors including breed, degree of body fatness, diet composition, sanitary
status, and other climatic factors (humidity, draught, etc.). In addition, pre-experimental rearing
conditions and their potential related effects on the animal response to heat stress during the
experimental period could be an additional source of variation. Differences in the experimental
protocol between studies (e.g. duration of acclimation period to temperature, treatment before
the experimental period) could also explain some of this variability. Under practical housing
conditions and especially under tropical conditions where buildings are often semi-opened, pigs

3,000
2,500
75 kg BW
ADFI (g/d)
2,000
50 kg BW
1,500 25 kg BW
1000
500
16 18 20 22 24 26 28 30 32
Temperature (°C)
1,100
900
ADG (g/d)
700 75 kg BW
50 kg BW
500 25 kg BW
300
100
16 18 20 22 24 26 28 30 32
Temperature (°C)
3.5
3.3
3.1
75 kg BW
FRC (kg/kg)
2.9
2.7 50 kg BW
2.5
2.3 25 kg BW
2.1
1.9
1.7
1.5
16 18 20 22 24 26 28 30 32
Temperature (°C)
Figure 5. Effect of ambient temperature and pig BW on daily feed intake (ADFI, g/d), average daily gain (ADG,
g/d) and feed conversion ratio (FRC, kg feed/kg gain) (adapted from Renaudeau et al., 2011).
are usually subjected to fluctuations in Ta within a day or over successive days. Consequently, the
application of results collected in closely controlled and constant environments may be inappropriate
when applied to variable Ta conditions. In fact, some evidence shows that pigs are partially capable
of maintaining feed consumption when exposed to a cycling diurnal Ta challenge. For example,
pigs maintained in a hot, cyclic Ta, partially shift their feeding activity to the cooler periods of the
day to minimize the detrimental effect of high Ta on feed intake (Feddes et al., 1988; Renaudeau et
al., 2006). However, in growing-finishing pigs, the extra feed consumed during the cooler period

of the day can compensate for the reduced feed intake during the hot periods of the day as long as
the temperature variation does not exceed ±1.5 °C (Quiniou et al., 2000a).
As shown for feed intake, responses of ADG and FCR to Ta are curvilinear and they are both
strongly affected by pig BW (Figure 5).According to Rinaldo and Le Dividich (1991) and Nienaber
et al. (1987), the Ta range for which the ADG is maximal is 15 to 25 °C in young pigs and 10 to
20 °C in growing-finishing pigs, respectively (Figure 6). The corresponding values for a minimal
FCR are 25 to 30 °C, and 20 to 25 °C, respectively. Under cold conditions, the ADG and FCR
decrease at rates of 8 g/d/°C and 0.040/°C from 18.5 to 12.0 °C in young pigs and 5 g/d/°C and
0.052/°C from 15 to 5 °C in growing finishing pigs (Figure 6). In addition, under cold conditions,
the energy content of the BW gain ([E]gain) remains constant in young pigs between 25 and
15 °C (Rinaldo and Le Dividich, 1991) whereas it increases +22 kcal/kg/°C for growing finishing
MEm (MJ/d/kg BW0.60)
2.5 ME intake (x MEm) 2.0

1.8
2.0
1.6
1.5 1.4
1.2
1.0 1.0
5 10 15 20 25 30 35 5 15 25 35
Temperature (°C) Temperature (°C)
FCR (kg feed/kg gain)
4.0 800
3.5 700
ADG (g/d)
3.0 600
2.5 500
2.0 400
1.5 300
1.0 200
5 10 15 20 25 30 35 5 10 15 20 25 30 35
Retained energy (MJ/d)
20 16
18 14
[E]ADG (MJ/kg)
16 12
14 10
12 8
10 6
8 4
6 2
5 15 25 35 5 15 25 35
Figure 6. Effect of Ta on energy intake (ME × MEm), feed conversion ratio (FCR, kg feed/kg gain) and
average daily gain (ADG, g/d) in 20 kg-pigs pig (▲) and in 65 kg-pigs (○) (Nienaber et al., 1987; Rinaldo
and Dividich, 1991).

pigs between 20 and 5 °C (Nienaber and Hahn, 1987) in relation with an increase of carcass
fatness (Figure 6). Whatever the physiological stage, the reduction of FCR at low Ta is mainly the
consequence of a reduction of the feeding level, expressed as a multiple of MEm and the use of
a fraction of the energy intake for thermoregulation. The effect of the feeding level on FCR may
be emphasized by the increase of [E]gain especially in heavier animals.
At a Ta higher than the optimal range, ADG decreases primarily as a result of a severe decline
in feed intake. The increase of FCR in hot conditions (Ta >24 to 25 °C) depends on both the Ta
level and the pig BW (Figures 5 and 6). From 24 to 32 °C, the rates of increase in FCR average
0.020 and 0.035/°C in 25 and 75 kg BW pigs, respectively (Figure 5). This increase is not linear
and is significant only for extreme Ta (>30 °C). This increased FCR reflects a reduced proportion
of ME intake available for tissue growth, which is mainly explained by the strong reduction of
ME intake rather than an increase in MEm which is not markedly changed. In fact, the slight
decrease of MEm at high Ta is more a consequence of the reduced feed intake (Koong et al., 1985;
De Lange et al., 2002; Labussière et al., 2011) than a direct effect of the high Ta. Under ad libitum
feeding, elevated Ta does not markedly reduce body fatness and [E]gain (as feed restriction will
do at normal Ta) which emphasizes the negative effect of high Ta on FCR.
Pregnant sows
There is no evidence that cold conditions impair the normal course of pregnancy in the sow.
However, pregnant sows are often exposed to a Ta lower than their LCT which can compromise
the building of body reserves, unless feeding level is increased. From a compilation of literature,
Noblet et al. (1997) showed that the daily increment of heat production with Ta reduction varies
with many factors (housing conditions, floor type, feeding level, etc). These latter authors reported
that the daily heat increment ranges from 2 to 2.5 kcal/°C·kg BW-0.75 in group housed sows or in
sows close to the LCT or at high feeding levels to 3.6 to 4.3 kcal/°C·kg BW-0.75 in individually-
housed sows or at low Ta. For the latter case, it can then be calculated that the amount of feed
required to compensate for the increase in heat production is about 60 to 80 g/d/°C for a 220 kg
BW sow fed a standard diet.
In pigs, a certain degree of seasonality of reproduction has been reported in many countries.
Major manifestations of seasonal infertility in sows include delayed onset of puberty, a prolonged
weaning to estrus interval and a reduced proportion of mated sows that farrow. Numerous studies
have highlighted the multi-factorial nature of seasonal infertility problem in sows. They are partly
explained by reduced semen quality of the boar (Wettmann et al., 1976; Suriyasomboon et al.,
2004). However, changes in the photoperiod and/or thermal challenges occurring during the
mating period or in the early gestation period (Edwards et al., 1968) or the extent to which
maternal body reserves are mobilized during the preceding lactation period (Prunier et al., 2003)
also affect reproductive performance. Logically, summer infertility and the related increase in the
number of non-productive days have a direct negative consequence on the amount of feed given
to pregnant sows per born piglets. In addition, severe heat stress may also result in significant
embryo (Wildt et al., 1975) and fetal mortality (Omtvedt et al., 1971), and may emphasize the
deterioration of feed efficiency in pregnant sows during the hottest periods of the year.

Lactating sows
According to its low LCT (15 to 18 °C), cold is not a problem for lactating sows. In contrast,
high Ta is much more problematic for lactating sows which have a high metabolic rate related
to milk production. The reduction in HP under hot conditions is mainly achieved by a decrease
in feed intake. The average rate of ME reduction is 645 kcal/d/°C (Figure 7). However, as shown
for growing pigs, the effect of heat stress on feed consumption becomes more pronounced
as the Ta increases (Quiniou and Noblet, 1999). Concomitantly, the amount of nutrients and
energy available for milk synthesis is reduced with a negative consequence on the litter growth
rate (-50 g/d/°C on average; Figure 7) and the litter BW at weaning. In thermoneutral Ta, the
energy requirement for milk production generally exceeds feed intake capacity, especially in
modern hyper prolific sows. Consequently, the mobilization of body reserves is inevitable to
compensate for the energy deficit. On average, the marginal rate of rise in maternal BW loss is
0.046 kg/°C (Figure 7). Sows losing excessive amounts of body weight have a prolonged weaning
to estrus intervals. Under thermal stress, the energy deficit and occurrence of reproductive failure
are accentuated, especially in young sows (Prunier et al., 1996). Variation in milk production in
hot conditions is mainly explained by reduced feed consumption. However, direct effects of heat
stress on milk production and on the mobilization of body reserves have also been reported in
primiparous (Messias de Bragança et al., 1998) and multiparous sows (Renaudeau et al., 2003a).
a b
120 3,500
Litter BW gain (g/d)
100 3,000
ME intake (MJ/d)
80 2,500
2,000 References
60
Gourdine et al., 2004
1,500
40 Johnston et al., 1999
1000 Laspuir et al., 2001
20 Lynch et al., 1977
500
Messia de Bragança et al., 1998
16 20 24 28 32 16 20 24 28 32
Mullan et al., 1992
Temperature (°C) Temperature (°C) Prunier et al., 1997
c d Quiniou et al., 2000
2.0 60 Quiniou et al., 2001
Quiniou and Noblet, 1999
(MJ/kg litter BW gain)
1.5 50 Renaudeau et al., 2001

BW loss (kg/d)
Renaudeau et al., 2003

ME intake
1.0 40 Schoenherr et al., 1989

Stansbury et al., 1987
0.5 30 Vidal et al., 1991
0.0 20
-0.5 10
16 20 24 28 32 16 20 24 28 32
Figure 7. Compilation of the literature data on the effects of elevated ambient temperature on sow performance
during lactation: (a) metabolizable energy intake (MJ/d), (b) litter body weight (g/d), (c) body weight loss
(kg/d), and (d) metabolizable energy intake (MJ/kg litter BW gain) (adapted from Renaudeau, 2008).

Total feed efficiency for milk production, defined as the ME requirement per unit of litter BW gain,
declines when Ta increases (Figure 7). This reduction of the apparent efficiency of ME utilization
for milk production could be explained to some extent by an increase in the relative importance
of MEm requirement compared to the total ME intake in hot conditions. In addition, heat stress
during lactation may also decrease feed efficiency in pregnant sows by increasing the weaning to
oestrus interval and/or in the growing-finishing phase by decreasing piglet BW at weaning.
Strategies for alleviating the effects of thermal stress on feed efficiency
In cold or hot conditions, solutions for improving feed efficiency can be classified into three
groups; those promoting a higher energy intake, those decreasing the maintenance requirement
and especially extra energy requirement for thermoregulation, and those decreasing the direct
effect of thermal stress on pig metabolism, physiology, and health.
Changes in diet composition
Some evidence shows an interaction between dietary energy content and Ta on pig metabolism
and performance. In growing pigs, TEF is higher in high fiber diets as compared to high starch
diets in thermoneutral conditions, whereas, it is similar for both diets in cold conditions (Noblet
et al., 1985). Similar results are reported in pregnant sows (Noblet et al., 1989). From these results,
the efficiency of ME for energy retention is independent of Ta level for high energy diets but it
increases at low Ta for low energy density diets. In fact, TEF of high fiber diets is partly used
to meet the pig’s elevated maintenance needs in cold whereas it is likely lost or wasted in pigs
housed in Ta higher than LCT. In other words, high heat increment diets could be useful in cold
conditions because HP related to feed utilization can minimize the amount of the other nutrients
or tissues that must be used for extra thermoregulatory HP. Quite logically, in a cold environment,
FCR is similar in growing finishing pigs fed high fiber diets compared to those fed high energy
corn-soybean meal diets (Coffey et al., 1982). On the other hand, the introduction of fiber rich
ingredients in diets has negative consequences on FCR and growth rate under warm conditions.
Pregnant sows are more frequently exposed to cold stress than growing pigs, especially when they
are individually housed. The contribution of dietary fiber to net supply of energy is much greater
in breeding sows than in growing pigs in connection with a higher ability to digest it (Noblet
and Le Goff, 2001) and the energy value of fiber rich ingredients is higher at low Ta (Noblet et
al., 1989). These results are clearly in favor of the use of and the economic interest in high fiber
ingredients in pregnant sows exposed to cold conditions.
Few studies have been done to test the effect of high fiber diets on lactating sows in hot conditions.
According to Schoenherr et al. (1989), the inclusion of 48% of wheat bran in lactation aggravates
the effect of thermal heat stress on sow performance. In this study, the efficiency of utilization of
ME intake for milk production at 32 °C was reduced in the high fiber diet in comparison with
the basal corn-soybean meal diet (8.0 vs. 6.6 Mcal ME/kg litter BW gain). In contrast, in tropical-
humid conditions (26.3 °C and 89% RH on average), we showed that a high fiber diet (36%
wheat bran) improved the sow’s milk production but had negative effects on the mobilization of
maternal body reserves (Renaudeau et al., 2003b).

In contrast to high fiber diets, low heat increment diets in hot conditions can reduce the amount
of HP to be dissipated, increase the portion of energy available for tissue synthesis and global
feed efficiency. Some trials were carried out to evaluate the performance of growing-finishing
pigs reared under low or high Ta conditions and fed high fat diets. In these studies, the economic
value of dietary fat was not only estimated from the impact on fat addition on the amount of
feed required per unit of gain, but also on the days required to reach market weight and the pig’s
carcass merit (fat and lean content). It has been demonstrated that the effects of dietary energy
density on FCR depends on the Ta level and to a lesser extent on the BW. In cold conditions, each
extra 100 kcal increase in dietary ME content results in a decline of FCR of about 0.10 kg feed/kg
gain whereas it remains constant when expressed as Mcal ME intake/kg gain (10.5 Mcal ME/kg
gain on average; Coffey et al., 1982; Stahly and Cromwell, 1979). In these studies, pigs tended
to adjust their voluntary feed intake to the diet energy density for maintaining a constant level
of ME intake and then were not markedly improved which can practically limit the economic
interest of fat supplementation in cold conditions. In hot conditions, the response of FCR to
dietary fat supplementation in growing-finishing pigs is rather similar to that in cold conditions
(-0.12/100 Kcal ME on average; Stahly and Cromwell, 1979; Stahly et al., 1981; Katsumata et al.,
1996; Spencer et al., 2001). However, dietary fat supplementation at high Ta improved growth
performance in growing and in finishing pigs. In hot conditions, high energy diets are better
tolerated because these (1) are associated with a reduced TEF, which must be dissipated to the
environment and (2) they result in higher energy intake (especially in finishing pigs), which
counterbalances to some extent the reduction in feed intake occurring in warm conditions. At
thermoneutrality or at high Ta, pigs fed high fat diets are generally fatter than pigs fed control
diets and this effect depends on the level of fat supplementation (Stahly and Cromwell, 1979;
Stahly et al., 1981; Katsumata et al., 1996; Spencer et al., 2001). This high adiposity is mainly
related to an excess of energy relative to protein intake. However, body fat deposition is more
strongly affected by level of ME intake at high Ta than at thermoneutrality (Katsumata et al.,
1996). This result is partly explained by the large increase in ME intake at elevated Ta. However,
using a pair-feeding experimental design, Le Bellego et al. (2001) demonstrated that maximal
protein deposition rate can be limited by a direct effect of high Ta and they suggested that the
increase in energy intake in hot conditions by dietary manipulations benefits mainly or more to
the lipid than to the protein deposition and increases carcass fatness.
The response to high dietary fat addition- diets has also been studied in heat stressed lactating
sows but to a lesser extent. When high-fat diets were used without an increase in CP or essential
amino acids (AA) to maintain a constant protein/ME ratio, the BW loss during lactation and the
milk yield were not affected by the dietary treatment (McGlone et al., 1988). In contrast, when
the diet was correctly balanced for protein or lysine to ME ratio, the increase in diet nutrient
density in heat stressed sows improved litter BW gain, via an increase in milk-fat content, but it
did not limit the mobilization of body reserves (Dove and Haydon, 1994; Quiniou et al., 2000b;
Schoenherr et al., 1989). However, the magnitude of these effects seems to be dependent on the
rate of dietary fat in the diet.
The utilization of low heat increment diets based on a reduction of dietary crude protein level
has been tested both in growing pigs and in lactating sows in order to attenuate the effect of
high Ta on voluntary feed intake and performance. As reviewed by Renaudeau et al. (2008), the

utilization of low CP diets at high ambient temperature moderately improves energy intake in
hot conditions; the effect seems to be significant only in finishing pigs (Spencer et al., 2001). In
lactating sows, the results show great variability. In response to a reduced protein level from 16.8
to 14.3%, Quiniou and Noblet (1999) did not report any improvement in performance of lactating
sows housed at 29 °C. Johnston et al. (1999) observed an increase in litter BW gain (+60 g/d) in
the hot season for mixed parity sows fed a low CP diet (13.7 vs. 16.5%). In this study, the increase
in daily weight gain of the litter was attributed to an increase in the mobilization of sow body
reserves due to a probable imbalance of some essential AA (threonine, tryptophan, and valine)
in the LP diet. In experiments in which essential AA supplies were kept constant and adequately
balanced, heat stressed sows fed low CP diets tended to consume more energy and mobilize fewer
body reserves (Renaudeauet al., 2001; Silva et al., 2009a).
Other feeding strategies
With respect to the postnatal management of the piglet, provision of an early intake of colostrum
or milk replacer for all the piglets or only to the weak, light, and latter born piglets could be
useful to promote survival especially at low Ta. A variety of energy compounds including glucose,
lactose, and lipids have been tested as energy supplements for neonatal pigs with little success
(Noblet et al., 2001). For example, providing supplemental energy as oral doses of corn oil during
the 48 h post partum did not significantly improve pre-weaning mortality (Pettigrew et al., 1986).
The growth of nursing piglets is directly dependent on the ability of the sow to produce milk. As
mentioned above, milk yield is reduced in heat stressed sows which lowers the litter BW gain. To
supplement milk production, producers can provide piglets with creep feed or starter diets based
predominately on cereals and various sources of animal proteins. At thermoneutrality, creep
feed intake is generally low and highly variable among the litters within each study and between
each study (see review of Pluske et al., 1995). When offered during the last week of lactation, the
amount of creep feed consumed before weaning by nursing piglets was higher when sows were
kept in hot conditions (Renaudeau and Noblet, 2001). This higher intake was interpreted as an
adaptation to compensate for insufficient milk production by the heat-stressed sows. Moreover,
Azain et al. (1996) showed that providing litters with a milk substitute during the warm season
is more effective in improving litter BW gain than during the cool season (+38 vs. +10%). In
addition, it has been shown that supplying milk replacer to piglets during periods of heat stress
could attenuate the maternal BW loss and improve post-weaning reproductive performance
(Spencer et al., 2003). These results suggest that the provisioning of supplemental feed or milk
substitutes can attenuate the reduced growth rate of nursing piglets during heat stress and have a
subsequent positive effect on post weaning growth performance. However, these practices need
to be economically evaluated in regard to the high cost of these types of substitutes.
Water is an essential nutrient for livestock, especially during a thermal stress. Water intake during
heat stress is a limiting factor for survival and performance since water has a fundamental role in
the heat exchange system required for temperature regulation and maintenance of water balance
(Mroz et al., 1995). Whatever the physiological stage, water restriction enhances the negative
effect of thermal stress on animal performance and especially on FCR (Nienaber and Hahn,
1984; Leibbrandt et al., 2001). In hot conditions, water losses increase (evaporation by panting)

and water ingested in feed and generated by metabolism is reduced. Consequently, drinking
water consumption has to increase to meet the requirements of a heat-stressed animal. In warm
climates, a key husbandry practice is to provide an abundant clean source of drinking water
close to the feeding area. Moreover, in many high temperature regions, drinking water provided
to pigs is often warm. Some studies demonstrated that providing chilled water would improve
animal performance by reducing body temperature through absorbed heat energy (West, 2003;
Jeon et al., 2006). However, it must be considered that most animals are limited in chilled water
consumption. According to Jeon et al. (2006), chilled water (22 to 15 °C) can provide sufficient
cooling to allow lactating sows to increase their feed intake (+40%) and milk production (+20%)
during heat stress.
Environmental modifications
A broad spectrum of environmental and technical solutions can be used to temper the effects of
high or low Ta on feed efficiency in pigs. Whatever the climate, in order to reduce the effect of
Ta on FCR, approaches leading to reduce the extra energy expenses for thermoregulation and/
or to increase energy intake must be favoured. Classically, these approaches can be divided in
two groups, those modifying the climatic environment to prevent or limit the degree of heat/cold
stress to which the animals are exposed or those enhancing/decreasing heat exchanges between
the animal and the environment (Renaudeau et al., 2011a). Basically, the decision on the degree
to modify the animal environment depends upon the cost of providing improved performance
compared with the value of improved performance. In other words, the merit of maximizing feed
efficiency from an environment standpoint may be questionable according to the relative price
of feed versus energy.
Maintaining the Ta at or higher than the LCT during the 7 to 14 d period following weaning helps
to avoid loss of excessive body fat and hence limits the reduction of thermal insulation and cold
stress and associated consequences of weaning on health (diarrhoea) and on performance (post
weaning growth lag) (Le Dividich et al., 1998). Thereafter, once regular feed intake is established,
Ta can be reduced by about 2 to 3 °C until the Ta to be maintained in the finishing house is reached
(Noblet et al., 2001). During this period, a moderate reduction of Ta does not have marked
detrimental effects on growth performance because the pig is generally able to compensate for the
extra energy requirement for thermoregulation by increasing its voluntary feed intake (Rinaldo
and Dividich, 1991). The related increase in FCR and the associated economic loss can be avoided
by maintaining the post weaning house at an optimal Ta. In practice, especially during the winter
season in temperate climates, a high amount of heating in weaning houses is required to maintain
Ta within the thermoneutral zone and the cost of providing this improved environment has
to be considered. Several ways of reducing the heating requirement while maintaining piglet
performance at acceptable levels have been investigated. As reviewed by Le Dividich and Herpin
(1994), these solutions include the provision of a microenvironment and/or the reduction in
nocturnal temperature. The creation of a microenvironment within the post-weaning crate is
considered the simplest way to save heating costs. For example, hovers and covers are effective
devices to reduce draughts and to capture heat generated by the piglets. According to Shelton
and Brumm (1984), pigs subjected to these energy management techniques generally performed
similar to, or better than, pig housed at recommended air temperatures. The reduction in

nocturnal temperature is based on the fact that the pig displayed marked circadian variation
in its metabolic rate; its LCT is lower during the night time than during the daytime (Van Der
Hel et al., 1984). Compared with a constant Ta, a 4 to 10 °C reduction in nocturnal Ta did not
affect pig performance but it decreased heating costs by 16 to 35% (Brumm et al., 1985; Brumm
and Shelton, 1988; Brumm and Shelton, 1991; Nienaber and Hahn, 1989). However, pigs on the
reduced nocturnal treatment where the reduction in Ta was higher than 8 °C tended to have a
higher incidence and severity of scouring. In some cases, the reduced nocturnal temperature can
deteriorate FCR due to an increase in feed intake, especially during the night (Rinaldo, 1989).
When compared with post weaning pigs, growing-finishing pigs allowed ad libitum access to feed
are less sensitive to cold Ta. Once the other climatic parameters (draughts, relative humidity)
are controlled, the effect of Ta on FCR can be alleviated by providing supplemental heat to the
building. However, this can be costly and is not practical in cases using lower-cost open-front
housing systems. In this situation, providing wind protection for the pigs on three sides and dry
bedding (straw, wood shavings, or other similar materials) in the sleeping area are adequate to
prevent cold stress.
From an economical point of view, heat stress is a greater problem for growing-finishing pigs
than is cold stress because of its negative effects on both FCR and growth rate. As the group size
can influence critical temperatures (Close and Clark, 1981), the number of pigs per pen should
be reduced with the aim to increase the minimum floor space allowed to the pigs. One of the
most effective methods to promote heat losses under high Ta involves the addition of water to the
skin with or without supplemental air flow to increase the rate of evaporation of the additional
water. Water evaporation occurs by absorbing heat directly from the body of the animal and
also by absorbing water to the surrounding air. The efficiency of these systems depends on the
magnitude of thermal stress, the pig BW and the conditions of the water application (Renaudeau
et al., 2011a). The interval between wettings, the duration of water application, and the amount
of supplied water per wetting need to be considered to maximize evaporative heat losses. With
sprinklers operating intermittently (1 min on and 14 min off) below 29.5 °C and continuously
above 29.5 °C, feed intake and average daily gain were improved by 13% in heat stressed finishing
pigs (Nichols et al., 1987). In that study, the reduction of water flow rate from 1.6 to 0.8 l/pig/h did
not change the performance of the spray cooling system. The advantage of wetting finishing pigs
during heat stress is supported by the results presented in Table 2. Whatever the wetting system,
the growth performance was improved when compared to the control group. The experiment
shows that the intermittent wetting system is more effective than continuous fogging system and
the effect of intermittent sprinkling on voluntary feed intake were large enough to improve FCR.
In fact, for evaporative cooling to be valuable, water must be applied intermittently to permit time
for evaporation of the moisture at the skin surface. This rate of evaporation can be improved by
providing supplemental air flow (Turner et al., 1997).
As described before, the thermal requirement of lactating sows differs from that of suckling
piglets. In practice, Ta in the farrowing barn is kept at a compromise between the requirement
of the sow and the piglets. This means that Ta is too low for the litter, especially during the first
days following the farrowing, and too high for the sows. This thermal dilemma is accentuated
either in cold or in heat stress situations. For avoiding this problem, two microclimates have

Table 2. Effects of the cooling system on growth performance in finishing pigs (initial BW=51 kg) during
summer (adapted from Nichols et al., 1979).
Cooling system
Control Fogger1 Sprinkler2
No. of pigs 40 40 40
Feed intake (kg/d) 1.95 2.21 2.36
Average daily gain (kg/d) 0.521 0.580 0.698
FCR (kg/kg)3 3.75 3.84 3.39
1 System that generated low volume, small droplet nozzles and operated continuously when Ta exceeded 26.7 °C.
2 System that generated high volume, large droplet nozzles and operated intermittently (1 min on followed by 29
min off ) when Ta exceeded 26.7 °C
3 FRC = feed conversion ratio.
to be used to cool the sows without detrimental effects on piglet’s well-being and/or to warm
piglets without causing heat stress in sows. An infrared lamp located over the creep area and/or
a heated floor are the most common methods used to provide a thermal environment suitable
for the newborn piglet (Kuhn, 1990). However, the provision of an adequate environment must
take into account the behaviour of the newborn piglet. During the first days of life, the newborn
prefers to lie close to the sow udder rather than in the heat area which increases the risk of being
crushed by the sow (Hrupka et al., 1998). To avoid this problem, a movable infrared lamp is
placed at the rear of the sow during parturition to avoid excessive changes in the piglet’s body
temperature. Thereafter, the lamp is used to direct the piglet away from the zone of the sow and
attract them progressively towards the heating device (heating lamp or map) in the creep area.
Covering the creep area was found to be effective in reducing heat losses in piglets. Because of
the high susceptibly of the lactating sows to high Ta, a lot of alternatives were tested in order to
increase heat exchange between the animal and the environment. Provision of supplemental fresh
air directly on the sow’s head and neck (‘snout cooler’) was found to be an efficient way to improve
performance under heat stress. At 30 °C, the average daily feed intake was increased by 25 to 35%
when snout coolers were used to cool the sow (McGlone et al., 1988; Stansbury et al., 1987). In
these latter studies, voluntary feed intake and milk production were also increased by 24% and
19% by using a drip cooling method based on the application of a water drip for 3 min every
10 min. As previously described, the reduction of energy intake in heat stressed animals is mainly
related to the inability to dissipate sufficient body heat to the environment. As the lactating sow
spends about 85 to 90% of the day resting (Renaudeau et al., 2002), most of the time a large part
of its body is in contact with the floor. Thus, cooling the floor under the sow in the farrowing
crate should increase the dissipation of sow’s body heat by conduction (Table 3). It is interesting
to note that the benefit of a floor cooling system for milk production seems to be higher than for
feed intake. This result also confirms the direct effect of high Ta on mammary gland metabolism.
Finally, use of the floor cooling system did not reduce piglet mortality; it has been suggested that
the surface temperature was not low enough to keep the piglet away (Wagenberg et al., 2006).

Table 3. Effect of floor cooling system on performance of lactating sows and their litter1.
Reference Treatment No. of sows ADFI (kg/d) BW loss (g/d) LBWg (kg/d) WEI (d)
Wagenberg et al. (2006)2 control 29 4.90 1,060 2.52 _

cool 29 5.50 943 2.93 _
Silva et al. (2006)3 control 20 5.60 276 1.70 4.3
cool 20 6.47 -133 2.17 3.9
Silva et al. (2009b)4 control 14 4.71 580 1.96 5.7
cool 15 5.48 328 2.57 4.0
1 Abbreviations used: ADFI = average daily feed intake; BW loss = maternal BW loss during the whole lactation period;
LBWg = litter BW gain during lactation; WEI = weaning to estrus interval.

2 Study performed in Netherlands with an average Ta on 24 °C
3 Study performed in Brazil with an average Ta on 23.5 °C
4 Study performed in Brazil with an average Ta on 25.5 °C
Genetic strategies
Whatever the physiological stage, the depressed effect of thermal stress on FCR cannot be totally
eliminated by management solutions. Moreover, most of these practices are not economically or
technically feasible in all livestock production systems. As a consequence, selection for pigs able
to maintain a low FCR whatever the Ta, if successful, could be the most cost-effective method for
mitigating the effects of thermal stress on FCR.
In growing pigs, an important part (60-70%) of voluntary feed consumption can be explained by
maintenance and production requirements (Dekkers and Gilbert, 2010). The remaining variation,
referred to as residual feed consumption (RFC), may be used for direct selection of feed efficiency
without any change in production. In practice, RFC of individual growing pigs can be estimated
as the residual of a multiple linear regression for feed intake that includes growth rate and body
composition (backfat thickness or carcass lean content) as covariates to account for production
requirements, along with metabolic BW to account for maintenance requirements (Mrode and
Kennedy, 1993). As far as we know, only two selection experiments for RFC have been conducted
in growing pigs, one at INRA (Gilbert et al., 2007) and the other one at Iowa State University
(Cai et al., 2008), both in purebred Large White/Yorkshire pigs. Details on the selections designs
have been given by Gilbert et al. (2007) and Cai et al. (2008). The selection on RFC has shown to
be effective in creating lines of pigs having different feed intake levels but a rather similar growth
potential. Among others, individual variations in RFC can reflect differences in digestion and/
or metabolic utilization of feed which includes differences in heat production and/or in body
composition. From a comparison of two divergent lines with a low (RFC–) or a high RFC (RFC+)
over 4 generations of selection from the INRA experiment, it has been shown that digestibility
of a conventional cereals-soybean meal-diet did not differ between the two lines (Barea et al.,
2010). However, the efficiency of ME utilization for growth was higher in RFC- than in RFC+
pigs, in connection to a reduced total HP (9% reduction in RFC- pigs compared to RFC+ pigs;

Table 4). This difference is mainly due to a reduced basal metabolic rate and, to a lesser extent,
lower physical activity (9% reduction in RFC– pigs; Table 4). According to the importance of
HP on thermoregulatory responses, selection for RFC is assumed to be able to change the ability
of the pig to cope with thermal stress. We have tested this hypothesis through two experiments
conducted at INRA by exposing pigs to a thermal heat stress in which they were exposed to
24 °C (thermoneutrality) during one week and, thereafter, to a constant high Ta (30 or 32 °C)
for 2 or 3 weeks. A first experiment was designed to determine the effect of selection for RFC
on energy utilization during a thermal heat acclimation period, in pigs individually housed in
respiration chambers for a 28 d period. As expected, feed intake and total HP were significantly
reduced under thermal stress (Figure 8). However, these changes were not significantly affected
by line. From data obtained in a second experiment on a larger number of pigs, we found that pigs
with a low RFC tended to have a greater ADG (+50 g/d) and a lower FCR (-0.40) at 30 °C when
compared to pig with a high RFC (Figure 9). These preliminary results would suggest that the
selection for RFC could improve the capacity of pigs to cope with heat stress. However, further
studies are required to validate these results in growing pigs under practical conditions such as
those encountered in tropical areas, and to identify the correlated effects of selection for FCR on
the ability of other physiological stages (piglets, lactating sows) to tolerate cold or hot conditions.
Conclusion
Thermal stress negatively impacts the global feed efficiency on a pig farm by changing the energy
intake and the metabolic utilization of energy for tissue deposition, by reducing the overall
productivity of sows, and/or by increasing the mortality rate whatever the physiological stages.
This lower feed efficiency, when combined with other direct and indirect losses of production
related to climate effects, results in very important economic impacts on the pig industry. Current
and past research has resulted in significant improvements in pig management in cold and hot
conditions. This chapter provides a non exhaustive review of the strategies available to alleviate
Table 4. Effect of a divergent selection for the residual feed consumption (RFC) on energy balance (kcal·kg of
BW-0.60/d) in growing pigs (adapted from Barea et al. 2010).
RFC+ RFC-
No. of pigs 12 12
Average BW (kg) 60.4 60.0
ME intake 668a 622b
Heat production (HP)
Fasting HP 202a 184b
Activity HP 60a 52b
TEF1 94 93
Total 352a 332b
a,b Rows with diferent superscript are significantly differerent (P<0.05).
1 TEF = thermic effect of feeding.

a 3,000
2,500
Feed intake (g/d)

2,000
1,500 RFC+
1000 RFC-
500
0
wk1 wk2 wk3 wk4
(24°C) (32°C) (32°C) (32°C)
b
Total HP(kJ.kg BW-0.60/d)
1,600
1,200
800 RFC+
RFC-
400
0
wk1 wk2 wk3 wk4
(24°C) (32°C) (32°C) (32°C)
Figure 8. Effects of the line and the duration of exposure to 32 °C on (a) feed intake and (b) total heat
production (HP) in growing pigs. Pigs (5/line) were kept at 24 °C for 7 days and thereafter at 32 °C for 21 days
(Renaudeau, Frances, Dubois, Gilbert, Noblet, unpublished results).
a 3.0
FCR (kg feed/kg gain)
2.5
2.0
1.5 RFC+
1.0 RFC-
0.5
0.0
24°C 30°C
b 1.2
1.0
ADG (kg/d)
0.8
RFC+
0.6 RFC-
0.4
0.2
24°C 30°C
Figure 9. Effect of line and ambient temperature on (a) average daily BW gain (ADG) and (b) feed conversion
ratio (FCR) in growing pigs. Pigs (17 RFC+ and to 19 RFC-) were kept at 24 °C for 7 days and thereafter at 30 °C
for 14 days (Campos, Renaudeau, Gilbert, Noblet, unpublished results).

the effects of Ta changes on feed efficiency and, more generally, on pig performance. Maximizing
production levels and the feed efficiency of pig farms is important but economic considerations
largely determine the level of environmental or feeding manipulations used to cope with thermal
stress. In the future, efforts have to be developed to select animals with high production efficiency
and optimal heat tolerance.
References
Azain, M. J., T. Tomkins, J. S. Sowinski, R. A. Arentson, and D. E. Jewell. 1996. Effect of supplemental pig
milk replacer on litter performance: seasonal variation in response. J. Anim. Sci. 74:2195-2202.
Barea, R., Dubois, S., Gilbert, H., Sellier, P., Van Milgen, J., and Noblet, J. 2010. Energy utilization in pigs
selected for high and low residual feed intake. J. Anim Sci. 88:2062-2072.
Berthon, D., P. Herpin, and J. l. Dividich. 1994. Shivering thermogenesis in the neonatal pig. J. Therm. Biol.
19:413-418.
Black, J. L., B. P. Mullan, M. L. Lorschy, and L. R. Giles. 1993. Lactation in the sow during heat stress. Livest.
Prod. Sci. 35:153-170.
Brown-Brandl, T. M., R. A. Eigenberg, J. A. Nienaber, and S. D. Kachman. 2001. Thermoregulatory profile
of a newer genetic line of pig. Livest. Prod. Sci. 71:253-260.
Brumm, M. C., and D. P. Shelton. 1988. A modified reduced nocturnal temperature regimen for early-
weaned pigs. J. Anim. Sci. 66:1067-1072.
Brumm, M. C., and D. P. Shelton. 1991. Two reduced nocturnal temperature regimens for early-weaned
pigs. J. Anim. Sci. 69:1379-1388.
Brumm, M. C., D. P. Shelton, and R. K. Johnson. 1985. Reduced nocturnal temperatures for early weaned
pigs 1, 2, 3. J. Anim. Sci. 61:552-558.
Cai, W., D. S. Casey, and J. C. M. Dekkers. 2008. Selection response and genetic parameters for residual feed
intake in Yorkshire swine. J. Anim. Sci. 86:287-298.
Canario, L., E. Cantoni, E. Le Bihan, J. C. Caritez, Y. Billon, J. P. Bidanel, and J. L. Foulley. 2006. Between-
breed variability of stillbirth and its relationship with sow and piglet characteristics. J. Anim. Sci.
84:3185-3196.
Close, W. H. 1987. The influence of the thermal environment on the productivity of pigs. BSAP Occasional
Publication 11:9-24.
Close, W. H., and J. A. Clark. 1981. The climatic requirements of the pig Environmental aspects of housing
for animal production No. Butterworths. p 149-166. Butterworths, London.
Coffey, M. T., R. W. Seerley, D. W. Funderburke, and H. C. McCampbell. 1982. Effect of heat increment and
level of dietary energy and environmental temperature on the performance of growing-finishing swine.
J. Anim. Sci. 54:95-105.
Comberg, G., E. Stephan, and H. Spath. 1972. The reaction of German Landrace boar progeny groups to
different climatic pen conditions. Pt. 3. Zuchtungskunde 44:402-415.
Curtis, S. E. 1983. Environmental management in animal agriculture, Ames/Iowa.
De Lange, C. F. M., J. van Milgen, J. Noblet, S. Dubois, and S. H. Birkett. 2002. Previous feeding level
influences fasting heat production in growing pigs. J. Anim. Sci. 80 (suppl. 1):63.
Dekkers, J. C. M., and H. Gilbert. 2010. Genetic and biological aspect of residual feed intake in pigs, Leipzig,
Germany. p 8 pp.
Dove, C. R., and K. D. Haydon. 1994. The effect of various diet nutrient densities and electrolyte balances
on sow and litter performance during two seasons of the year. J. Anim. Sci. 72:1101-1106.

Edwards, R. L., I. T. Omtvedt, E. J. Turman, D. F. Stephens, and G. W. A. Mahoney. 1968. Reproductive

performance of gilts following heat stress prior to breeding and in early gestation. J. Anim. Sci. 27:1634-1637.
Feddes, J. J. R., J. A. DeShazer, and A. M. Parkhurst. 1988. Dynamic responses of growing pigs to high cyclic
and constant temperature, St. Joseph, Mich. p 85-92.
Geuyen, T. P. A., J. M. F. Verhagen, and M. W. A. Verstegen. 1984. Effect of housing and temperature on
metabolic rate of pregnant sows. Animal Production 38:477-485.
Gilbert, H., J. P. Bidanel, J. Gruand, J. C. Caritez, Y. Billon, P. Guillouet, H. Lagant, J. Noblet, and P. Sellier.
2007. Genetic parameters for residual feed intake in growing pigs, with emphasis on genetic relationships
with carcass and meat quality traits. J. Anim Sci. 85:3182-3188.
Henry, Y., and J. Noblet. 1986. Alimentation énergétique Le porc et son élevage. p. 233-260.
Herpin, P., M. Damon, and J. Le Dividich. 2002. Development of thermoregulation and neonatal survival
in pigs. Livest. Prod. Sci. 78:25-45.
Herpin, P., J. Le Dividich, J. C. Hulin, M. Fillaut, F. De Marco, and R. Bertin. 1996. Effects of the level of
asphyxia during delivery on viability at birth and early postnatal vitality of newborn pigs. J. Anim. Sci.
74:2067-2075.
Holmes, C. W., and W. H. Close. 1977. The influence of climatic variables on energy metabolism and
associated aspects of productivity in the pig. p 51-73 in Nutrition and the climatic environment. Studies
in the agricultural and food sciences. Butterworths, London, UK.
Hrupka, B. J., V. D. Leibbrandt, T. D. Crenshaw, and N. J. Benevenga. 1998. The effect of farrowing crate heat
lamp location on sow and pig patterns of lying and pig survival. J. Anim. Sci. 76:2995-3002.
Huynh, T. T. T., A. J. A. Aarnink, M. W. A. Verstegen, W. J. J. Gerrits, M. J. W. Heetkamp, B. Kemp, and T.
T. Canh. 2005. Effects of increasing temperatures on physiological changes in pigs at different relative
humidities. J. Anim. Sci. 83:1385-1396.
Ingram, D. L. 1964. The effect of environemental temperature on body temperature, respiratory frequency
and pusle rate in the young pig. Research Veterinary Science 5:348-356.
Jeon, J. H., S. C. Yeon, Y. H. Choi, W. Min, S. Kim, P. J. Kim, and H. H. Chang. 2006. Effects of chilled
drinking water on the performance of lactating sows and their litters during high ambient temperatures
under farm conditions. Livestock Science 105:86-93.
Johnston, L. J., M. Ellis, G. W. Libal, V. B. Mayrose, and W. C. Weldon. 1999. Effect of Room Temperature
and Dietary Amino Acid Concentration on Performance of Lactating Sows. J. Anim. Sci. 77:1638-1644.
Kammersgaard, T. S., L. J. Pedersen, and E. Jorgensen. 2011. Hypothermia in neonatal piglets: Interactions
and causes of individual differences. J. Anim. Sci. 89:2073-2085.
Katsumata, M., Y. Kaji, and M. Saitoh. 1996. Growth and carcass fatness responses of finishing pigs to dietary
fat supplementation at a high ambient temperature. Anim. Sci. 62:591-598.
Kemp, B., M. W. A. Verstegen, L. A. den Hartog, and H. J. G. Grooten. 1989. The effect of environemental
temperature on metabolic rate and partitoning of energy intake in breeding boars. Livest. Prod. Sci.
23:329-340.
Koong, L. J., C. Ferrell, and J. A. Nienaber. 1985. Assessment of interrelationships among level of intake and
production, organ size and fasting heat production in growing animals. J. Nutr. 115:1383-1390.
Kovacs, F., and P. Rafai. 1973. Metabolism in newborn and young pigsAz ujszulott es fiatal malacok
anyagcserejenek vizsgalata. Magy. Allatorv. Lapja 28:182-187.
Kuhn, J. 1990. Klimatisierung von Abferkelstallen. Deutsche Geflugelwirtschaft und Schweineproduktion
28:830-835.
Labussière, E., J. van Milgen, C. F. M. de Lange, and J. Noblet. 2011. Maintenance energy requirements of
growing pigs and calves are influenced by feeding level. The Journal of Nutrition 141:1855-1861.

Le Bellego, L., J. van Milgen, and J. Noblet. 2001. Effect of high temperature and energy intake on energy
utilization in growing pigs. J. Anim. Sci. 79 (Suppl.1):211.
Le Dividich, J. 1999. A review – Neonatal and weaner pig: management to reduce variation, Adelaide, South
Australia. p 135-155.
Le Dividich, J. 2006. The issue of colostrum in piglet survival: energy and immunity. Page 89-102. Nutritional
biotechnology in the feed and food industries: Proceedings of Alltech’s 22nd Annual Symposium,
Lexington, Kentucky, USA, 23-26 April 2006.
Le Dividich, J., and P. Herpin. 1994. Effects of climatic conditions on the performance, metabolism and
health status of weaned piglets: a review. Livest. Prod. Sci. 38:79-90.
Le Dividich, J., and J. Noblet. 1981. Colostrum intake and thermoregulation in the neonatal pig in relation
to environmental temperature. Neonatology 40:167-174.
Le Dividich, J., J. Noblet, P. Herpin, J. van Milgen, N. Quiniou, J. Wiseman, M. A. Varley, and J. P. Chadwick.
1998. Thermoregulation progress in pig science. p 229-263. Nottingham University Press, Nottingham.
Le Dividich, J., M. Vermorel, J. Noblet, J. C. Bouvier, and A. Aumaitre. 1980. Effects of environmental
temperature on heat production, energy retention, protein and fat gain in early weaned piglets. Br. J.
Nutr. 44:313-323.
Leibbrandt, V. D., L. J. Johnston, G. C. Shurson, J. D. Crenshaw, G. W. Libal, and R. D. Arthur. 2001. Effect of
nipple drinker water flow rate and season on performance of lactating swine. J. Anim. Sci. 79:2770-2775.
McGlone, J. J., W. F. Stansbury, and L. F. Tribble. 1988. Management of lactating sows during heat stress:
effects of water drip, snout coolers, floor type and a high energy-density diet. J. Anim. Sci. 66:885-891.
Mellor, D. J., and F. Cockburn. 1986. A comparison of energy metabolism in the new-born infant, piglet and
lamb. Exp. Physiol. 71:361-379.
Messias de Bragança, M., A. M. Mounier, and A. Prunier. 1998. Does feed restriction mimic the effects of
increased ambient temperature in lactating sows? J. Anim. Sci. 76:2017-2024.
Mount, L. E., J. L. Monteith, and L. E. Mount. 1974. The concept of thermal neutrality Heat loss from animals
and man. p 425-439. Butterworths, London.
Mrode, R. A., and B. W. Kennedy. 1993. Genetic variation in measures of food efficiency in pigs and their
genetic relationships with growth rate and backfat. Animal Production 56:225-232.
Mroz, Z., A. W. Jongbloed, N. P. Lenis, and K. Vreman. 1995. Water in pig nutrition: physiology, allowances
and environmental implications. Nutr. Res. Rev. 8:137-164.
Nichols, D. A., D. R. Ames, and R. H. Hines. 1979. Evaporative cooling systems for swine. Report of Progress,
Agricultural Experiment Station, Kansas State University:6-9.
Nichols, D. A., R. C. Thaler, J. P. Murphy, R. H. Hines, and J. L. Nelssen. 1987. The value of drip versus
spray cooling at two flow rates to reduce heat stress of finishing pigs. Report of Progress, Agricultural
Experiment Station, Kansas State University:58-60.
Nienaber, J. A., and G. L. Hahn. 1983. Performance of growing-finishing swine in response to the thermal
environment. In: Transactions of the American Society of Agricultural Engineers. p 1-32.
Nienaber, J. A., and G. L. Hahn. 1989. Cool nighttime temperature and weaning age effects on 3 to 10 week
old pigs. Transactions of the American Society of Agricultural Engineers 32:691-695.
Nienaber, J. A., G. L. Hahn, and J. T. Yen. 1987. Thermal environment effects of growing-finishing swine.
I. Growth, feed intake and heat production. Transactions of the American Society of Agricultural
Engineers 30:1772-1775.
Nienaber, J. A., and G. LeRoy Hahn. 1984. Effects of water flow restriction and environmental factors on
performance of nursery-age pigs. J. Anim. Sci. 59:1423-1429.

Noblet, J., J. Y. Dourmad, M. Etienne, and J. Le Dividich. 1997. Energy metabolism in pregnant sows and
newborn pigs. J. Anim. Sci. 75:2708-2714.
Noblet, J., J. Y. Dourmad, J. Le Dividich, and S. Dubois. 1989. Effect of ambient temperature and addition
of straw or alfalfa in the diet on energy metabolism in pregnant sows. Livest. Prod. Sci. 21:309-324.
Noblet, J., and M. Etienne. 1987. Metabolic utilization of energy and maintenance requirements in lactating
sows. J. Anim. Sci. 64:774-781.
Noblet, J., H. Fortune, X. S. Shi, and S. Dubois. 1994a. Prediction of net energy value of feeds for growing
pigs. J. Anim. Sci. 72:344-354.
Noblet, J., C. Karege, and S. Dubois. 1994b. Prise en compte de la variabilité de la composition corporelle
pour la prévision du besoin énergetique et de l’efficacité alimentaire chez le porc en croissance. Journées
de la Recherche Porcine en France 26:267-276.
Noblet, J., C. Karege, S. Dubois, and J. van Milgen. 1999. Metabolic utilization of energy and maintenance
requirements in growing pigs: effects of sex and genotype. J. Anim. Sci. 77:1208-1216.
Noblet, J., and J. Le Dividich. 1981. Energy Metabolism of the Newborn Pig during the First 24 h of Life.
Neonatology 40:175-182.
Noblet, J., J. Le Dividich, and T. Bikawa. 1985. Interaction between energy level in the diet and environmental
temperature on the utilization of energy in growing pigs. J. Anim. Sci. 61:452-459.
Noblet, J., J. Le Dividich, and J. Van Milgen. 2001. Thermal environment and swine nutrition. p 26 in Swine
Nutrition. A. J. Lewis and L. L. Southern, eds. CRC Press, Boca Raton London New York Washington,
D.C., USA.
Noblet, J., and G. Le Goff. 2001. Effect of dietary fibre on the energy value of feeds for pigs. Anim.Feed.Sci.
Tech. 90:35-52.
Omtvedt, I. T., R. E. Nelson, R. L. Edwards, D. F. Stephens, and E. J. Turman. 1971. Influence of heat stress
during early, mid and late pregnancy of gilt. J. Anim. Sci. 32:312.
Pettigrew, J. E., S. G. Cornelius, R. L. Moser, T. R. Heeg, H. E. Hanke, K. P. Miller, and C. D. Hagen. 1986.
Effects of oral doses of corn oil and other factors on preweaning survival and growth of piglets. J. Anim.
Sci. 62:601-612.
Pluske, J. R., I. H. Williams, F. X. Aherne, and M. A. Varley. 1995. Nutrition of the neonatal pig The Neonatal
Pig, Development and Survival. p 187-235. CAB international, Walligford, UK.
Prunier, A., H. Quesnel, M. Messias de Bragança, and A. Y. Kermabon. 1996. Environmental and seasonal
influences on the return-to-oestrus after weaning in primiparous sows: a review. Livest. Prod. Sci.
45:103-110.
Prunier, A., N. M. Soede, H. Quesnel, B. Kemp, J. R. Pluske, J. Le Dividich, and M. W. A. Verstegen. 2003.
Productivity and longevity of weaned sows Weaning the pig. Concepts and consequences. p 385-419.
Wageningen Academic Publishers, Wageningen, the Netherlands.
Quiniou, N., P. Massabie, and R. Granier. 2000a. Diurnally variation of ambient temperature around 24 or 28
°C: Influence on performance and feeding behavior of growing pigs, Des Moines, Iowa, USA. p 232-239.
Quiniou, N., D. Gaudré, S. Rapp, and D. Guillou. 2000b. Influence de la température ambiante et de la
concentration en nutriments de l’aliment sur les performances de lactation de la truie primipare. Journée
des Recherches Porcines en France 32:275-282.
Quiniou, N., and J. Noblet. 1999. Influence of high ambient temperatures on performance of multiparous
lactating sows. J. Anim. Sci. 77:2124-2134.
Renaudeau, D. 2008. Nutrition of the lactating sows in hot conditions 3rd CLANA congress. Colegio
Latinamericano de Nutricion Animal, Cancun, Q. Roo, Mexico

Renaudeau, D., C. Anais, L. Tel, and J. L. Gourdine. 2010. Effect of temperature on thermal acclimation in
growing pigs estimated using a nonlinear function. J. Anim Sci. 88:3715-3724.
Renaudeau, D., Collin, A., Yahav, S. de Basilio, V., Gourdine, J.L. and R.J. Collierl. 2011a. Adaptation to
tropical climate and research strategies to alleviate heat stress in livestock production: a review. Animal
(doi:10.1017/S1751731111002448).
Renaudeau, D., M. Giorgi, F. Silou, and J. L. Weisbecker. 2006. Effect of breed (lean or fat pigs) and sex on
performance and feeding behaviour of group housed growing pigs in a tropical climate. Asian-Australas.
J. Anim. Sci. 19:593-601.
Renaudeau, D., J. L. Gourdine, B. A. N. Silva, and J. Noblet. 2008. Nutritional routes to attenuate heat stress
in pigs. In: Livestock and Global Climate Change, Hammamet, Tunisia. p 134-138.
Renaudeau, D., J. L. Gourdine, and N. R. St-Pierre. 2011b. A meta-analysis of the effect of high ambient
temperature on growing-finishing pigs. J. Anim. Sci. 89:2220-2230.
Renaudeau, D., E. Huc, and J. Noblet. 2007. Acclimation to high ambient temperature in Large White and
Caribbean Creole growing pigs. J. Anim. Sci. 85:779-790.
Renaudeau, D., J. Noblet, and J. Y. Dourmad. 2003a. Effect of ambient temperature on mammary gland
metabolism in lactating sows. J. Anim. Sci. 81:217-231.
Renaudeau, D., N. Quiniou, S. Dubois, and J. Noblet. 2002. Effect of high ambient temperature and dietary
protein level on feeding behaviour of multiparous lactating sows. Anim. Res. 51:227-243.
Renaudeau, D., and J. Noblet. 2001. Effects of exposure to high ambient temperature and dietary protein
level on sow milk production and performance of piglets. J. Anim. Sci. 79:1540-1548.
Renaudeau, D., N. Quiniou, and J. Noblet. 2001. Effects of exposure to high ambient temperature and dietary
protein level on performance of multiparous lactating sows. J. Anim. Sci. 79:1240-1249.
Renaudeau, D., J. L. Weisbecker, and J. Noblet. 2003b. Effect of season and dietary fibre on feeding behaviour
of lactating sows in a tropical climate. Anim. Sci. 77:429-437.
Rinaldo, D. 1989. Influence de la température ambiante sur le métabolisme énergétique et tissulaire et le
besoin en lysine du porc en croissance. Mise en évidence de l’intérêt d’une température élevée. PhD,
Université de Rennes I.
Rinaldo, D., and J. le Dividich. 1991. Assessment of optimal temperature for performance and chemical body
composition of growing pigs.Livest. Prod. Sci. 29:61-75.
Rinaldo, D., and J. Le Dividich. 1991. Influence de la température ambiante sur les performances de
croissance du porc. Prod. Anim. (Paris) 4:57-65.
Schoenherr, W. D., T. S. Stahly, and G. L. Cromwell. 1989. The effects of dietary fat or fiber addition on yield
and composition of milk from sows housed in a warm or hot environment. J. Anim. Sci. 67:482-495.
Shelton, D. P., and M. C. Brumm. 1984. Response of nursery pigs to reduced nocturnal temperatures. ASAE
paper N°84-4021 St. Joseph, MI.
Silva, B. A. N., J. Noblet, J. L. Donzele, R. F. M. Oliveira, Y. Primot, J. L. Gourdine, and D. Renaudeau.
2009a. Effects of dietary protein level and amino acid supplementation on performance of mixed-parity
lactating sows in a tropical humid climate. J. Anim Sci. 87:4003-4012.
Silva, B. A. N., A. I. G. Oliveira, J. L. Donzele, H. C. Fernandez, M. L. T. Abreu, J. Noblet, and C. G. V.
Nunes. 2006. Effect of floor cooling on performance of lactating sows during summer. Livest. Prod.
Sci. 105:176-184.
Silva, B. A. N., R. F. M. Oliveira, J. L. Donzele, H. C. Fernandes, A. L. Lima, D. Renaudeau, and J. Noblet.
2009b. Effect of floor cooling and dietary amino acids content on performance and behaviour of
lactating primiparous sows during summer. Livestock Science 120:25-34.

Spencer, J. D., R. D. Boyd, R. Cabrera, and G. L. Allee. 2003. Early weaning to reduce tissue mobilization in
lactating sows and milk supplementation to enhance pig weaning weight during extrme heat stress. J.
Anim. Sci. 81:2041-2052.
Spencer, J. D., A. M. Gaines, G. Rentfrow, W. Cast, J. L. Usry, and G. L. Allee. 2001. Supplemental fat and/or
reduced dietary protein crude protein effects on growth performance, carcass characteristics, and meat
quality of late finishing barrows reared in controlled hot environment. J. Anim. Sci. 79:66.
Stahly, T. S., and G. L. Cromwell. 1979. Effect of environmental temperature and dietary fat supplementation
on the performance and carcass characteristics of growing and finishing swine. J. Anim. Sci. 49:1478-1488.
Stahly, T. S., G. L. Cromwell, and J. R. Overfield. 1981. Interactive effects of season of year and dietary fat
supplementation, lysine source and lysine level on the performance of swine. J. Anim. Sci. 53:1269-1277.
Stansbury, W. F., J. J. McGlone, and L. F. Tribble. 1987. Effects of season, floor type, air temperature and snout
cooler on sow and litter performance. J. Anim. Sci. 65:1507-1513.
Suriyasomboon, A., N. Lundeheim, A. Kunavongkrit, and S. Einarsson. 2004. Effect of temperature and
humidity on sperm production in Duroc boars under different housing systems in Thailand. Livest.
Prod. Sci. 89:19-31.
Tokach, M. D., B. Goodband, J. L. Nelssen, and L. J. Kats. 1992. Influence of the weaning weight and growth
during the first week post-weaning on subsequent pig performance. Kansas State University Day, report
of progress, No. 667.
Turner, L. W., H. J. Monegue, R. S. Gates, and M. D. Lindemann. 1997. Fan, sprinkler, and sprinkler plus
fan systems for cooling growing-finishing swine. In: ASAE Annual International Meeting, Minneapolis,
Minnesota, USA, 10-14 August, 1997. p 13 pp.
Van der Hel, W., M. W. A. Verstegen, W. Baltussen, and H. Brandsma. 1984. The effect of ambiant temperature
on diurnal rhythm in heat production and activity in pigs kept in groups. Int. J. Biometeorol. 28:303-315.
Van Milgen, J., N. Quiniou, and J. Noblet. 2000. Modelling the relation between energy intake and protein
and lipid deposition in growing pigs. Anim. Sci. 71:119-130.
Van Wagenberg, A. V., C. M. C. Van de Peet-Schwering, G. P. Binnendijk, and P. J. P. W. Claessen. 2006. Effect
of floor cooling on farrowing sow and litter performance: field experiment under Dutch conditions.
Transactions of the ASABE 49:1521-1527.
Vanderhaeghe, C., J. Dewulf, S. Ribbens, A. de Kruif, and D. Maes. 2010. A cross-sectional study to collect
risk factors associated with stillbirths in pig herds. Anim. Reprod. Sci. 118:62-68.
Varley, M. A. 1995. Introduction. The neonatal pig, development and survival. p 1-16. CAB international,
Wallingford, UK.
Verhagen, J. M. F., R. Geers, and M. W. A. Verstegen. 1988. Time taken for growing pigs to acclimate to
change in ambient temperature. Neth. J. Agric. Sci. 36:1-10.
West, J. W. 2003. Effects of heat-stress on production in dairy cattle. J. Dairy Sci. 86:2131-2144.
Wettmann, R. P., M. E. Wells, I. T. Omtvedt, C. E. Pope, and E. J. Turman. 1976. Influence of elevated ambient
temperature on reproductive performance of boars. J. Anim. Sci. 42:664-669.
Wildt, D., G. Riegle, and W. Dukelow. 1975. Physiological temperature response and embryonic mortality
in stressed swine. American Journal of Physiology 229:1471-1475.
Yang, P. G., W. D. Fang, I. T. Yu, S. H. Wang, C. Y. Tsay, W. B. Chung, and M. Rea-Sen Liu. 1996. Pathological
studies of abortion, stillbirth and neonatal deaths on a swine herd in Taiwan. Journal of the Chinese
Society of Veterinary 22:222-228.

10. Fueling the immune response: what’s the cost?
R.W. Johnson
Department of Animal Sciences and Division of Nutritional Sciences, University of Illinois, 4
Animal Sciences Laboratory, 1207 W. Gregory Drive, Urbana, IL 61801, USA; rwjohn@illinois.edu
Abstract
Pigs kept in environments where they are exposed to a high number of pathogenic microbes
have reduced feed intake and growth, even when no obvious acute illness exists. This chronic
drain on production is called immunological stress. Sentinel immune cells (e.g. macrophages)
‘sense’ the diverse microbial environment by detecting pathogen-associated molecular patterns
(PAMPs), which are molecules associated with groups of pathogens. The immune sentinels detect
PAMPs mainly with Toll-like receptors (TLRs). Stimulation of macrophages through their TLRs
leads to the synthesis and secretion of pro-inflammatory cytokines and prostaglandins, thereby
initiating the inflammatory response that recruits both soluble immune molecules and circulating
immune cells. Pro-inflammatory cytokines enable the immune system to communicate with
other disparate physiological systems. They rearrange the animal’s metabolic priorities, resulting
in re-partitioning of nutrients away from productive processes towards responses that support
the immune system. Thus, the immune system, through detection of PAMPs and production of
pro-inflammatory cytokines, is the critical chain link connecting the pathogenic environment
to productivity. Estimates suggest that at maintenance a healthy animal uses about 0.5-2% of
the body’s lysine for leukocytes, antibodies, and acute phase proteins. When mounting a robust
response to an infectious pathogen the immune response is estimated to account for about 9% of
the body’s lysine. Thus, the cost of immunological stress, in terms of lysine utilization, must lie
somewhere on the gradient beginning at 2% (maintenance) and ending at 9% (robust immune
response). Providing additional lysine in the diet does not alleviate the reduced growth caused
by immunological stress because the capacity for protein accretion is inhibited. Thus, minimizing
exposure to pathogenic microbes with sound environmental management practices must remain
a high priority.
Introduction
Pigs can grow and function most efficiently when they exist in harmonious balance with the other
living forms and the physical and chemical factors in the environment. Each of these three general
factors – the animal, its biologic environment, and its physical environment – affects the others.
When one or more of the actions change in intensity, imbalance can arise. Such imbalance
may lead to abnormal function, which is a sign of disease. Hence, viewed in the broad sense,
the physical and biological environments and the host animal itself are co-determinants of
infectious disease.
The primary pathogenic agent is part of the pig’s biologic environment, of course. To resist the
never-ending challenges of the biologic environment, the pig is equipped with an assortment
of defense mechanisms. These defenses afford protection against infectious microorganisms
so that even in high-density production systems where pigs live surrounded by pathogenic

R.W. Johnson
microorganisms, they become ill relatively infrequently. This protection can be costly as it is well
established that pigs reared in unsanitary environments that afford a high level of host-pathogen
interaction grow poorly compared to pigs kept in more sanitary environments. In the scheme
of things, immunological defenses against infectious pathogens are of the highest priority. After
all, failure to resist may have lethal consequences. During a challenge by an infectious pathogen,
valuable nutrients that might have otherwise gone to support production are readily redirected
to support the host’s immunological defenses. Therefore, effective management of the pathogenic
environment is critical for efficient pig production – not to mention overall health and well-being
of the herd.
Anything that inhibits a host’s immunological defenses renders that individual more susceptible
to pathogenic microorganisms present in the environment. Whereas the pig with a robust
immunological defense can often effectively contend with a pathogen and prevent infection,
the immune-compromised pig may not. Environmental factors often combine to influence the
intensities of both the specific challenge and the pig’s defenses. It is through these effects that
the total environment influences the frequency and the severity of infectious disease. These
secondary pathogenic agents either enhance the primary challenge or impair the defenses. In
effect, they predispose the animal to the debilitating effects of the primary agent.
Feeding a diet that fails to provide essential nutrients in adequate amounts and in forms that are
palatable and bioavailable will ultimately immune-compromise the animal and render it more
susceptible to infectious disease. Thus, a poor diet can be a secondary pathogenic agent. Because
nutrient requirements to support optimal productivity are well defined, marked deficiencies in
protein, amino acids, or trace nutrients are not likely to occur in healthy pigs reared in commercial
situations. However, infectious pathogens can trigger an immune response leading to an amazing
array of metabolic changes, so nutrient requirements to support optimal performance in this
circumstance can differ from those prescribed by the National Research Council (NRC, 1998).
Therefore, understanding the nutritional expense of fueling the immune response is necessary to
maximize feed efficiency. The purpose of this chapter is to briefly discuss the relationship between
the pathogenic microbial environment and growth performance, the paraphernalia of the innate
immune system that senses the microbial environment, and the metabolic effects and cost of the
immune response.
Relationship between disease and growth performance
Pigs kept in environments where they are exposed to a high number of pathogenic microbes
have reduced feed intake and growth, even when no obvious acute illness exists. This chronic
drain on production is called immunological stress (Johnson, 1997). Early research by Hill et al.
(1952) and Coates et al. (1963) showed that chicks kept in a clean, disinfected environment or
a germ-free environment grew faster than those kept under ‘normal’ conditions, suggesting a
negative relationship between the microbial pathogenic environment and growth performance.
This relationship can be further corroborated by two practical examples. First, even when
management practices remain constant, it is clear that the initial groups of animals that pass
through a new facility have superior growth performance compared to subsequent groups (Harris
and Alexander, 1999). Disease producing microorganisms accumulate in the environment from

one group to the next and so on. These pathogenic microbes can keep re-infecting the herd, thus
producing chronic immunologic stress. Second, antibiotics are dispensed to animals not only
for therapeutic treatment of infectious disease, but also for disease prophylaxis and growth. The
growth promotion caused by feeding subtherapeutic levels of antibiotics is greater if pigs are kept
in dirty, poorly sanitized environments than in clean sanitary ones (Cromwell, 2002). Indeed,
in animals maintained under clean conditions where disease producing microorganisms have
not accumulated, prophylactic antibiotics have little or no effect on growth (Hill et al., 1952;
Coates et al., 1963; Roura et al., 1992). Antibiotics are believed to promote growth by reducing
the pathogenic microorganisms the animal must contend against. Unfortunately, prophylactic
antibiotics (and vaccines) lessened the importance of good environmental management practices.
To some extent, producers, researchers and veterinary practioners relied on feeding antibiotics
to enhance growth instead of environmental management procedures to minimize primary
and secondary pathogens. However, the concern about antibiotic-resistant bacteria and the
potential ban on prophylactic antibiotics for growth has stimulated new interest in environmental
management.
One group capitalized on the well-defined relationship between immunological stress and
growth to investigate how nutrient requirements might be affected by the pathogenic microbial
environment (Williams et al., 1997a,b,c). Two management schemes were used to compare pigs
with minimal or widespread exposure to infectious pathogens. On the one hand, a medicated,
early-weaning scheme was implemented wherein sows from which pigs were derived were
vaccinated before farrowing for diseases that had previously been identified in the herd of origin.
Sows were farrowed in sanitized rooms. Piglets were treated with antibiotics at 1, 3, 5, 8, and 11 d
of age. Pigs were weaned at 12 d of age while they still possessed colostrum antibodies and were
given yet more antibiotics and an anti-parasitic. At weaning, pigs were placed in an isolated,
sanitized nursery and given a milk-based diet. On the other hand, sows were not vaccinated
for diseases known to exist in the herd of origin. Sows were farrowed in an unsanitized room.
Pigs were not given antibiotics. Pigs were weaned later (19 d of age) to increase the likelihood of
transfer of disease from the sow. Pigs were weaned into an unsanitary nursery that had previously
been occupied by pigs from the herd of origin. After weaning, in both management schemes, pigs
were provided a diet with 0.60, 0.90, 1.20, or 1.50% lysine. The key features of the two management
schemes are summarized in Figure 1. It is easy to appreciate the disparity in immunological stress
created by the two schemes.
Not surprisingly, from 6 to 27 kg, pigs in the high pathogen exposure group tended to eat less,
grow slower, and be less efficient compared to low exposure pigs (Williams et al., 1997b). The
most important finding, however, was that the degree of immunological stress significantly
impacted the dietary lysine requirement for growth. For example, the low pathogen exposure pigs
responded to greater dietary lysine concentrations. Both growth and efficiency of feed utilization
were maximized by a dietary lysine concentration of 1.50% and daily lysine intakes of 14.7 g in
low pathogen exposure pigs compared to 1.20% and 8.8 g, respectively, for pigs experiencing a
larger pathogen burden. The results were similar for daily accretion rate of body protein (Figure
1B). Thus, immunological stress induced by rearing pigs in an unsanitary environment reduced
both the absolute quantity and the concentration of dietary lysine required to maximize weight
gain and feed efficiency.

R.W. Johnson
A • Sows vaccinated
• Farrowing room sanitized
• Pigs given prophylactic antibiotics
• Sows not vaccinated • Weaned at 12 d into an
• Farrowing room not sanitized Pathogens isolated, sanitized nursery
• Pigs not given antibiotics • Given anti-parasitic
• Weaned at 19 d into an Pathogens
unsanitized nursery
• Not given an anti-parasitic
High (‘dirty’) Low (‘clean’)

Exposure to infectuous pathogens
B
Body protein accretion rates
from 6 to 27 kg BW
120
‘Clean’
100 ‘Dirty’
Req. for maximal protein accretion
Protein (g/d)
80
60 Clean Dirty
Dietary lysine
40 concentration 1.5% 1.2%
20 Daily lysine intake 14.7g 8.8g
0
0.60 0.90 1.20 1.50
Dietary lysine (%)
Figure 1. Immunological stress reduced protein accretion and the amount of lysine required for maximum
growth. (A) Two management schemes were used to compare pigs with minimal or widespread exposure to
infectious pathogens. (B) Protein accretion and lysine required for maximal protein accretion for ‘clean’ and
‘dirty’ pigs (adapted from Williams et al., 1997b).
Similar effects of immunological stress have been reported in chicks, indicating the effects
of infectious disease on growth and nutrient requirements are conserved across species.
A good example is a study where chicks were given repeated injections of Escherichia coli
lipopolysaccharide (LPS) to mimic a chronic infection (Webel et al., 1998a). It should be noted
that this study used a slow-growing genetic line of chicken that remained responsive to repeated
exposure to LPS – most animals develop tolerance and fail to respond after several exposures. The
greater rates of weight gain and protein accretion in control chicks were associated with a higher
dietary lysine requirement when expressed on the basis of absolute intake. However, due to the
reduced feed intake observed in LPS-treated chicks, there was no difference in the concentration
of dietary lysine required to maximize performance when comparing control- and LPS-treated
chicks. Klasing and Barnes (1988) reported that immunogen-injected chicks may require a lower
concentration of dietary lysine to maximize performance than control chicks. Thus, although it

is not entirely clear whether immunologically stressed animals require a lower concentration of
dietary lysine to maximize growth performance, it seems clear that in cases in which voluntary
feed intake is reduced due to immunological stress, it is not effective to increase the dietary lysine
concentration in an effort to bring lysine intake of stressed animals up to the same levels as that
of non-stressed animals. Immunological stress reduces the capacity for body protein accretion
and overfeeding expensive nutrients that can not be utilized is an obvious waste.
How does the immune system sense the pathogenic environment?
The previous section highlighted the relationship between the microbial pathogenic environment
and growth performance. But how does a pathogen in the external environment induce metabolic
changes within the host animal and create deficits in growth performance? To appreciate this, one
needs a rudimentary understanding of the immune defenses used by the host, and an appreciation
for how the host ‘senses’ the diverse microbial environment.
Host defense systems that provide the first line of defense against pathogenic microorganisms
make up the innate immune system. They are inherent, and the capacity to respond does not
change or adapt from one infection to another. The epithelial barriers present the first impediment
to infection (Table 1). However, if an infectious pathogen is highly virulent, present in numbers
that simply overwhelm the system, or the system is injured or weakened, it can penetrate the
epithelial barrier and infect the host. Microbes that penetrate an epithelial barrier are encountered
by the three types of sentinel immune cells in the tissues: macrophages, mast cells and immature
dendritic cells. These sentinels can distinguish debris from normal tissue turnover and foreign
microbial agents by detecting pathogen-associated molecular patterns (PAMPs), which are
molecules associated with groups of pathogens. The immune sentinels detect PAMPs mainly
with Toll-like receptors (TLRs). The PAMPs recognized by TLRs are molecules that are necessary
for the integrity, function, or replication of the microorganism, so the infectious pathogen cannot
change a PAMP to evade detection – doing so would be suicide. Stimulation of macrophages
or mast cells through their Toll-like receptors leads to the synthesis and secretion of pro-
inflammatory cytokines and prostaglandins, thereby initiating the inflammatory response that
recruits both soluble immune molecules and circulating immune cells. Stimulation of dendritic
cells through their TLRs initiates an adaptive immune response.
Table 1. Epithelial barriers to infection.
Mechanical epithelial cells joined by tight junctions

ciliated epithelial cells and mucin trap and remove pathogens
Chemical bactericidal enzymes in saliva, sweat, tears, and gut
low pH in stomach
antibacterial peptides
Microbiological gut microbiota produce antibacterial substances and compete against
pathogenic microorganisms

R.W. Johnson
Thus far, thirteen TLRs have been identified in mammals (Table 2), 10 of which are functional
in humans (Moresco et al., 2011). Genes corresponding to the 10 functional TLR genes that
have been identified in humans have also been cloned in pigs (Uenishi and Shinkai, 2009). The
various TLRs are needed because the PAMPs are molecularly diverse and include nucleic acids,
proteins, lipids and polysaccharides. Furthermore, depending on the PAMP they recognize, some
TLRs are localized to the cell surface whereas others are localized to intracellular membranes.
Two infectious microorganisms affecting swine can be used to illustrate these important points
– Escherichia coli and porcine reproductive and respiratory syndrome virus (PRRSV) (Figure 2).
On the one hand, LPS, the primary ligand for TLR4, is an essential component of the outer
surface of Escherichia coli and other Gram-negative bacteria. Ablation of a gene necessary for
synthesis of LPS is lethal to Gram-negative bacteria (Galloway and Raetz, 1990). Monocytes
and macrophages express TLR4 and two accessory molecules, CD14 and MD-2, on their outer
membrane, and detect LPS that is bound to LPS-binding protein (LBP). The combined actions
of LBP, CD14, TLR4, and MD-2 allow cells of the innate immune system to respond to small
amounts of LPS. Activation of TLR4 initiates intracellular signaling pathways that ultimately
lead to the production of pro-inflammatory cytokines. Many studies have used E. coli LPS to
stimulate the innate immune system in pigs. Bolus i.p. injection of LPS increases plasma levels
of several pro-inflammatory cytokines and induces symptoms indicative of bacterial infection
including reduced feed intake, fever, and lethargy (Johnson and vVon Borell, 1994; Webel et al.,
1997, 1998b).
On the other hand, PRRSV is an enveloped, positive-sense, single-stranded RNA virus

(Cavanagh, 1997) that primarily infects and replicates in monocytes and macrophages (Duan
et al., 1997). During replication of PRRSV and other RNA viruses, double-stranded RNA is
produced as an obligatory intermediate (Luo et al., 2008). Whereas TLR7 recognizes single-
stranded RNA, TLR3 recognizes double-stranded RNA (Alexopoulou et al., 2001; Lund et
Table 2. Mammalian Toll-like receptors and their ligands.
TLR2+TLR1 bacterial lipoproteins

TLR2+TLR6 bacterial lipoproteins, lipoteichoic acid, yeast cell wall mannans
TLR2+? GPI anchors (parasites), bacterial porins, HMGB1
TLR3 dsRNA
TLR4 LPS, HSPs, HMGB1, some viral proteins
TLR5 bacterial flagellin
TLR7 ssRNA (viral)
TLR8 ssRNA (viral)
TLR9 CpG-containing DNA (viral and bacterial)
TLR10 unknown
TLR11 Toxoplasma profilin
TLR12 unknown
TLR13 unknown

E. coli
LBP
LPS
MD-2
CD14
TLR4
Macrophage (cell exterior)
Cytosol
PRRSV
ssRNA
dsRNA
TLR7
TLR3
DNA
Cytokine promoter site(s)
Pro-inflammatory
cytokines
Figure 2. Macrophages express TLR4 and two accessory molecules, CD14 and MD-2, on their outer membrane,
and detect LPS that is bound to LPS-binding protein (LBP). Activation of TLR4 initiates intracellular signaling
pathways that ultimately lead to the production of pro-inflammatory cytokines. TLR3 and TLR7 are localized
to intracellular membranes and recognize double-stranded RNA and single-stranded RNA, respectively.
TLR recognition of a pathogen-associated molecular pattern initiates a signaling cascade that results
in activation of transcription factors ultimately leading to production of pro-inflammatory cytokines,
chemokines, and anti-viral cytokines

R.W. Johnson
al., 2004). Because viral replication is an intracellular event, TLR3 and TLR7 are localized to
intracellular membranes. Recognition of single- or double-stranded RNA initiates a signaling
cascade that results in activation of transcription factors ultimately leading to production of
pro-inflammatory cytokines, chemokines, and anti-viral cytokines (Takeuchi and Akira, 2007).
The involvement of TLR3 in PRRSV infection is evident in two recent studies. In one, pigs
experimentally infected with PRRSV had increased serum levels of IL-1β, IL-6, TNFα, and IFNγ,
as well as fever and reduced growth performance. Infection by PRRSV also increased mRNA for
TLR3 and TLR7 in the tracheobronchial lymph nodes (Miguel et al., 2010). In the other, pigs
injected i.p. with polyinosinic:polycytidylic acid (poly I:C) had increased production of pro-
inflammatory and anti-viral cytokines and symptoms common to PRRSV infection (Dilger and
Johnson, 2010). Poly I:C is a synthetic analog of double-stranded RNA, and as such, interacts with
TLR3. Examination of PRRSV infectivity in porcine alveolar macrophages in vitro indicated that
pre-treatment with poly I:C suppressed PRRSV infection (Sang et al., 2008; Miller et al., 2009).
Thus, the up-regulation of TLR3 and TLR7 in PRRSV infected pigs may restrict viral replication.
How does the immune system affect growth?
The production of pro-inflammatory cytokines upon TLR stimulation is an essential part of the
immune response to infectious pathogens, serving to recruit soluble immune molecules and
immune cells. Pro-inflammatory cytokines enable the immune system to communicate with
other disparate physiological systems and in doing so, set off a systemic alarm – the acute phase
response. They inhibit muscle protein synthesis, enhance muscle protein degradation, stimulate
protein synthesis by the liver, and stimulate fat cells to secrete hormones, to name a few. In short,
cytokines antagonize productive processes. They rearrange the animal’s metabolic priorities,
resulting in a re-partitioning of dietary nutrients away from productive processes towards responses
that support the immune system. This includes the production of acute phase proteins which serve
different physiological functions for the immune system. A shift in the balance between anabolic
and catabolic processes forms the basis for impaired production in pigs subjected to pathogenic
agents. Thus, the immune system, through detection of PAMPs and production of pro-inflammatory
cytokines, is the critical chain link connecting the pathogenic environment to productivity.
The relationship between PRRSV infection, pro-inflammatory cytokines, and growth performance
has been explored to some degree (Toepfer-Berg et al., 2004; Escobar et al., 2006, 2007; Miguel et
al., 2010; Che et al., 2011a,b, 2012) and can be illustrated by a study from Escobar et al. (2004).
In the study, nursery pigs were kept in disease containment chambers to control the pathogenic
environment and inoculated with PRRSV. Infection with PRRSV provoked marked reductions
in feed intake, weight gain and whole body protein accretion. The reduction in weight gain and
protein accretion caused by experimental infection with PRRSV is consistent with that seen in pigs
maintained in an unsanitary environment that provided a high level of host-pathogen interaction
(Williams et al., 1997a,b,c). In the PRRSV study, feed intake, weight gain, and protein accretion were
negatively correlated with circulating levels of IL-1β (Pearson correlation coefficients were -0.80,
-0.72, and -0.72, respectively). Based on a covariance analysis wherein differences in feed intake
for control and infected pigs was corrected, the decrease in protein accretion of PRRSV-infected
pigs was not explained entirely by the reduction in feed intake, suggesting that the systemic pro-
inflammatory cytokine response altered metabolic processes in disparate physiological systems.

Cytokines are known to act directly on skeletal muscle to reduce the efficacy of anabolic
hormones such as IGF-I and insulin (Broussard et al., 2001). However, the study by Escobar
et al. revealed another possibility. Growth differentiation factor-8 (i.e. myostatin) is a member
of the transforming growth factor-β superfamily of secreted growth and differentiation factors.
Myostatin is known to negatively regulate muscle mass in both mice (McPherron et al., 1997;
Szabo et al., 1998) and cattle (Grobet et al., 1997; Kambadur et al., 1997). This appears true
for pigs, too. In a recent study, pigs were immunized against myostatin (Long et al., 2009).
Immunization reduced muscle myostatin expression, increased carcass lean percentage, and
decreased intramuscular fat. Increased myostatin has been reported in several conditions
associated with increased inflammation (Gonzalez-Cadavid et al., 1998; Kirk et al., 2000; Sharma
et al., 2001). In PRRSV-infected pigs, reductions in weight gain and protein accretion were
accompanied by increases in myostatin mRNA in skeletal muscle (Escobar et al., 2004). Thus,
whereas circulating IL-1β was negatively correlated with weight gain and protein accretion, it was
positively correlated with muscle myostatin. Because myostatin is a negative regulator of muscle
mass, these results suggest that the immune system may, in part, regulate protein accretion and
muscle mass during infection by causing an increase in myostatin in skeletal muscle. Therefore,
the magnitude of increases in pro-inflammatory cytokines and myostatin following a disease
challenge may be predictive of the decreases in growth that can be expected.
It has also been speculated that the degradation of skeletal muscle provides substrate to fuel
acute phase protein synthesis by the liver. Acute phase proteins are produced in response to
inflammation and are used clinically as markers of inflammation and infection (Eckersall and
Bell, 2010). Acute phase proteins serve different physiological functions for the immune system
including destroying or inhibiting growth of microbes and enhancing phagocytosis. There is
a substantial quantity of acute phase proteins synthesized shortly after an infectious insult. In
fact, it has been estimated for humans that up to 850 mg acute phase proteins are synthesized
per kg body weight in a typical acute phase response (Reeds et al., 1994). A mismatch in amino
acid composition of muscle and acute phase proteins, suggests more than 1,900 mg of skeletal
muscle protein must be degraded to provide a sufficient quantity of phenylalanine to support
synthesis of the acute phase proteins (Reeds et al., 1994). Substantial skeletal muscle protein
degradation appears to occur in pigs during an immune response. For example, pigs deprived of
feed revealed a threefold increase in plasma urea nitrogen after i.p. injection of LPS (Webel et al.,
1997). Interestingly, the increase in plasma urea nitrogen was only evident after the induction of
several pro-inflammatory cytokines.
Serum levels of the acute phase proteins α1-acid glycoprotein (AGP), haptoglobin, and CRP have
been explored as measures of stress in swine herds. Specific pathogen-free pigs have threefold
lower levels of AGP when compared with pigs infected with Actinobacillus pleuropneumonia and
atrophic rhinitis (Itoh et al., 1993). In addition, Williams et al. (1997a,c) reported that the level
of immune stimulation and the plasma concentration of AGP were positively correlated. Serum
levels of CRP and haptoglobin have also been shown to be elevated in infected swine (Burger et
al., 1992; Hall et al., 1992).

R.W. Johnson
What does it cost to nourish the immune response?
By now it should be clear that to appreciate the total cost of an immune response, one must
account for the cells of the immune system as well as the metabolic responses elicited by pro-
inflammatory cytokines. To do so, Kirk Klasing at the University of California-Davis estimated
the quantitative needs of the immune system for lysine in young chicks by summing up the major
components of the immune response (immune cells and acute phase proteins) and estimating
the amounts of lysine in them relative to the rest of the body (Klasing, 2007). He rationalized
that lysine would be the best indicator since it has few purposes other than for protein synthesis.
Using this approach it was estimated that at maintenance a healthy young broiler chick uses about
0.5-2% of the body’s lysine for leukocytes, antibodies, and acute phase proteins. When mounting
a robust response to an infectious pathogen the immune response is estimated to account for
about 9% of the body’s lysine. The response may include rapid proliferation of lymphocytes,
synthesis and secretion of antibodies, and generation of new phagocytic cells. However, most of
the increase in nutrient needs is to support production of acute phase proteins, which is in line
with the earlier projections by Reeds et al. (1994). Estimates have not been made for pigs, but
given the conserved nature of the immune response across species, it is reasonable to suggest
they would be similar. Thus, the cost of immunological stress, in terms of lysine utilization,
must lie somewhere on the gradient beginning at 2% (maintenance) and ending at 9% (robust
immune response). Since providing additional lysine in the diet does not alleviate the reduced
growth caused by immunological stress (Williams et al., 1997b; Webel et al., 1998a), minimizing
exposure to pathogenic microbes with sound environmental management practices must remain
a high priority.
Acknowledgements
I wish to acknowledge two former mentors, Professors Stanley Curtis and David Baker, who
passed away in 2010 and 2009, respectively. Dr. Curtis was instrumental in developing concepts
about the relationship between the pathogenic environment and animal performance. Indeed,
a portion of this chapter’s Introduction was prepared for what was to be a 2nd edition of Curtis’
classic textbook, ‘Environmental Aspects of Animal Management’. Unfortunately, the 2nd edition
could not be brought to fruition. Dr. Baker introduced me to nutritional sciences when I a was
young assistant professor and developed in me an appreciation for how nutrition and the immune
system intersect. Several of our collaborative projects are discussed herein. Drs. Curtis and Baker
were giants in the animal science community and both were passionate about efficient animal
production. It is appropriate to acknowledge them here.
References
Alexopoulou, L., A. C. Holt, R. Medzhitov, and R. A. Flavell. 2001. Recognition of double-stranded RNA
and activation of NF-kappaB by Toll-like receptor 3. Nature 413:732-738.
Broussard, S., J. H. Zhou, H. D. Venters, R. M. Bluthé, R. W. Johnson, R. Dantzer, and K. Kelley. 2001. At
the interface of environment-immune interactions: cytokines and growth factor receptors. J. Anim. Sci.
79:E268-284.

Burger, W., E. M. Fennert, M. Pohle, and H. Wesemeier. 1992. C-reactive protein--a characteristic feature
of health control in swine. Zentralbl Veterinarmed A. 39:635-638.
Cavanagh, D. 1997. Nidovirales: a new order comprising Coronaviridae and Arteriviridae. Archives of
Virology 142:629-633.
Che, T. M., R. W. Johnson, K. W. Kelley, K. A. Dawson, C. A. Moran, and J. E. Pettigrew. 2012. Effects of
mannan oligosaccharide on cytokine secretions by porcine alveolar macrophages and serum cytokine
concentrations in nursery pigs. J. Anim. Sci. 90: 657-668.
Che, T. M., R. W. Johnson, K. W. Kelley, W. G. Van Alstine, K. A. Dawson, C. A. Moran, and J. E. Pettigrew.
2011a. Mannan oligosaccharide improves immune responses and growth efficiency of nursery pigs
experimentally infected with porcine reproductive and respiratory syndrome virus. J. Anim. Sci.
89:2592-2602.
Che, T. M., R. W. Johnson, K. W. Kelley, W. G. Van Alstine, K. A. Dawson, C. A. Moran, and J. E. Pettigrew.
2011b. Mannan oligosaccharide modulates gene expression profile in pigs experimentally infected with
porcine reproductive and respiratory syndrome virus. J. Anim. Sci. 89:3016-3029.
Coates, M. E., R. Fuller, G. F. Harrison, M. Lev, and S. F. Suffolk. 1963. A comparison of the growth of chicks
in the Gustafsson germ-free apparatus and in a conventional environment, with and without dietary
supplements of penicillin. Br. J. Nutr. 17:141-150.
Cromwell, G.L. 2002. Why and how antibiotics are used in swine production. Animal Biotechnology 13:7-27.
Dilger, R. N. and R. W. Johnson. 2010. Behavioral assessment of cognitive function using a translational
neonatal piglet model. Brain Behav. Immun. 24:1156-1165.
Duan, X., H. J. Nauwynck, and M. B. Pensaert. 1997. Virus quantification and identification of cellular
targets in the lungs and lymphoid tissues of pigs at different time intervals after inoculation with porcine
reproductive and respiratory syndrome virus (PRRSV). Vet. Microbiol. 56:9-19.
Eckersall, P. D. and R. Bell. 2010. Acute phase proteins: Biomarkers of infection and inflammation in
veterinary medicine. Vet. J. 185:23-27.
Escobar, J, T. L. Toepfer-Berg, J. Chen, W. G. Van Alstine, J. M. Campbell and R. W. Johnson. 2006.
Supplementing drinking water with Solutein did not mitigate acute morbidity effects of porcine
reproductive and respiratory syndrome virus in nursery pigs. J. Anim. Sci. 84:2101-2109.
Escobar, J., W. G. Van Alstine, D. H. Baker, and R. W. Johnson. 2004. Decreased protein accretion in pigs
with viral and bacterial pneumonia is associated with increased myostatin expression in muscle. J. Nutr.
134:3047-3053.
Escobar, J., W. G. Van Alstine, D. H. Baker, and R. W. Johnson. 2007. Behavior of pigs with viral and bacterial
pneumonia. J. Appl. Anim. Behav. 105:42-50.
Galloway, S. M. and C. R. Raetz. 1990. A mutant of Escherichia coli defective in the first step of endotoxin
biosynthesis. J. Biol. Chem. 265:6394-6402.
Gonzalez-Cadavid, N. F., W. E. Taylor, K. Yarasheski, I. Sinha-Hikim, K. Ma, S. Ezzat, R. Shen, R. Lalani, S.
Asa, M. Mamita, G. Nair, S. Arver, and S. Bhasin. 1998. Organization of the human myostatin gene and
expression in healthy men and HIV-infected men with muscle wasting. Proc. Natl. Acad. Sci. U.S.A.
95:14938-14943.
Grobet, L., L. J. Martin, D. Poncelet, D. Pirottin, B. Brouwers, J. Riquet, A. Schoeberlein, S. Dunner, F.
Menissier, J. Massabanda, R. Fries, R. Hanset, and M. Georges. 1997. A deletion in the bovine myostatin
gene causes the double-muscled phenotype in cattle. Nat. Genet. 17:71-74.
Hall, W. F., T. E. Eurell, R. D. Hansen, and L. G. Herr. 1992. Serum haptoglobin concentration in swine
naturally or experimentally infected with Actinobacillus pleuropneumoniae. J. Am. Vet. Med. Assoc.
201:1730-1733.

R.W. Johnson
Harris, D. L. and T. J. L. Alexander. 1999. Methods of disease control. Pages 1077-1110 in Diseases of Swine. B.
E. Straw, S. D’Allaire, W. L. Mengeling, and D. J. Taylor, eds. Iowa State University Press, Ames, IA, USA.
Hill, D. C., H. D. Branion, S. J. Slinger, and G. W. Anderson. 1952. Influence of environment on the growth
response of chicks to penicillin. Poult. Sci. 32:464-466.
Itoh, H., K. Tamura, M. Izumi, Y. Motoi, K. Kidoguchi, and Y. Funayama. 1993. The influence of age and
health status on serum alpha-1 acid glycoprotein levels of conventional and pathogen-free pigs. Can.
J. Vet. Res. 57:74-78.
Johnson, R. W. 1997. Inhibition of growth by pro-inflammatory cytokines: an integrated view. J. Anim. Sci.
75:1244-1255.
Johnson, R. W. and E. von Borell. 1994. Lipopolysaccharide-induced sickness behavior in pigs is inhibited
by pretreatment with indomethacin. J. Anim. Sci. 72:309-314.
Kambadur, R., M. Sharma, T. P. Smith, and J. J. Bass. 1997. Mutations in myostatin (GDF8) in double-
muscled Belgian Blue and Piedmontese cattle. Genome Res. 7:910-916.
Kirk, S., J. Oldham, R. Kambadur, M. Sharma, P. Dobbie, and J. Bass. 2000. Myostatin regulation during
skeletal muscle regeneration. J. Cell. Physiol. 184:356-363.
Klasing, K. C. 2007. Nutrition and the immune system. Br. Poult. Sci. 48:525-537.
Klasing, K. C. and D. M. Barnes. 1988. Decreased amino acid requirements of growing chicks due to
immunologic stress. J. Nutr. 118:1158-1164.
Long, D. B., K. Y. Zhang, D. W. Chen, X. M. Ding, and B. Yu. 2009. Effects of active immunization against
myostatin on carcass quality and expression of the myostatin gene in pigs. Anim. Sci. J. 80:585-590.
Lund, J. M., L. Alexopoulou, A. Sato, M. Karow, N. C. Adams, N. W. Gale, A. Iwasaki, and R. A. Flavell.
2004. Recognition of single-stranded RNA viruses by Toll-like receptor 7. Proc. Natl. Acad. Sci. U.S.A.
101:5598-5603.
Luo, R., S. Xiao, Y. Jiang, H. Jin, D. Wang, M. Liu, H. Chen, and L. Fang. 2008. Porcine reproductive and
respiratory syndrome virus (PRRSV) suppresses interferon-beta production by interfering with the
RIG-I signaling pathway. Mol. Immunol. 45:2839-2846.
McPherron, A. C., A. M. Lawler, and S. J. Lee. 1997. Regulation of skeletal muscle mass in mice by a new
TGF-beta superfamily member. Nature 387:83-90.
Miguel, J. C., J. Chen, W. G. Van Alstine, and R. W. Johnson. 2010. Expression of inflammatory cytokines
and Toll-like receptors in the brain and respiratory tract of pigs infected with porcine reproductive and
respiratory syndrome virus. Vet. Immunol. Immunopathol. 135:314-319.
Miller, L. C., K. M. Lager, and M. E. Kehrli, Jr. 2009. Role of Toll-like receptors in activation of porcine
alveolar macrophages by porcine reproductive and respiratory syndrome virus. Clin. Vaccine Immunol.
16:360-365.
Moresco, E. M., D. LaVine and B. Beutler. 2011. Toll-like receptors. Curr. Biol. 21:R488-493.
NRC. 1998. Nutrient requirements of swine. 10th ed. National Academy Press, Washington, D.C.
Reeds, P. J., C. R. Fjeld, and F. Jahoor. 1994. Do the differences between the amino acid compositions of
acute-phase and muscle proteins have a bearing on nitrogen loss in traumatic states? J. Nutr. 124:906-910.
Roura, E., J. Homedes, and K. C. Klasing. 1992. Prevention of immunologic stress contributes to the growth-
permitting ability of dietary antibiotics in chicks. J. Nutr. 122:2383-2390.
Sang, Y., C. R. Ross, R. R. Rowland, and F. Blecha. 2008. Toll-like receptor 3 activation decreases porcine
arterivirus infection. Viral Immunol. 21:303-313.
Sharma, M., B. Langley, J. Bass, and R. Kambadur. 2001. Myostatin in muscle growth and repair. Exerc.
Sport Sci. Rev. 29:155-158.

Szabo, G., G. Dallmann, G. Muller, L. Patthy, M. Soller. and L. Varga. 1998. A deletion in the myostatin gene
causes the compact (Cmpt) hypermuscular mutation in mice. Mamm. Genome 9:671-672.
Takeuchi, O. and S. Akira. 2007. Signaling pathways activated by microorganisms. Curr. Opin. Cell Biol.
19:185-191.
Toepfer-Berg, T. L., J. Escobar, W. G. Van Alstine, D. H. Baker, J. Salak-Johnson, and R. W. Johnson. 2004.
Vitamin E supplementation does not mitigate the acute morbidity effects of porcine reproductive and
respiratory syndrome virus in nursery pigs. J. Anim. Sci. 82:1942-1951.
Uenishi, H. and H. Shinkai. 2009. Porcine Toll-like receptors: the front line of pathogen monitoring and
possible implications for disease resistance. Dev. Comp. Immunol. 33:353-361.
Webel, D. M., B. N. Finck, D. H. Baker, and R. W. Johnson. 1997. Time course of increased plasma cytokines,
cortisol, and urea nitrogen in pigs following intraperitoneal injection of lipopolysaccharide. J. Anim.
Sci. 75:1514-1520.
Webel, D. M., R. W. Johnson, and D. H. Baker. 1998a. Lipopolysaccharide-induced reductions in food intake
do not decrease the efficiency of lysine and threonine utilization for protein accretion in chickens. J.
Nutr. 128:1760-1766.
Webel, D. M., D. C. Mahan, R. W. Johnson, and D. H. Baker. 1998b. Pretreatment of young pigs with
vitamin E attenuates the elevation in plasma interleukin-6 and cortisol caused by a challenge dose of
lipopolysaccharide. J. Nutr. 128:1657-1660.
Williams, N. H., T. S. Stahly, and D. R. Zimmerman. 1997a. Effect of chronic immune system activation on
body nitrogen retention, partial efficiency of lysine utilization, and lysine needs of pigs. J. Anim. Sci.
75:2472-2480.
Williams, N. H., T. S. Stahly, and D. R. Zimmerman. 1997b. Effect of chronic immune system activation on
the rate, efficiency, and composition of growth and lysine needs of pigs fed from 6 to 27 kg. J. Anim.
Sci. 75:2463-2471.
Williams, N. H., T. S. Stahly, and D. R. Zimmerman. 1997c. Effect of level of chronic immune system activation
on the growth and dietary lysine needs of pigs fed from 6 to 112 kg. J. Anim. Sci. 75:2481-2496.

11. Influence of health on feed efficiency
S.S. Dritz
Kansas State University, College of Veterinary Medicine, Diagnostic Medicine/Pathobiology,
1800 Denison Avenue, Manhattan, KS 66506, USA; dritz@vet.k-state.edu
Abstract
The concept of improved health leading to better feed efficiency is a concept that makes intuitive
biologic sense but in reality is difficult to quantify. This is due to the complex interaction between
the pig’s immune response, pathogens that cause disease, and the external environment where the
pig is raised. Mortality causes a direct effect on feed efficiency, since pigs that die consume feed
but do not contribute weight gain in the calculation of feed efficiency. Further influence of health
on feed efficiency arises from chronic immune system stimulation that affects composition of
gain. Since lean tissue gain requires less feed per unit of gain in comparison to fat tissue, healthier
pigs that increase lean tissue gain at a greater rate than fat tissue gain will be more feed efficient.
Evidence indicates that in multi-site pig production systems, chronic immune stimulation is
reduced due to age segregation, leading to less pathogen transmission across age groups and
via the environment from batch to batch. Unfortunately, multi-site pig production systems have
not eliminated all pathogens and clinical disease in pig populations. The predominant pathogen
specific effect of health on feed efficiency in these systems is produced by mortality. A lower
magnitude of health influence on feed efficiency is mediated through metabolic alterations
of survivors. This appears to be the result of the dynamic nature of the pig’s immune system
mitigating the influence in the survivors. Furthermore, it appears that non-specific health-
enhancing factors such as biosecurity and sanitation protocols to prevent lateral transmission
of infectious pathogens and reduced environmental contamination and result in considerable
improvements in feed efficiency.
Introduction
In general, the concept of improved health leading to better feed efficiency is a concept that makes
intuitive biologic sense. However, in the complex realities of swine production systems that are
resource constrained, the practical reality is virtually no population of pigs will be raised free of
infectious pathogens and thus completely free of immune stimulation. Also, the nature of the
immune stimulation will be dependent on specific pathogen, dose of pathogen and interaction
of the pathogen with other pathogens and other environmental factors. Additional complexity
arises from the fact that the immune system is dynamic in nature and influences the clearance of
pathogens and building resistance to subsequent insult.
It is well accepted that stimulation of the immune system by pathogens reduces voluntary feed
intake (Kyriazakis and Doeschl-Wilson, 2009). However, the reduction in feed intake also leads
to a reduction in growth rate. If the effects on growth rate and feed intake are proportional
then the effect on feed efficiency will be limited. A lipopolysaccharide (LPS) model of immune
stimulation has demonstrated that approximately 2/3 of the reduction in growth rate can be
attributed to the reduction in feed intake and approximately 1/3 due to a reduction in feed

S.S. Dritz
efficiency (Dritz et al., 1996). A common model to study immune system stimulation in the
pig and other species utilizes the LPS challenge model (Touchette et al., 2002; Wright et al.,
2000). This is a well-accepted model where a defined dose can be administered without the
complexities of using an infectious pathogen. The LPS challenge model has been quite useful
for understanding the mechanism of endocrine immune function that affects the growth axis.
However, differences between LPS stimulation and challenge with live Salmonella organisms
have been demonstrated (Balaji et al., 2000). An additional layer of complexity arises from the
difficulty translating research under highly controlled conditions that may use a single pathogen
challenge and an optimum environment. These studies are unable to replicate the transmission
and exposure dynamics typically encountered with larger populations under less controlled field
conditions. Also, further complexity is provided by the evolutionary capability of some infectious
pathogens to change over time to influence their virulence and adapt to survival in modern swine
production systems (Osorio, 2010).
Therefore, in order to set the stage for some of the discussion of influences of health on feed
efficiency, some historical perspective is necessary. Infectious pathogen control has played
a critical role in the evolution of pig production systems. There has been a progression from
the observation that vertical infectious pathogen transfer from dam to off-spring could be
prevented through cesarean derivation of the off-spring and lateral pathogen transfer prevented
by segregation during rearing. The challenges with this procedure were the complications of
doing surgery on large numbers of sows and the difficulty raising colostrum deprived pigs.
A fundamental advance eliminated surgical derivation to allow natural colostrum intake
(Alexander et al., 1980). Natural maternal immunity protection and administration of vaccines
and antibiotics prevented vertical pathogen transfer. Further protection was provided by using an
early weaning age and segregation to an isolated off-site location. This eliminated the challenges
of raising colostrum deprived piglets. This led to the observation that these procedures could
be modified at the commercial level and led to significant improvements in health and financial
efficiency (Harris, 2000). These modifications were termed multi-site production and led to wide
scale adoption around the world. Further evolution was provided from the understanding that
the age segregation rather than the early weaning age is the critical factor in the success of multi-
site pig production influencing production traits (Patience et al., 2000; Main et al., 2004). The
major implication is that most pigs in the world are raised in age segregated groups at some level.
The preferred age range at weaning is less than one week of age. Also, segregation can either be
by airspace, barn or site with the predominate thinking that moving up in degree of segregation
from airspace to site results in improved health status.
Direct effects of mortality
The first influence of health on feed efficiency is based on the way feed efficiency is calculated in
most commercial record keeping systems. Many times, especially in nutritional research where
mortality is assumed to be unrelated to treatment, the weight gain of dead pigs is used or feed
intake is adjusted for the loss of the pig. In this scenario, the adjustments are performed to improve
the sensitivity of the experiments and prevent undue influence of the mortality on treatment
means. In contrast, the accepted formula for determining feed efficiency in most commercial

record keeping systems is total feed consumed divided by group live weight gain where gain is
the quantity of weight sold from the group minus quantity of weight started in the group.
The first implication from this formula is that dead pigs will have contributed starting weight and
consumed feed but not contributed weight sold. Thus, there is a direct effect of mortality on feed
efficiency that will vary in magnitude depending on the quantity of pigs that die and when they died.
For example pigs that die later in the finishing period will have a large magnitude of impact on feed
efficiency. These impacts can be quantified by assuming that all dead pigs have consumed 25, 50 or
75% of their indicated feed budget (Table 1). These assumptions can be used to represent the impact
of early-, mid- or late finishing period mortality. This modeling assumes a base feed efficiency of
2.85 in the absence of mortality and a weight gain of 104 kg. The implications of this modeling are
that the direct effect of mortality can have a substantial direct impact on feed efficiency.
Feed efficiency also can be influenced by the direct effect of sow herd health. If accounting for
feed usage for the whole life cycle, health impacts on sow productivity will affect whole herd feed
efficiency. For example, the infection of a sow farm with Porcine Reproductive and Respiratory
Syndrome (PRRS) virus causing reproductive loss, sow mortality and elevated mortality rate
before weaning will lower the number of weaned pigs produced. This leads to a lower amount
of weight produced per sow while having a relatively low impact on sow feed usage. Thus, there
will be a lower weight produced leading to poorer whole herd feed efficiency. Therefore, in a
general sense, any health related improvements in productive response are going to have a direct
influence on feed efficiency.
Chronic immune stimulation
The second area where health affects feed efficiency is based on the direct metabolic effects of
infectious disease on partitioning nutrients between maintenance, lipid and protein deposition.
For example, during an acute infection, body temperature is elevated and maintenance energy
requirements are elevated relative to productive output and lowering the efficiency of feed
utilization (Baracos et al., 1987). Also, the rate of skeletal muscle protein turnover is enhanced
in order to partition amino acids for immune function, leading to lower muscle protein growth
(Klasing, 1988). Since muscle growth is more feed efficient than lipid growth and a reduction in
Table 1. Direct effect of mortality on finisher feed efficiency (23 to 127 kg BW).
Dead pig feed1 Mortality (%)
0 2 4 6 8 10 12
25% 2.85 2.86 2.88 2.90 2.91 2.93 2.95

50% 2.85 2.88 2.91 2.94 2.97 3.01 3.04
75% 2.85 2.89 2.94 2.99 3.04 3.09 3.14
1 Dead pig feed assumes that all dead pigs had consumed the indicated amount of their feed budget prior to death.

S.S. Dritz
growth rate leads to a larger proportion of maintenance for growth, lower growth rates should
lead to less efficient feed utilization.
The quantitative impact on chronic continuous exposure of the immune system is exemplified by
the research of Williams et al. (1997a). A key objective of these experiments was to evaluate the
impact of health status on the rate, efficiency and composition of pig growth and the interaction
with lysine requirements. These experiments were performed by creating high and low health
groups of pigs from the same herd of origin that had evidence of infection with Actinobacillus
pleuropneumoniae (APP), Mycoplasma hyopneumoniae (M. hyo), PRRS virus, Swine Influenza
Virus (SIV) and Transmissible Gastroenteritis (TGE) virus. The high health pigs were weaned
early, age segregated and placed into sanitized isolated facilities. In contrast, the low health status
pigs were weaned at an older age and placed into facilities that had not been cleaned after removal
of the previous group of pigs. In addition, pigs were housed in rooms containing multiple age
groups of pigs with different age groups of pigs continually reintroduced. These pigs had evidence
of multiple pathogen exposure throughout the growing period that included APP, M. hyo, PRRS
virus, SIV and TGE virus. In contrast, the high health status group lacked evidence of exposure
to these pathogens or non-specific stimulation of the immunes system.
The most striking results from these data were that when adequate dietary lysine levels were
provided to the high health pigs, there was a striking improvement in rate of gain, feed efficiency,
carcass muscle percentage and efficiency of carcass muscle gain. These data were pivotal in
understanding the magnitude that chronic immune stimulation from continual exposure
to infectious pathogens can have on lean growth rate and feed efficiency. These studies also
demonstrated the impact chronic immune stimulation can have on carcass lipid to protein
accretion ratios to increase carcass muscle content. The investigators also examined the efficiency
of energy and nitrogen utilization in the two immune exposure groups (Williams et al., 1997b,c).
These data indicated that there was no difference in the maintenance energy requirement or the
efficiency of energy utilization for protein or fat deposition.
Therefore, the key lesson from these data is that the improvements in rate of growth and feed
efficiency were due to composition of gain rather than altered maintenance requirements or
partial efficiencies of gain. Although the energy required for protein and fat synthesis is similar
(NRC, 1998). Fat free lean gain is approximately 3 g of moisture per 1 g of protein. In contrast, fat
tissue gain has little moisture associated and is approximately 90% fat. Thus, when the composition
of growth is shifted from fat tissue to lean gain the efficiency of energy utilization is increased
without affecting the partial efficiency of energy use for protein or fat deposition. Thus, when
lean growth rate is increased relative to fat tissue growth, the feed efficiency of gain will improve.
Production responses to in-feed antimicrobials in multi-site production
Cesarean-derived pigs free of pathogens have little improvement in growth rate. This suggests that
poorer production responses to in-feed antimicrobials should be expected in higher health status
pigs. Therefore, the production responses attributed to in-feed antimicrobial supplementation
should provide some non-pathogen specific evidence of the impact of health on feed efficiency.
According to NRC (1998), the combined results of the two summary studies (Hays, 1978;

Zimmermann, 1986) showed an improvement in feed efficiency of 6.9% in nursery pigs (7 to

25 kg body weight), 4.5% in growing pigs (17 to 49 kg) and 2.2% in finishing pigs (24 to 89 kg).
There was little change in the responses over time between the Hays (1977) and Zimmerman
(1986) summaries. In contrast, an evaluation pooling feed antimicrobial influences across 5
nursery and 4 finisher trials failed to find an influence on feed efficiency across three different
multi-site pig production systems (Dritz et al., 2002). The change in production systems is one
major difference between the Hays (1997) and Zimmerman (1986) and the Dritz et al. (2002)
studies. The lower response to in-feed antimicrobials suggests that multi-site pig production
reduces the infectious pathogen burden on pigs. Additionally, the sanitation and hygiene steps
employed in multi-site production between groups in a given facility reduce the environmental
pathogen burden to which susceptible, arriving young pigs are exposed.
Field data
Unfortunately, with the wide-scale implementation of multi-site, age segregated production

systems, clinical disease is still evident. Multi-site pig production has clearly reduced the chronic
burden of infectious disease and improved productivity. However, due to the complex interplay
of host, pathogen and environment, multi-site pig production has influenced the expression of
clinical disease. These systems can create large populations of immunologically naïve pigs that
can result in epidemics of clinical disease, when a pathogen is introduced into an isolated age
segregated population. In these systems viral pathogens, those that stimulate poorly protective
immunity, or can survive for extended periods in the environment are predominant. Some
pathogens such as Porcine Circovirus Type 2 (PCV2) which is the causative agent of Porcine
Circoviral Associated Disease (PCVAD) are both viral in nature and resistant to inactivation in
the environment. Therefore, the expression of disease in multi-site production is epidemic in
nature and difficult to replicate and model under controlled research conditions.
Surprisingly, there is relatively little data available under field conditions to quantify the influences
of health on feed efficiency. Fortunately, another critical advantage of multi-site pig production
systems is that the distinct age-segregated growing pig groups facilitate collection of production
data for calculation of mortality, growth rate and feed efficiency. Therefore, the objective of this
section is to present some observations on influences of health based on large scale field data.
Porcine Reproductive and Respiratory Syndrome is arguably the most economically important
pathogen on a global basis (Osario, 2010). It has been estimated to cost the US industry over
$US 500 million that has been recently updated to over $US 600 million (Neumann et al., 2005).
Therefore, the first field data evaluation will focus on the effects of PRRS virus infection on
production data. This evaluation is based on data from a production system geographically
located in southern Minnesota and northern Iowa. Sow farms were categorized as actively
shedding PRRS virus based on a subsampling of serum from piglets at weaning for detection of
PRRS virus by PCR. Pigs were then commingled in nurseries at weaning on the basis of presence
or absence of PRRS virus detected in the subsample. Subsequently, after approximately 7 weeks
in the nursery, pigs were moved to finisher sites. The nurseries and finishers were managed on an
all-in-all-out by site basis. Capacity of nursery sites ranged from 2,000 to 6,500 pigs and finisher
site capacity from 2,000 to 4,800 pigs. No PRRS virus vaccine was administered to either sows

S.S. Dritz
or growing pigs during the course of the evaluation. The data were obtained for a 1 year period
from September through the following August and adjusted by dividing the time of group close
out into 4 time periods and including time period in the statistical model to account for seasonal
changes. A subsample of serum was obtained from pigs at the end of the nursery and finisher
periods and evaluated for presence or absence of PRRS virus antibody. Groups where PRRS virus
was not detected at weaning but antibody was detected at the end of the nursery period were
categorized as nursery positive. Groups without detection of PRRS virus antibody at the end
of the nursery period but positive at the end of the finisher period were categorized as finisher
positive. Once either PRRS virus or antibodies were detected in a growth cycle, further testing
was discontinued. Groups without detection of virus at weaning and antibodies at the end of
the nursery and finisher phase were categorized as negative. Finisher group performance was
evaluated based on these categories and feed efficiency was adjusted to a common starting and
ending weight.
The mortality plus cull rate was significantly higher in sites categorized as having PRRS virus
exposure compared to those that were negative (Table 2). Note the reported mortality/cull rate
was only during the finisher period and did not include losses in the nursery. Also, those groups
where PRRS virus viremia was detected at weaning had higher mortality rates compared to
groups where PRRS virus antibodies were detected at the end of the nursery or finisher period.
Growth rate followed a similar pattern as mortality/cull percentage. Feed efficiency was poorer
for pigs categorized as sow farm PRRS virus positive compared to those groups categorized as
negative and positive in the nursery phase. Feed efficiency means were adjusted by assuming the
Table 2. Effect of Porcine Reprodction and Respiratory Syndrom Virus (PRRS virus) category on finisher period
close out mortality/cull percentage, growth rate and feed efficiency.
PRRS virus category1
Item Negative Finisher Nursery Sow
n groups 207 42 58 60
Mortality/culls (%) 4.3a 6.1b 5.5b 9.5c
ADG2 (g) 822a 790b 790b 763c
F:G3 2.59a 2.65ab 2.62a 2.69b
1 Each group was categorized as negative if no PRRS virus was detected in a subsample of serum at weaning or PRRS
virus antibody detected in a sub sample at the end of the nursery or finisher period. Categorized as finisher group
positive if no PRRS virus was detected at weaning or PRRS virus antibody detected at the end of the nursery period
but was detected at the end of the nursery. Categorized as nursery positive if there was no PRRS virus detected at
weaning but antibodies were detected at the end of the nursery period, and sow farm PRRS virus positive if PRRS
virus was detected in the subsample of serum samples at weaning.
3 F:G (Feed:Gain ratio) adjusted to a common start weight of 23 kg and market weight of 113 kg.
a,b,c Means within a row without a common superscript differ at P<0.05.

average pig that died or was culled had consumed half of the targeted feed consumption per pig.
After this adjustment, the feed efficiency was similar across all categories. It appears from this
evaluation that the direct effect of mortality accounts for a large portion of the difference in feed
efficiency across categories. These data are supported by the observations that no differences
in feed efficiency between PRRS virus challenged and unchallenged pigs was detected during
the first 14 days after challenge in one short-term PRRS virus challenge study in nursery pigs
(Toepfer-Berg et al., 2004). However, it is not supported in a similar follow up study (Escobar et
al., 2006) where differences were detected.
The second production data set is derived from asking the question what are health factors that
impact production responses when placing high health pigs into a pig dense geographic area? The
evaluation consisted of both pathogen specific and pathogen non-specific factors. The pathogen
specific effects evaluated were infection with PRRS virus and presence of clinical ileitis. The non-
specific factors were a subjective score of biosecurity/sanitation practices, geographic pig density
and size of production site.
Farm description. Pigs were derived from two different production systems where the sow farms
were located in low pig density geographic regions. Pigs from the first production system were
produced from two different sow sites where pigs where comminged at weaning and transported
to nurseries within the pig dense region. The second system consisted of pigs derived from a
single sow farm and raised in multi-site nursery sites in the same geographic site as the sow
farm. Pigs were then transported to finisher sites in the pig dense region. Feeding program and
management protocols were similar across systems but genotype was different across system.
The finisher sites were located in Sioux County, IA (1,489 km2). Pig density for the county where
the pigs were raised was 722 pigs/km2 based on December 1, 2008 pig inventory for the county
(NASS, 2008). This is the third highest density county for pig production in the United States.
Database description. Data from one growth cycle of pigs at one site were considered a group. A
total of 200 finisher groups across 42 different finisher sites were used in the analysis. There were
105 groups from the first system and 95 groups used from the second system, closed out from
June 2005 through April 2008.
Classification variables. Each group was characterized as PRRS virus positive by either detection
of PRRS virus in diagnostic samples provided from routine clinical investigations. Also, in the
absence of detection of PRRS virus prior to marketing, a sub-sample of serum from 5 to 10 pigs
per group were evaluated for the presence of PRRS virus antibodies. Groups were characterized as
ileitis positive or negative by the presence of clinical signs that were confirmed via histopathologic
evaluation of tissues. Further pathogen specific classification was performed for groups that had
a prior history of either PRRS virus positive or ileitis positive closeout groups.
For the non-specific factors evaluated, the biosecurity/sanitation score was categorized as positive
or negative based on the presence or absence of organic matter present in the facility prior to pig
placement or a subjective assessment of compliance with biosecurity protocols performed after
the group was closed out. Pig density was estimated as the number of pig spaces within a 2.4 km
radius (18.1 km2) of each site at the conclusion of the time period (April 2008). Multiple groups

S.S. Dritz
from the sites were classified based on the April 2008 classification (Range 0 to 15,000, median
5,000). Subsequently, sites classified by pig density were categorized as less than 5,000 and 5,000
or greater number of pig spaces in the 2.4 kg radius. Site capacity (range 300 to 5,000 pigs; median
2,400 pigs) was categorized as less than or greater than 2,500 pigs.
The outcome variables were ADG, feed efficiency, mortality and opportunity cost per top
pig marketed. The opportunity cost was a proprietary economic index to quantify lost profit
opportunity. The index was based on close out group feed economic performance, mortality
economic cost and substandard revenue received from cull pigs. The intent of the index was to
provide a measure of group economic opportunity relative to a benchmark target.
Based on this evaluation, the groups categorized as PRRS virus positive grew slower, were
less feed efficient and had higher mortality rate than those categorized as PRSS virus negative
(Table 3). A large part of the feed efficiency appeared to be mediated by the direct effects of
mortality. The opportunity cost was estimated at $4.71 per pig greater in the PRRS infected
groups compared to the uninfected groups. In contrast, the ileitis category did not influence
ADG, feed efficiency, mortality rate or opportunity cost index. One could argue that there could
have been a misclassification of ileitis category. However, the ileitis prevalence was significantly
greater in groups that had a prior history of ileitis in the previous group (61 vs. 29%; P<0.001). In
contrast, prior history of PRRS virus category had no association with current group prevalence.
Table 3. Effect of pathogen specific and non-pathogen specific factors on growth performance in a pig dense
geographic region.
Item ADG1 (g) Feed:Gain ratio Mortality (%)
PRRS virus
Negative 835a 2.88a 4.2a
Positive 804b 2.93b 5.7b
Ileitis
Negative 817 2.94 5.3
Positive 822 2.89 4.6
Biosecurity/sanitation
Pass 831a 2.88a 4.8
Fail 804b 2.97b 5.5
Site capacity
<2,500 817 2.93 5.3
2,500+ 817 2.91 5.0
Pig density
<5,000 822 2.92 5.2
5,000+ 808 2.93 5.2
a,b Means within a column and item without a common superscript differ at P<0.05.

This would be consistent with the suggestion that Lawsonia intracellularis, the causative agent of
ileitis, is more difficult to inactivate in an environment outside of the pig compared to the PRRS
virus (Amass and Baysinger, 2006; McOrist and Gebhart, 2006).
Interestingly, the subjective categorizations of groups as passing biosecurity/sanitation protocols

were associated with better average daily gain and feed efficiency as well as numerically lower
mortality rates. Since PRRS virus positive percentage was 73% in groups with a failing categorization
some of this effect was associated with PRRS prevalence. However, it does not explain all of the
difference since 48% of the passing groups was categorized as PRRS virus positive. This certainly
lends support to the non-specific nature biosecurity/sanitation has on production performance.
These data are supported by the work of Madec et al. (1998) that indicate longer down times
between batches of pigs reduce residual non-specific bacterial carryover between groups of pigs.
The clear conclusion from these data is that the economic cost of individual pathogens can be high
in modern multi-site production systems but the magnitude of the impact appears dependent on
the specific pathogen. When evaluating pathogen non-specific factors, geographic pig density or
production site capacity did not appear to have a significant influence on production parameters.
In contrast, biosecurity/sanitation factors had a significant influence.
In addition to the effects of PRRSv, pathogen specific effects on feed efficiency have been provided
by field studies with PCV2 vaccine to control disease associated with PCV2. Clearly, these
vaccines have been effective in enhancing survival in the presence of PCVAD (Horlen et al., 2008;
Desrosiers et al., 2009; Neumann et al., 2009). These field studies on the responses to vaccination
have been critical to understanding the magnitude of impact PCVAD was causing in the North
American swine herd. The improvement in mortality rates in PCV2-vaccinated pigs suggests
PCVAD has a direct effect on feed efficiency mediated through mortality.
Also, the impacts of PCV2 vaccination on growth rate have been relatively well documented as
well (Horlen et al., 2008; Jacela et al., 2011, Potter et al., 2012). Additionally, these impacts on
growth rate are similar among the whole population of survivors as indicted by a shifting of the
distribution of growth rates rather than increasing the variability of growth rates (Horlen et al.,
2008; Potter et al., 2012). However, the direct evidence evaluating feed efficiency is more limited
due to the difficulty in collecting feed intake data under field conditions. Also, prior to wide scale
PCV2 vaccination, there were limited diagnostic tools available to categorize groups in regard
to PCVAD.
Although the primary lesions in cases of PCVAD are in the lymphoid system, other components
such as muscle wasting, clinical signs of icterus and swollen kidneys with foci of inflammatory
infiltrates indicate compromises in function of these major metabolic organs, and lesions in
the intestinal tract seem to indicate potential for compromises in efficient feed utilization.
Immunosuppression is also thought to be a major component of the clinical picture of PCVAD
so influences of other potential infectious pathogens may be important in the clinical picture
as well. Therefore, based on biologic mechanisms, the PCVAD clinical syndrome affects many
processes that will impact the efficiency of feed utilization.

S.S. Dritz
In a study by Jacela et al. (2011) PCV2 vaccination improved feed efficiency in one trial with a
tendency for improvement in the second trial where clinical evidence of PCVAD was documented
in unvaccinated pigs. This study was conducted in a finisher barn under commercial conditions.
In this study, weight gain of pigs that died was used in the calculation of feed efficiency to account
for the direct effect of mortality. Averaged over the two trials, feed efficiency was improved by
1.2% for PCV2 vaccinated pigs compared to unvaccinated control pigs. Although this is about
1/3 of the 3.5% improvement in growth rate in PCV2 vaccinated pigs, the economic impact of
the feed efficiency improvement could be greater than US$ 1 per pig depending on feed cost.
In another study, feed efficiency was evaluated in PCV2 vaccinated and unvaccinated pigs that
evaluated dietary lysine requirements in grower (40 to 65 kg BW for barrows and 40 to 60 kg BW
for gilts) and finisher phases (98 to 118 kg BW barrows and 100 to 125 kg BW gilts) (Shelton et al.,
2012). This study was performed in the same commercial finisher facilities as those reported by
Jacela et al. (2011). Removal rate was 6.2 times greater from pens housing unvaccinated compared
to PCV2 vaccinated pigs between d 0 and 71 after weaning (3.2% for PCV2 vaccinated pigs
compared to 19.8% for unvaccinated pigs). This period corresponded to the period between
weaning and just prior to initiation of the grower studies and a large proportion was due to pigs
with clinical evidence of muscle wasting consistent with PCVAD.
Feed efficiency was improved by 2.6% in PCV2 vaccinated pigs compared to unvaccinated pigs
despite the severe PCVAD associated clinical disease. In comparison, in the study by Williams
et al. (1997a) using the chronic immune challenge model, the pigs housed under the high health
conditions and fed above their lysine requirement had over a 15% better feed efficiency compared
to those in the low health status category.
Although the lysine requirement was increased in PCV2 vaccinated pigs when expressed on a
grams per day basis, there was no evidence of an influence when expressed as lysine grams to
energy ratio. Also, it appears that PCV2 vaccination has little effect on carcass back fat or loin
muscle depth (Jacela et al., 2011; Venegas-Vargas et al., 2011; Potter et al., 2012). This would
suggest that PCV2 vaccination has a limited effect on the composition of gain that could influence
feed efficiency. Therefore, the influence of PCVAD on feed efficiency again appears to be mostly
related to direct effects of mortality.
Future advances
Diagnostic tools and vaccines to assist in the management of infectious disease will continue to be
needed in multi-site swine production systems. Especially, low-cost, high-throughput diagnostic
tools and those that distinguish vaccine produced immunity compared to immunity from natural
infection will be useful for categorizing production data. The characterization of production data
like feed efficiency that are difficult to model in controlled research settings will provide further
understanding of the influence of health on feed efficiency. Also, development of these tools will
be important for understanding the economic impact as new pathogens emerge or reemerge.
Certainly, genetic selection to reduce fat and increase lean will continue to improve feed efficiency.
Additionally, genetic selection that improves disease resistance will be useful in mediating

direct effects of mortality improvements on feed efficiency. Fortunately, adding survival traits to
selection indexes and evaluation of off-spring survival under large scale commercial conditions
are becoming standard practices among breeding stock companies.
Conclusion
Health effects on feed efficiency due to chronic immune stimulation are primarily mediated
through the composition of gain when adequate amino acids are provided in the diet. The
increased feed and amino acid intake leads to an increased rate and efficiency of muscle growth.
Since muscle growth is more feed efficient compared to fat growth, this results in improvements
in feed efficiency. Further evidence is provided by the lack of difference in partial efficiency of
energy utilization for protein and lipid accretion between high and low health categorization.
Evidence indicates that in multi-site production systems, chronic immune stimulation is

reduced due to age segregation, leading to less pathogen transmission across age groups and
via the environment from batch to batch. Unfortunately, multi-site production systems have
not eliminated all pathogens leading to pathogen specific direct effects of mortality with fewer
effects mediated through metabolic alterations of survivors. Furthermore, it appears that non-
specific health-enhancing factors such as biosecurity and sanitation protocols to prevent lateral
transmission of infectious pathogens and reduced environmental contamination result in
considerable improvements in feed and financial efficiency.
References
Alexander, T. J. L., K. Thornton, G. I. Boon, R. J. Lysons, A. F. Gush. 1980. Medicated early weaning to obtain
pigs free from pathogens endemic in the herd of origin. Vet. Rec. 106:114-119.
Amass, S. F. and A. Baysinger. 2006. Swine disease transmission and prevention. in Diseases of Swine 9th
ed. B. E. Straw, J. J. Zimmerman, S. D’Allaire, and D. J. Taylor, eds. Blackwell Publishing Professional,
Ames, IA, USA.
Balaji, R., K. J. Wright, C. M. Hill, S. S. Dritz, E. L. Knoppel, and J. E. Minton. 2000. Acute phase responses
of pigs challenged orally with Salmonella typhimurium. J. Anim. Sci. 78:1885-1891.
Baracos, V. E., W. T. Whitmore, and R. Gale. 1987. The metabolic cost of fever. Can. J. Physiol. Pharmacol.
65:1248-1254.
Desrosiers, R., E. Clark, D. Tremblay, R. Tremblay, D. Polson. 2009. Use of a one-dose subunit vaccine to
prevent losses associated with porcine circovirus type 2. J. Swine Health Prod. 17:148-154.
Dritz, S. S., K. Q. Owen, R. D. Goodband, J. L. Nelssen, M. D. Tokach, M. M. Chengappa, and F. Blecha. 1996.
Influence of lipopolysaccharide-induced immune challenge and diet complexity on growth performance
and acute-phase protein production in segregated early-weaned pigs. J. Anim. Sci. 74:1620-1628.
Dritz, S. S., M. D. Tokach, R. D. Goodband, and J. L. Nelssen. 2002. Effects of administration of antimicrobials
in feed on growth rate and feed efficiency of pigs in multisite production systems. J. Am. Vet. Med.
Assoc. 220:1690-1695.
Escobar, J., T. L. Toepfer-Berg, J. Chen, W. G. Van Alstine, J. M. Campbell, and R. W. Johnson. 2006.
Supplementing drinking water with Solutein did not mitigate acute morbidity effects of porcine
reproductive and respiratory syndrome virus in nursery pigs. J. Anim. Sci. 84:2101-2109.
Harris, D. L. 2000. Mulit-stite Pig Production. Iowa State University Press, Ames, IA.

S.S. Dritz
Hays, V. W. 1978. Effectiveness of feed additive usage of antibacterial agents in swine and poultry production.
Report to the Office of Technology Assessment, U.S. Government Printing Office, Washington, D.C.
Horlen, K. P., S. S. Dritz, J. C. Nietfeld, S. C. Henry, R. A. Hesse, R. Oberst, M. Hays, J. Anderson, and R. R.
R. Rowland. 2008. A field evaluation of pig mortality, performance and infection following commercial
vaccination against porcine circovirus type 2. J. Am. Vet. Med. Assoc. 232:906-912.
Jacela, J. Y., S. S. Dritz, J. M. DeRouchey, M. D. Tokach, R. D. Goodband, and J. L. Nelssen. 2011. Field
evaluation of a PCV2 vaccine on finishing pig growth performance and mortality rate in a herd with a
history of porcine circovirus disease. J. Swine Health Prod. 19:10-18.
Klasing, K. C. 1988. Nutritional aspects of leukocytic cytokines. J. Nutr. 118:1436-1446.
Kyriazakis, I. and A. Doeschl-Wilson. 2009. Anorexia during infection in mammals: variation and its
sources. in Voluntary Feed Intake in Pigs. D. Torrallardona and E. Roura, eds. Wageningen Academic
Publishers, Wageningen, the Netherlands.
McOrist, S., and C. J. Gebhart. 2006. Proliferative enteropathies. in Diseases of Swine 9th ed. B. E. Straw, J.
J. Zimmerman, S. D’Allaire, and D. J. Taylor, eds. Blackwell Publishing Professional, Ames, IA, USA.
Madec, F., N. Bridoux, S. Bounaix, and A. Jestin. 1998. Measurement of digestive disorders in the piglet at
weaning and related risk factors. Prev. Vet. Med. 35:53-72.
Main, R. G., S. S. Dritz, M. D. Tokach, R. D. Goodband, and J. L. Nelssen. 2004. Increasing weaning age
improves pig performance in a multi-site production system. J. Anim. Sci. 82:1499-1507.
NASS. 2008. Hog County Estimates, Iowa, December 1, 2008. Accessed January 4, 2012. http://www.nass.
usda.gov/Statistics_by_State/Iowa/Publications/County_Estimates/reports/hogCty%20Est.pdf.
Neumann E. J., J. B. Kliebenstein, C. D. Johnson, J. W. Mabry, E. J. Bush, A. H. Seitzinger, A. L. Green, and
J. J. Zimmerman. 2005. Assessment of the economic impact of porcine reproductive and respiratory
syndrome on swine production in the United States. J. A. V. M. A. 227:385-392.
Neumann, E., S. Simpson, J. Wagner, and B. Karaconji. 2009. Longitudial field study of the effect of
commercial porcine circovirus type 2 vaccine on postweaning mortality in New Zealand farms. Journal
of Swine Health and Prod. 17:204-209.
NRC. 1998. Nutrient requirements of swine: 10th revised edition. Natl. Acad. Press, Washington, D.C.
Osorio F. 2010. PRRSV infections: a world-wide update (2010). Acta. Scientiae Veterinariae 38 (Supl
1):s269-s275.
Patience, J. F., H. W. Gonyou, D. L. Whittington, E. Beltranena, C. S. Rhodes and A. G. Van Kessel. 2000.
Evaluation of site and age of weaning on pig growth performance. J. Anim. Sci. 78:1726-1731.
Potter, M. L., L. M. Tokach, S. S. Dritz, S. C. Henry, J. M. DeRouchey, M. D. Tokach, R. D. Goodband, J. L.
Nelssen, R. R. R. Rowland, R. A. Hesse, and R. Oberst. 2012. Genetic line influences pig growth rate
responses to vaccination for porcine circovirus type 2. J. Swine Health Prod. 20:34-43.
Shelton, N. W., M. D. Tokach, S. S. Dritz, R. D. Goodband, J. L. Nelssen, J. M. DeRouchey, and J. L. Usry.
2012. Effects of porcine circovirus type 2 vaccine and increasing standardized ileal digestible lysine:ME
ratio on growth performance and carcass composition of growing and finishing pigs. J. Anim. Sci. 90:
361-372.
Toepfer-Berg, T. L., J. Escobar, W. G. Van Alstine, D. H. Baker, J. Salak-Johnson, and R. W. Johnson. 2004.
Vitamin E supplementation does not mitigate the acute morbidity effects of porcine reproductive and
respiratory syndrome virus in nursery pigs. J. Anim. Sci. 82:1942-1951.
Touchette, K. J., J. A. Carroll, G. L. Allee, R. L. Matteri, C. J. Dyer, L. A. Beausang, and M. E. Zannelli. 2002.
Effect of spray-dried plasma and lipopolysaccharide exposure on weaned pigs: I. Effects on the immune
axis of weaned pigs. J. Anim. Sci. 80:494-501.

Venegas-Vargas, M. C., R. Bates, and R. Morrison. 2011. Effect of porcine circovirus type 2 vaccine on
postweaning performance and carcass composition. J. of Swine Health Prod. 19:233-237.
Williams, N. H., T. S. Stahly, and D. R. Zimmerman. 1997a. Effect of level of chronic immune system
activation on the growth and dietary lysine needs of pigs fed from 6 to 112 Kg. J. Anim. Sci. 75:2481-
2496.
Williams, N. H., T. S. Stahly, and D. R. Zimmerman. 1997b. Effect of chronic immune system activation on
the rate, efficiency, and composition of growth and lysine needs of pigs fed from 6 to 27 Kg. J. Anim.
Sci. 75:2463-2471.
Williams, N. H., T. S. Stahly, and D. R. Zimmerman. 1997c. Effect of chronic immune system activation on
body nitrogen retention, partial efficiency of lysine utilization, and lysine needs of pigs. J. Anim. Sci.
75:2472-2480.
Wright, K. J., R. Balaji, C. M. Hill, S. S. Dritz, E. L. Knoppel, and J. E. Minton. 2000. Integrated adrenal,
somatotropic, and immune responses of growing pigs to treatment with lipopolysaccharide. J. Anim.
Sci. 78:1892-1899.
Zimmerman., D. R. 1986. Role of subtherapeutic levels of antimicrobials in pig production. J. Anim. Sci.
62(Suppl. 3):6-17.

12. Physiology of feed efficiency in the pig: emphasis on the
gastrointestinal tract and specific dietary examples
J.R. Pluske
School of Veterinary and Biomedical Sciences, Murdoch University, South Street, Murdoch WA
6150, Australia; J.Pluske@murdoch.edu.au
Abstract
A sound and robust structure of the gastrointestinal tract (GIT) with appropriate physiological
functions is critical during the various life phases of pig growth, to ensure optimum health of
the GIT that in turn ensures optimum pig production. This is especially the case in the period
immediately following weaning, when there are a number of abrupt simultaneous insults
imposed on young pigs that cause temporal changes to the structure and function of the GIT.
These alterations contribute to the well-described post-weaning ‘growth check’, and although
it is reasonably well understood what causes these changes, continuation of the ‘growth check’
in commercial pig production around the world continues largely unabated. Given the close
connection between the physiology of the GIT and pig performance and health, this chapter
focuses on a number of influences, before and after weaning, that influence the structure and
function of the GIT that in turn impact feed efficiency. This chapter, as with others described
in this book, demonstrates that the physiology of the GIT, and indeed the entire pig, is the
cornerstone for improved feed efficiency.
Introduction
Differences in feed efficiency in pigs arise primarily as a consequence of variation in the factors
influencing growth and development through the animal’s life, such as body composition, feeding
patterns, management, genetics, the efficiency of digestion and absorption, the level of physical
activity, dietary factors (e.g. quality, effects of processing, energy to amino acid ratios), disease/
immune function, thermoregulation and tissue metabolic rates (Lkhagvadorj et al., 2010). The
pig’s ‘physiology’ (including the physiology associated with behavior) is the scientific principle
underpinning and (or) explaining the mechanisms behind these factors. Each of these factors/
subjects is covered by other authors in this book and is therefore outside the scope of my
particular chapter. Given this, I have decided to focus specifically on a number of factors that
influence the efficiency of gastrointestinal tract (GIT) function through effects on physiology as I
reason this would be an appropriate contribution to this book. Specifically, I will pay attention to
influences of secretions from the sow on the efficiency of piglet growth before and after weaning,
consequences of the weaning process in relation to feed efficiency, and some relationships
between diet composition, post-weaning diarrhea and feed efficiency.
Secretions from the lactating sow and piglet growth efficiency
It is very well recognized that insufficient colostrum intake is one of the major causes of neonatal
mortality in pig production. Approximately 50% of pre-weaning piglet mortality occurs within

J.R. Pluske
3 days after birth (Tuchscherer et al., 2000) and mainly affects piglets characterized by low birth
weight and low weight gain, which in turn can be related to the failure of a piglet to achieve a
regular and sufficient intake of colostrum (De Passillè and Rushen, 1989). After birth, the piglet
ingests first colostrum, and then milk, both of which are high in fat and low in carbohydrate,
implying that the intestine must be functional at the time of birth and the piglet is rapidly able
to synthesize glucose by gluconeogenesis to supply its glucose-dependent tissues, and to oxidize
fats (Le Dividich et al., 2005). In this context the ingestion of colostrum, which provides both
energy and maternal antibodies that protect the piglets until their own immune system matures,
is of the utmost importance for survival while also being the continuation of the mother–infant
bond (Le Dividich et al., 2005).
The neonatal piglet, therefore, needs energy to meet its requirements for maintenance, including
thermoregulation and physical activity, and growth. Requirements for energy and nutrients,
expressed on a body-weight (BW) basis, are maximal at birth. Le Dividich et al. (1994) and
Marion and Le Dividich (1999) estimated that under conditions of thermal neutrality and
minimum energy expenditure (associated with feeding and physical activity), the energy required
for maintenance is 66 kcal/kg BW. Le Dividich et al. (1998) calculated, however, that the newborn
pig needed extra energy required for thermoregulation amounting to 0.48 kcal/kg BW/h/°C
below the lower critical temperature. Overall and in thermoneutral conditions, Le Dividich et
al. (2005) determined that the minimum net energy requirement for survival of a 1.0 kg piglet at
birth is approximately 167 kcal during the first postnatal day, and may be as high as 215-227 kcal
in moderately cold conditions (Figure 1).
1000
Fat
Glycogen
800 Thermoregulation
Energy retention
Energy (kJ/kg)
600 Physical activity

Maintenance
400
200
0
Energy available Thermal neutrality Below thermal
at birth neutrality
Figure 1. Energy reserves available at birth (fat; glycogen) and estimated net energy requirements during
the first 24 h of life of piglets surviving to weaning in conditions of thermal neutrality or 5 °C below thermal
neutrality (maintenance; physical activity; energy retention; thermoregulation) (taken from Le Dividich et
al., 2005).

12. Physiology of feed efficiency in the pig
Colostrum intake, therefore, plays a major role in piglet development by providing energy for
thermoregulation (Le Dividich et al., 2005), enabling antibody immune transfer from the sow
(Rooke and Bland, 2002), and in stimulating GIT development (Xu, 1996). Clearly the digestibility,
absorption and overall efficiency of use of the colostrum, as well as the milk that appears from 24-
36 h after birth, must be very high to permit these functions. Lin et al. (2009) determined that the
apparent fecal digestibilities of crude protein (CP) and dry matter (DM) in colostrum, defined as
the percentage of ingested colostral crude proteins and DM that disappeared in the gut, averaged
96.9±0.4% and 98.3±0.2%, respectively. Digestibility of total amino acids (protein-bound plus
free amino acids) in the colostrum was 98.3±0.1%, with the values being 98.5±0.3, 98.2±0.4,
and 98.3±0.3%, respectively, for lysine, threonine and arginine, confirming that protein-bound
and free amino acids in porcine colostrum were highly digestible and available to neonatal pigs
(Table 1). Earlier, the true digestibility values for lysine, threonine and arginine in mature milk
for 17-d old pigs was found to be 92.1, 84.3, and 90.4%, respectively (Mavromichalis et al., 2001),
which were lower than the digestibility of amino acids in colostrum obtained by Lin et al. (2009).
However, considering that the digestibility obtained in their study was an apparent value, the true
digestibility of proteins and DM in porcine colostrum would be essentially 100%.
In another study demonstrating the excellent biological value of sows’ colostrum and milk, Le
Dividich et al. (2007) used 24 newborn piglets to evaluate the digestibility of sow colostrum and
milk and the efficiency of milk utilization by the piglet at day 8 of lactation. Milk energy digestibility
and metabolizability (ME/GE × 100) and nitrogen digestibility were 98.2±1.2 (SEM), 96.8±1.4 and
98.3±1.3%, respectively. Corresponding values for colostrum were statistically lower, averaging
95.2±2.8, 92.6±3.1 and 95.3±2.9%, respectively. The efficiency of using milk metabolizable energy
(ME) for energy retention (determined by indirect calorimetry or comparative slaughter) was
Table 1. Apparent digestibility (%) of dry matter (DM) and amino acids in porcine colostrum by neonatal pigs
(Lin et al., 2009).
Item Mean±SE Items Mean±SE
DM 98.3±0.2 CP2 96.9±0.4

Amino acid1
Ala 98.1±0.4 Lys 98.5±0.3
Arg 98.3±0.3 Met 98.5±0.2
Asp+Asn 98.1±0.3 Phe 98.1±0.3
Cys 98.4±0.3 Pro 98.7±0.2
Glu+Glu 98.8±0.2 Ser 98.2±0.6
Gly 97.9±0.4 Thr 98.2±0.4
His 98.6±0.2 Trp 98.3±0.2
Ile 98.3±0.3 Val 98.5±0.4
Leu 98.5±0.3
1 Data are means±SE, n=12.
2 DM basis.

J.R. Pluske
similar and averaged 0.72±0.02. The energy cost of 0.24 kcal of protein deposition was 0.42
(±0.009) kcal (efficiency, 0.56), whereas the energy cost of 0.24 kcal of fat deposition was not
different to 0.24 kcal. These data are in accord with those found previously by Le Dividich et al.
(1994) and Marion and Le Dividich (1999), with the efficiency of ME utilization for total energy
retention and for energy retained as protein being higher for colostrum than for milk (0.91 vs.
0.72; 0.90 vs. 0.56, respectively).
In summary, it is evident that the colostrum and milk of sows have evolved to provide nutrients
to ensure the survival, growth and health of neonatal pigs. Indeed, the excellent efficiency of
utilization of milk has, for many years, been the basis of feeding cows’ whole milk to pigs both
before and after weaning (Pluske et al., 1995). For example, feeding ewes’ or cows’ whole milk
to 28-day-old weaned pigs for 5 days after weaning caused exemplary growth rates and feed
efficiencies (Pluske et al., 1996a,b) that simply cannot be matched by feeding contemporary solid
diets after weaning.
Growth factors in colostrum and milk
Like any mammalian secretions, the colostrum and milk produced by sows also contain an array
of bioactive compounds (e.g. hormones, growth factors, proteins, enzymes, neutrophils and
lymphocytes) that are crucial for not only protection and resistance against infections, but also for
tissue growth and maturation of the intestinal epithelial mucosa. There are numerous excellent
papers and reviews pertaining to this subject (e.g. Cera et al., 1987; Kelly, 1994; Xu, 1996; Pluske
et al., 1997; Burrin and Stoll, 2002; Xu et al., 2002; Van Barneveld and Dunshea, 2011) and their
contents will not be repeated here. Nevertheless, and because of the well-documented effects of
these bioactive compounds on the physiological development and maturation of the GIT, there
has always been interest in the use of colostrum to enhance the survival of the newborn piglet;
for example, artificial colostrum products based on freeze-dried colostrum from dairy cows have
been available to pig producers for many years.
The use of bovine colostrum has also attracted some attention as a means to stimulate GIT
growth and development and (or) ameliorate the deleterious changes that occur in the GIT of
the pig after weaning (Pluske et al., 1997), which individually or collectively could enhance feed
efficiency when included in a diet for young pigs. Growth factors in bovine colostrum include
insulin-like growth factors-I and-II (IGF-I and IGF-II), transforming growth factor-β1 and -β2
(TGF-β1 and TGF-β2) and epidermal growth lysozyme, lactoperoxidase, immunoglobulins (Igs)
and cytokines (IL-1β, IL-6, TNF-α, IFN-γ and IL-1ra) (Pakkanen and Aalto, 1997; Boudry et al.,
2008). These compounds in the colostrum of diary cows are similar to those found in the milk
of the sow and although they might have somewhat different physiological effects in vivo due to
animal specificity, some effects are likely to be the same.
Although I will discuss the changes that occur in the GIT after weaning more extensively later in
this review, it is pertinent to mention here that a recent review by Boudry et al. (2008) summarizes
studies on the use of bovine colostrum in diets in the immediate post-weaning period, in terms
of both bovine colostrum’s effects on prevention/repair of the GIT damage caused after weaning
and (or) an improvement in production indices. These authors reported, from the studies they

reviewed, that the beneficial effects observed could be explained by both an increase in feed
intake level during the first days following weaning and a likely direct effect on GIT health,
through preservation of GIT integrity and a reduction of the coliform population. Although
the improvements in production were not all necessarily related to an improvement in feed
efficiency, a study by Le Huërou-Luron et al. (2004; cited by Boudry et al., 2008) found that feed
efficiency was improved by 10% when pigs were kept in an ‘unclean’ environment compared to
other studies performed in cleaned rooms. Le Huërou-Luron et al. (2004; cited by Boudry et al.,
2008) concluded that bovine colostrum had a more positive effect in less sanitary conditions as
has been observed with plasma proteins (e.g. Van Dijk et al., 2001). Boudry et al. (2008) suggested
that the positive response of bovine colostrum might result from an improvement in the sanitary
status of the colostrum-treated piglets by a direct effect of the colostrum on GIT health.
As mentioned previously, bovine colostrum contains an extensive array of (characterized and as

yet uncharacterized) bioactive compounds that can modulate the GIT and potentially impart an
improvement in feed efficiency. One compound that has seen a reasonable amount of research
is epidermal growth factor (EGF) that in addition to its growth-promoting effect, appears to be
able to modulate enterocyte differentiation during the transition phase from maternal milk to
solid food in weaned piglets (Schweiger et al., 2003). In 21-day-old newly-weaned piglets, orally
administered EGF (372 μg per day) increased the specific activities of jejunal lactase (77%) and
sucrase (97%) measured after 3 days of feeding (Jaeger et al., 1990). Moreover, Kingsnorth et al.
(1990) reported increased tensile strength of gastric wounds in 20-kg pigs after 5 days of intra-
peritoneal infusion of EGF (0.5 μg per kg per day). These results suggest that supplementation
with EGF may aid in the recovery of damaged GIT tissues, and in support, Playford et al. (2000)
showed that a continuous infusion of EGF resulted in a remarkable restorative effect on gut
histology in a child with necrotizing enterocolitis (NEC) (Figure 2).
In summary, I have used the example of bovine colostrum as a product that may improve feed
efficiency either directly, for example by an increase in utilization of nutrients by the GIT, and
(or) by assisting the recovery process of the GIT after weaning that, as I explain in more detail
below, remains a major obstacle to enhanced feed efficiency following the weaning process. In
many ways it is similar in its physiological actions to spray-dried porcine plasma in its effects but
is obviously richer in some GIT-modulating compounds.
Changes in feed efficiency associated with weaning
Consequences of the weaning process
The weaning period is a critical phase in the young pig’s life. Under natural or semi-natural
conditions, weaning is a progressive process occurring between 12 and 17 weeks of age, but
in modern pig husbandry, weaning occurs abruptly at 3 to 4 weeks of age, inducing numerous
stressors on piglets. It therefore represents a period of nutritional, physiological and environmental
perturbation. As summarized for example by Pluske et al. (1997) and Boudry et al. (2008), amongst
many others, intestinal post weaning-induced changes occur in two successive periods: first,
a transitory anorexia-dependent alterative period that results in decreased absorptive surface,
modified epithelial barrier function, delayed adaptation of digestive enzymes, increased risks

J.R. Pluske
A B
Figure 2. Hematoxylin and eosin stain of small intestinal biopsies of a child with necrotizing enterocolitis
(200× magnification) before (A) and 7 d after (B) infusion of EGF. Before therapy the mucosa is virtually
completely ulcerated and after therapy the mucosa is almost completely regenerated (Playford et al., 2000).
of misbalance of the microbiota, and local inflammatory response triggering by enhanced pro-
inflammatory cytokine production; and second, a restorative period that lasts 1-2 weeks before
complete adaptation to the new diet. Burrin and Stoll (2003) similarly described these periods
as the ‘acute’ and ‘adaptive’ phases, respectively, that are associated with the general malaise that
follows weaning (Figure 3).
Brooks and Tsourgiannis (2003) reviewed a number of studies and surmised that although the
majority of pigs start to eat solid feed within 5 h after weaning, some of them take up to 54 h
before eating their first meal. Thereafter, ME intake gradually increases at a rate of approximately
24-29 kcal ME per kg0.75 per day. Regardless of the age of weaning, the level of ME intake attained
at the end of the first week after weaning tends to range between 167 and 191 kcal per kg0.75,
which accounts for 60-70% of the pre-weaning milk ME intake (Pluske et al., 1995; Boudry et
al., 2008). This reduction of ME intake results in a critical period of underfeeding (Le Dividich
and Sève, 2000), and many of the acute modifications of intestinal functions (described below)
are largely explained by the variation in feed intake that occurs (Pluske et al., 1997). This abrupt

300 200
Protein mass
Percent of pre-weaning
250
Percent of pre-weaning
DNA mass
150
200
Villus height
150 Crypt depth 100
Cell proliferation
100
50 50
Acute phase Adaptive phase Acute phase Adaptive phase
1 5 10 15 1 5 10 15
Days post-weaning Days post-weaning
Figure 3. The ‘acute’ and ‘adaptive’ phases associated with weaning the young pig (Burrin and Stoll, 2003).
reduction in voluntary feed intake immediately following weaning causes a ‘growth check’, and
its severity has a major impact on subsequent performance (e.g. Tokach et al., 1992).
Linked to low and erratic feed intake is bodyweight loss, and on the first day after weaning, pigs
lose anywhere from 100 to 250 g of body weight (Pluske et al., 1997; Le Dividich and Sève, 2000).
This loss of bodyweight is dependent on the age of weaning, with pigs weaned younger losing
proportionally more weight (Carroll et al., 1998). In another study addressing some of these
changes, Pluske et al. (2003) determined the interrelationships between sex, weaning age, and
weaning weight on aspects of physiological and gastrointestinal development in pigs in a factorial
arrangement with the respective factors being age at weaning (14 or 28 days), weight at weaning
(heavy or light), sex (boar or gilt), and time after weaning (1, 7, and 14 days). These authors found
that pigs weaned at 2 weeks of age and pigs weaned light-for-age at either 2 or 4 weeks had a less
developed gastrointestinal tract, and that its development after weaning might proceed differently
to that of pigs weaned older and heavier.
This period of transient anorexia, often followed by a short period of gorging, can lead to gut
dysfunction, increased sensitivity to enteric infections, and in some cases diarrhea and (or) death.
The most consistent and well characterized (patho) physiological changes affect the anatomy
and function of the small intestine (reviewed by Pluske et al., 1997), that in turn are likely to
influence feed conversion efficiency. In a comprehensive study of changes in the structure of the
small intestine in the weaned pig that pre-empted the figure depicted by Burrin and Stoll (2003;
Figure 3), Hampson (1986) reported that following weaning at 21 days of age, the villous height
was reduced to around 75% of pre-weaning values within 24 hours and subsequent reductions
along the small intestine were smaller but continued to decline until the fifth day after weaning, at
which point the villous height at most sites along the gut was approximately 50% of initial values
found at weaning. From five to eight days after weaning the villous height began to increase. In
contrast, unweaned pigs showed only slight reductions in villous height. Hampson (1986) also
reported that the number of cells in the crypts was not increased two days after weaning, but
increased steadily thereafter until the eleventh day. Crypt elongation (hyperplasia) also occurred

J.R. Pluske
in unweaned pigs, but the increase was greater in weaned animals and seemingly required the
consumption of food after weaning to occur. Importantly, these changes only occurred after
weaning if there was continuous absence from the sow, since pigs weaned for two days and then
returned to the dam for three days showed crypt elongation only equivalent to that of pigs weaned
for two days (Hampson, 1983). Furthermore, it appears that the lack of feed intake may not have
been necessary for villous atrophy to occur (Hampson, 1983).
In this regard, Lallès et al. (2004) reported a 20-30% reduction in mucosal weight associated
with villous atrophy, and numerous further studies have demonstrated that intestinal barrier
function is compromised, resulting in increased secretion of electrolytes and water and increased
permeability to potentially toxic substances (reviews by Vente-Spreeuwenberg and Beynen, 2003;
Lallès et al., 2004). In addition to alterations in gut physiology, weaning causes disturbances in
the homeostasis of the gut microbiota and delays the anatomical and functional development of
the mucosal immune system (review by Lallès et al., 2007).
An often unintended consequence of the post-weaning malaise can be post-weaning diarrhea

(PWD; Fairbrother et al., 2005), which is a condition in weaned pigs characterized by frequent
discharge of watery feces during the first two weeks following weaning, and can at times be a
major economic loss to the pig industry. Post-weaning diarrhea is a multifactorial disease and
its precise pathogenesis is still unclear. However it is typically associated with faecal shedding of
β-haemolytic enterotoxigenic E. coli (ETEC) serotypes that particularly proliferate in the small
intestine of both healthy and unhealthy piglets after weaning (Hampson, 1994). The adhesion
of ETEC to receptors in the small intestine normally results in their colonisation of the GIT,
which in turn enables them to produce one or more enterotoxins such as heat labile toxins (LT)
or heat stable toxins (ST; variants STa and STb). These enterotoxins activate cyclic guanosine
monophosphate (cGMP) and cyclic adenosine monophosphate (cAMP) systems with LT toxins
increasing secretion of sodium, chloride and hydrogen carbonate ions into the lumen, whilst
the ST toxins reduce the absorption of liquid and salts (Hampson, 1994). Both cases cause
hypersecretion of water and electrolytes into lumen that generally exceeds the absorptive capacity
of the colon, that then causes diarrhea, dehydration, reduced feed intake, reduced nutrient
digestibility, reduced growth and (or) even death (Pluske et al., 2002).
Prevention/control of PWD commercially has generally occurred, and continues to occur in

some parts of the world, by the use of antimicrobial substances such as antibiotics and (or)
mineral compounds (e.g. zinc oxide; ZnO) in the feed and (or) water supply. However these
compounds have been banned or severely restricted in their use in some pig-producing countries/
regions because of the survival of resistant bacterial species or strains. Genes encoding for this
resistance also can be transferred to other formerly susceptible bacteria, thus posing a threat
to both animal and human health (Barton, 2000; Hampson et al., 2001). Consequently, some
countries have banned (Sweden; in January 1986) or limited (European Union; in January 2000,
total withdrawal in January 2006) the general use of in-feed antibiotics as growth promotants
in animals (De Lange et al., 2010; Kil and Stein, 2010; Pluske et al., 2002, 2007). In the absence
of such antimicrobial compounds, attention has focused on feed additives/dietary intervention
strategies, or a combination of both, to reduce ETEC growth and proliferation to minimize the

expression of PWD. In this sense, both the source and level of dietary protein are known to
influence PWD in young pigs.
Relationships between dietary protein source and post-weaning diarrhea
It has long been recognized that certain plant protein sources can negatively impact on the growth
and GIT health of pigs after weaning. The pioneering work of Miller et al. (1984) showed the
effects of feeding soybean meal (SBM) on gut structure and function, and Li et al. (1990; 1991)
found that feeding soybean meal (SBM) as opposed to feeding dried skim milk powder similarly
caused transient hypersensitivity in early weaned pigs. In these studies, pigs fed the SBM diet
had a lower rate of gain and inferior feed efficiency, shorter villi, and a higher immunoglobulin
titre to SBM indicating the antigenic property of SBM. Earlier, Shimizu and Terashima (1982)
observed an increased incidence of diarrhea and appearance of ETEC in the feces of pigs offered
SBM compared to skim milk powder. Nevertheless there is some conflicting evidence regarding
the impacts of feeding SBM or milk powders on PWD (Wellock et al., 2008).
These pronounced morphological changes (reviewed by Lallès et al., 1993) led to the development
of a number of processing techniques for plant protein sources, e.g. extrusion, fermentation, to
effect an amelioration of the anti-nutritional factors contained in the grains with a subsequent
improvement in their nutritive value and feed efficiency. Nevertheless, animal protein sources
appear to have superior feeding value to plant protein sources when fed to weaned pigs partly
due to the fact that proteins of plant origin are less digestible than animal proteins. Although
long-term effects to slaughter weight may be negligible or even non-existent, feeding whey
protein concentrate and fish meal to pigs after weaning for example resulted in better growth
performance, and enhanced nutrient digestibility and indices of GIT morphology, compared with
feeding SBM, fermented soy protein, and rice protein concentrate (Yun et al., 2005).
Feeding plant or animal protein sources after weaning is also likely to alter the physiological
function of the GIT, including the composition and (or) number of microbiota in the GIT, which
in turn might influence the expression and (or) incidence of diarrhea (Heo et al., 2012). This is
then likely to reduce feed efficiency during the period in which pigs succumb to the disease.
However the levels of fermentable/non-fermentable carbohydrate and protein also require careful
consideration (Jeaurond et al., 2008). Non-pathogenic and pathogenic bacteria such as ETEC
utilize protein in the GIT as a nutrient source (Nollet et al., 1999), and thus manipulation of
the dietary crude protein (CP) level has been suggested as a nutritional strategy for reducing
PWD (Ball and Aherne, 1987). Diets for newly-weaned pigs generally contain between 170 and
240 g per kg of CP, with the CP level being determined by either diet cost (achieved by least-cost
formulation) or the availability (or lack of) of suitable antimicrobial additives; diets in Europe
for example tend to be lower in CP level than those in Australia or North America due to bans/
restrictions on the use of growth promoting antibiotics and mineral compounds. Under certain
conditions, feeding excessive dietary protein can increase the ileal protein flow (especially in
association with ETEC infection; Heo et al., 2010a), and consequently more dietary protein
enters the distal GIT materials where it is available for fermentation by the resident microbiota
(Heo et al., 2010a). This could contribute to an imbalance between the growth of proteolytic
(protein digesting) and saccharolytic (carbohydrate digesting) microbiota in the large intestine.

J.R. Pluske
Microbial fermentation of the undigested dietary protein can provoke PWD by contributing to an
increased production of potentially toxic by-products such as branched-chain fatty acids, indole,
phenols, ammonia and biogenic amines in the GIT. Indeed, scientific and practical experience
indicates that lowering the CP content of the diet is generally associated with reduced PWD
(summarized by Heo et al., 2012; see also Figure 4) and, under conditions of ETEC infection and
with appropriately formulated diets (see below), no deterioration in the feed conversion efficiency
(Heo et al., 2008; 2009; Table 2).
Although feeding lower protein diets can reduce scouring and improve fecal consistency after
weaning (Le Bellego and Noblet, 2002), growth can be reduced and feed efficiency worsened
when one or more of the indispensable amino acids are present in concentrations below the
requirement for optimum growth and feed conversion efficiency (Kil and Stein, 2010; Heo et
al., 2012). This is exacerbated under conditions of PWD. In contrast, pigs receiving a lower-
protein diet formulated to an ideal essential amino acid (AA) ratio (Chung and Baker, 1992) with
supplemental crystalline essential AA showed both significantly reduced expression of PWD and
comparable growth performance (growth rate, feed intake and feed conversion efficiency) to pigs
fed a higher protein diet (Heo et al., 2008, 2009, 2010a; Lordelo et al., 2008).
Collectively, these findings indicate that feeding a lower-protein diet may not hinder the growth
of pigs after weaning if pigs received similar digestible energy contents and adequate levels of
essential AA. However, in one contrasting study, Opapeju et al. (2009) found that pigs fed lower-
protein diets (decreasing from 210 to 170 g/kg CP) decreased growth performance although the
low-protein diet was supplemented with essential AA including Ile and Val for a 3-week period
after weaning. Opapeju et al. (2009) surmised that the use of early-weaned piglets and a shortage
of indispensable AA in the lower-protein diet might have caused the inferior performance.
HP LP5 LP7 LP10 LP14

50
40
Diarrhoea index
30
20
10
0
1-3 4-7 8-9 10-12 13-14
Days after weaning
Figure 4. The effect of feeding a high protein (HP) diet or low protein (LP) diets fed for either 5, 7, 10 or 124 days
after weaning, on the diarrhea index for 2 weeks after weaning. Diarrhea index is expressed as a proportion
of days with diarrhea with respect to total number of days (14 d). *P<0.05, **P<0.01 (Heo et al., 2008).

Table 2. The effect of feeding regimen and experimental β-hemolytic enterotoxigenic strains of Escherichia
coli infection on growth performance in weaned pigs1 (Heo et al., 2009).
Period2 Noninfected3 Infected3 Pooled P-value4

SEM
HP14 LP7 LP14 HP14 LP7 LP14 FR I FR×I
ADG (g)
d 1 to 14 129a 113a 150a 55b 63b 63b 7.0 0.304 <0.001 0.414
d 15 to 28 435a 412a 453a 352b 334b 296b 15.5 0.810 0.003 0.601
d 1 to 28 282a 301a 263a 204b 179b 199b 9.9 0.673 <0.001 0.524
ADFI (g)
d 1 to 14 192 186 235 196 206 189 9.4 0.374 0.692 0.321
d 15 to 28 562 461 581 501 486 418 33.8 0.598 0.391 0.464
d 1 to 28 377 408 324 349 304 346 25.9 0.598 0.354 0.379
G:F (g/g)
d 1 to 14 0.67a 0.63a 0.63a 0.28b 0.31b 0.34b 0.027 0.958 <0.001 0.775
d 15 to 28 0.79a 0.83a 0.79a 0.70b 0.71b 0.70b 0.020 0.408 0.023 0.269
d 1 to 28 0.76a 0.85a 0.74a 0.58b 0.58b 0.59b 0.030 0.518 0.003 0.677
a,b Means in the same row with different superscripts differ (P<0.05).
1 Pigs in infected group were experimentally challenged at 72, 96, and 120 h after weaning. Values are the mean
of 4 replicates for noninfected treatments (2 rooms with 6 pens per room) and 2 replicates for infected treatments
(1 room with 6 pens).
2 Abbreviations used: ADG = average daily gain; ADFI = average daily feed intake; G:F = Gain:Feed ratio.
3 HP14 = high protein diet fed for 14 d; LP7 = low protein diet fed for 7 d; LP14 = low protein diet fed for 14 d, and at
the conclusion of each feeding regimen, all pigs then received a second-phase diet until 4 wk after weaning.
4 FR = feeding regimen effect; I = infection effect.
In summary, and as advocated by Kil and Stein (2010), if no antibiotic growth promotants are
included in the immediate post-weaning diets, then it may be necessary to formulate diets
containing less than 180 g CP per kg to avoid scouring and intestinal malfunctions. In such diets,
it may be required that six crystalline amino acids, (lysine, methionine, threonine, tryptophan,
isoleucine, and valine) are included to meet the requirement for all amino acids (Heo et al.
2008, 2009). If problems with diarrhea persist at this level of crude protein, it may be necessary
to reduce CP levels even lower. This strategy, or feeding lower CP diets without balancing for
an ideal AA ratio, will usually cause slight reductions in growth rate and a deterioration in feed
efficiency, but pigs will compensate when they are later provided with adequate levels of protein
(e.g. Stein and Kil, 2006; Kim et al., 2011).

J.R. Pluske
ZnO as a growth-promoting compound to enhance feed efficiency after

weaning
Zinc (Zn) is an essential trace mineral (micronutrient) for pigs where it is a component of
metalloenzymes, transferases and many digestive enzymes, and is associated with effective
insulin action. Furthermore, Zn plays crucial roles in lipid, protein and carbohydrate metabolism.
Although the recommended dietary level of Zn for weaner pigs is 100 ppm (NRC, 1998),
pharmacological levels (up to 3,000 mg/kg of zinc in the diet) have been used as an effective
dietary tool to prevent and (or) ameliorate PWD, thereby acting as a ‘growth promoter’ after
weaning (Pluske et al., 2007). The exact mode(s) of action for these positive effects of ZnO
after weaning have not been fully elucidated but, and somewhat surprisingly given the efficacy
of the compound, continues to attract research attention. This is most probably because the
compound is very effective and the scientific process is occurring to, unravel its mechanism(s)
of action, possibly leading to the use/development of other efficacious compounds to mitigate
against the post-weaning malaise. The use of ZnO has the disadvantage that it has been banned
at pharmacological levels for environmental reasons in some countries.
Numerous reviews have been conducted summarizing the purported mechanisms of the positive
effects of ZnO (Poulsen, 1995; Pluske et al., 2002, 2007; Li et al., 2010; Heo et al., 2012). Reported
effects include increased gene expression of antimicrobial peptides in the small intestine, positive
effects on the stability and diversity of the microbiota particularly with respect to coliforms,
increased insulin-like growth factor-I and insulin-like growth factor-II expression in the small
intestinal mucosa, bactericidal functions, and reductions in electrolyte secretion in vitro from
enterocytes. However, clear evidence associating ZnO with physiological changes in the GIT
and then linking this to improved performance is not always available (Hedemann et al., 2006).
More recently, and to test an alternative physiological hypothesis, Ou et al. (2007) examined
the roles of intestinal-mucosal mast cells, whose maturation and proliferation is under the
control of stem cell factor (SCF), in the etiology of diarrhea by releasing histamine. Mast cells
contribute to the pathogenesis of diarrhea through the production and release of histamine,
prostaglandins, leukotrienes, 5-hydroxytryptamine, and tumor necrosis factor-α with available
evidence suggesting a critical role for mast-cell-derived histamine in the pathogenesis of diarrhea.
Ou et al. (2007) found that 3,000 mg/kg ZnO reduced the incidence of PWD in weanling pigs in
association with reductions in the expression of the SCF gene at both mRNA and protein levels,
the number of mast cells in the mucosa and submucosa of the small intestine, and histamine
release from mucosal mast cells (Figure 5). These findings may have important implications for
the discovery of alternative mechanisms in the prevention of weaning-associated diarrhea in
piglets, and highlight the importance of basic physiological research to address industry problems.
A consequence of using ZnO in diets after weaning, achieved by one or more of the physiological
mechanisms just outlined, is improved feed conversion efficiency. Heo et al. (2010b,c) conducted
a study designed to examine the interactive effects of dietary protein level, zinc oxide (ZnO)
supplementation and infection with ETEC on performance responses and gastrointestinal tract
characteristics. Ninety-six individually housed, 21-day-old weaned pigs were used in a split plot
experiment with the whole plot being challenge or no challenge with ETEC and the dietary
treatments used as subplots and arranged in a completely randomised 2×2 factorial design, with

70
60
Histamine release (%) 50 * *

40
30
20
10
0
0 µmol 25 µmol 50 µmol 100 µmol
ZnO ZnO ZnO ZnO
Figure 5. The effect of different levels of ZnO on concanavalin A-induced histamine release from piglet
intestinal-mucosal mast cells. * = P<0.05 vs. the 0 μmol ZnO treatment (Ou et al., 2007).
the factors being (1) two dietary protein levels (251 versus 192 g/kg crude protein); and (2)
addition or no addition of 2,500 mg/kg ZnO. No antibiotic was added to the diet. Data from
this experiment convincingly show first, the significant deterioration in feed conversion that
occurs due to an enteric infection and second, the improvement in feed conversion efficiency
that occurred from feeding ZnO.
Conclusions
The examples provided in this chapter demonstrate the intimate associations that exist between
dietary influences and the physiological functioning of the GIT, both before and after weaning,
on feed efficiency. It is evident that secretions from the sow in lactation provide the most efficient
nutrient source for the young pig, but the withdrawal of milk at weaning induces a number of
morphological and physiological changes in the GIT that perturb the homeostasis of this organ.
Continuation of feeding milk to pigs following weaning at an ad libitum level of intake, as shown
in Figure 6 (below), maintains the structure and function of the small intestine that in turn
promotes optimum feed conversion efficiency and therefore growth, in this case measured as
empty bodyweight gain. The challenge for nutritionists and physiologists working with post-
weaning pigs offered solid (non milk-based) diets is to find the correct combination of the
nutritional and management tools and strategies available to lessen the negative impacts weaning
has on all aspects of physiology of the GIT.

J.R. Pluske
Mean villous height (µm) 700
600
500
400
300
200
0 100 200 300 400 500 600
Dry matter intake (g/d) maintenance

b 2.5 × maintenance
ad libitum
600
Empty body-weight gain (g/d)
400
200
-200
300 400 500 600 700
Mean villous height (µm)
Figure 6. Relationships between (a) daily dry matter intake and mean villous height along the small intestine
[y = 286.10+0.54x, R2=0.68 (RSD=56.91); P<0.001], and (b) mean villous height along the small intestine and
daily empty body-weight gain [y = – 325.69 + 1.39x, R2=0.48 (RSD=127.10); P=0.002], in pigs offered cows’
liquid milk at maintenance, 2.5 times maintenance, or ad libitum energy intake for five days after weaning
(8 pigs per treatment). Each point represents a single piglet killed on the fifth day following weaning at 28
days of age (Pluske et al., 1995).

References
Ball, R. O., and F. X. Aherne, F. X. 1987. Influence of dietary nutrient density, level of feed intake and weaning
age on young pigs. II. Apparent nutrient digestibility and incidence and severity of diarrhea. Can. J.
Anim. Sci. 67:1105-1115.
Barton, M. D. 2000. Antibiotic use in animal feed and its impact on human health. Nutr. Res. Rev. 13:279-
299.
Boudry, C., J.-P. Dehoux, D. Portetelle, and A. Buldgen. 2008. Bovine colostrum as a natural growth promoter
for newly weaned piglets: a review. Biotechnol. Agron. Soc. Environ. 12:157-170.
Brooks, P. H. and C. A. Tsourgiannis. 2003. Factors affecting the voluntary feed intake of the weaned pig.
Pages 81-116 in Weaning the Pig: Concepts and Consequences. J. R. Pluske, J. Le Dividich, and M. W.
A. Verstegen, eds. Wageningen Academic Publishers, Wageningen, the Netherlands.
Burrin, D. G, and B. Stoll. 2002. Key nutrients and growth factors for the neonatal gastrointestinal tract.
Clin. Perinatol. 29:65-96.
Burrin, D. G., and B. Stoll. 2003. Intestinal nutrient requirements in weanling pigs. Pages 301-335 in
Weaning the Pig: Concepts and Consequences. J. R. Pluske, J. Le Dividich, and M. W. A. Verstegen, eds.
Wageningen Academic Publishers, Wageningen, the Netherlands.
Carroll, J. A., T. L. Veum, and R. L. Matteri. 1998. Endocrine responses to weaning and changes in post-
weaning diet in the young pig. Domest. Anim. Endocr. 15:183-194.
Cera, K., D. C. Mahan, and F. A. Simmen. 1987. in vitro growth-promoting activity of porcine mammary
secretions: initial characterization and relationship to known peptide growth factors. J. Anim. Sci.
65:1149-1159.
Chung, T. K., and D. H. Baker. 1992. Ideal amino acid pattern for 10-kilogram pigs. J. Anim. Sci. 70:3102-
3111.
De Lange, C. F. M., J. R. Pluske, J. Gong, and C. M. Nyachoti. 2010. Strategic use of feed ingredients and feed
additives to stimulate gut health and development in young pigs. Livest. Sci. 134:124-134.
De Passillè A. M, and J. Rushen. 1989. Using early suckling behavior and weight gain to identify piglets at
risk. Can. J. Anim. Sci. 69:535-544.
Fairbrother, J. M., É. Nadeau, and C. L. Gyles. 2005. Escherichia coli in postweaning diarrhea in pigs: an
update on bacterial types, pathogenesis, and prevention strategies. Anim. Health Res. Rev. 6:17-39.
Hampson, D. J. 1983. Post-weaning changes in the piglet small intestine in relation to growth-checks and
diarrhoea. PhD thesis, University of Bristol.
Hampson, D. J. 1986. Alterations in piglet small intestinal structure at weaning. Res. Vet. Sci. 40:32-40.
Hampson, D. J. 1994. Postweaning Escherichia coli diarrhoea in pigs. Pages 171-191 in Escherichia coli in
Domestic Animals and Humans. C. L. Gyles, ed. CAB Intl., Wallingford, UK.
Hampson, D. J., J. R. Pluske, and D. W. Pethick. 2001. Dietary manipulation of enteric disease. Pages 247-261
in Proceedings of the VIIIth International Symposium on Digestive Physiology in Pigs. J.-E. Lindberg,
and B. Ogle, eds. CAB Intl., Wallingford, U.K.
Hedemann, M. S., B. B. Jensen, and H. D. Poulsen. 2006. Influence of dietary zinc and copper on digestive
enzyme activity and intestinal morphology in weaned pigs. J. Anim. Sci. 84:3310-3320.
Heo, J. M., J.-C. Kim, C. F. Hansen, B. P. Mullan, D. J. Hampson, H. Maribo, N. Kjeldsen, and J. R. Pluske.
2010a. Effects of dietary protein level and zinc oxide supplementation on the incidence of post-weaning
diarrhoea in weaner pigs challenged with an enterotoxigenic strain of Escherichia coli. Livest. Sci.
134:210-213.

J.R. Pluske
Heo, J. M., J.-C. Kim, C. F. Hansen, B. P. Mullan, D. J. Hampson, and J. R. Pluske. 2008. Effects of feeding
low protein diets to piglets on plasma urea nitrogen, faecal ammonia nitrogen, the incidence of diarrhea
and performance after weaning. Arch. Anim. Nutr. 62:343-358.
Heo, J. M., J.-C. Kim, C. F. Hansen, B. P. Mullan, D. J. Hampson, and J. R. Pluske. 2009. Feeding a diet
withdecreased protein content reduces indices of protein fermentation and the incidence of postweaning
diarrhea in weaned pigs challenged with an enterotoxigenic strain of Escherichia coli. J. Anim. Sci.
87:2833-2843.
Heo, J. M., J.-C. Kim, C. F. Hansen, B. P. Mullan, D. J. Hampson, and J. R. Pluske. 2010b. Effects of dietary
protein level and zinc oxide supplementation on performance responses and gastrointestinal tract
characteristics in weaner pigs challenged with an enterotoxigenic strain of Escherichia coli. Anim. Prod.
Sci. 50:827-836.
Heo, J. M., J.-C. Kim, C. F. Hansen, B. P. Mullan, D. J. Hampson, and J. R. Pluske. 2010c. Feeding a diet
with a decreased protein content reduces both nitrogen content in the gastrointestinal tract and post-
weaning diarrhoea, but does not affect apparent nitrogen digestibility in weaner pigs challenged with
an enterotoxigenic strain of Escherichia coli. Anim. Feed Sci. Technol. 160:148-159.
Heo, J. M., F. O. Opapeju, J. R. Pluske, J.-C. Kim, D. J. Hampson, and C. M. Nyachoti. 2012. Gastrointestinal
health and function in weaned pigs: A review of feeding strategies to control post-weaning diarrhoea
without using in-feed antimicrobial compounds. J. Anim. Physio. Anim. Nutr. In press. DOI: http://
dx.doi.org/10.1111/j.1439-0396.2012.01284.x.
Jaeger L.A., C. H. Lama, T. R. Cline, and C. J. Cardona. 1990. Effect of orally administered epidermal growth
factor on the jejunal mucosa of weaned piglets. Am. J. Vet. Res. 51:471-474.
Jeaurond, E. A., M. Rademacher, J. R. Pluske, C. H. Zhu, and C. F. M. De Lange. 2008. Impact of feeding
fermentable proteins and carbohydrates on growth performance, gut health and gastrointestinal
function of newly weaned pigs. Can. J. Anim. Sci. 88:271-281.
Kelly, D. 1994. Colostrum, growth factors and intestinal development in pigs. Pages 151-166 in VIth
International Symposium on Digestive Physiology in Pigs. W.-B. Souffrant, and H. Hagemeister, eds.
EAAP Pub. No. 80. FBN, Dummerstorf, Germany.
Kil, D. Y., and H. H. Stein. 2010. Board Invited Review: Management and feeding strategies to ameliorate
the impact of removing antibiotic growth promoters from diets fed to weanling pigs. Can. J. Anim. Sci.
90:447-460.
Kim, J.-C., J. M. Heo, B. P. Mullan, and J. R. Pluske. 2011. Efficacy of a reduced protein diet on clinical
expression of post-weaning diarrhoea and life-time performance after experimental challenge with an
enterotoxigenic strain of Escherichia coli. Anim. Feed Sci. Technol. 170:222-230.
Kingsnorth A. N., R. Vowles, and J. R. G. Nash. 1990. Epidermal growth factor increases tensile strength in
intestinal wounds in pigs. Br. J. Surg. 77:409-412.
Lallès, J. P, G. Boudry, C. Favier, N. Le Floc’h, I. Luron, L. Montagne, I. P. Oswald, S. Pié, C. Piel, and B. Sève.
2004. Gut function and dysfunction in young pigs: physiology. Anim. Res. 53:301-316.
Lallès, J. P., P. Bosi, H. Smidt, and C. R. Stokes. 2007. Nutritional management of gut health in pigs around
weaning. Proc. Nut. Soc. 66:260-268.
Lallès, J. P., H. Salmon, N. P. M. Bakker, and G. H. Tolman. 1993. Effects of dietary antigens on health,
performance and immune system of calves and piglets. Pages 253-270 in Recent Advances of Research
in Antinutritional Factors in Legume Seeds. J. F. B. Van der Poel, J. Huisman, and M. S. Sani, eds.
Wageningen Pers, Wageningen, the Netherlands.
Le Bellego, L., and J. Noblet. 2002. Performance and utilization of dietary energy and amino acids in piglets
fed low protein diets. Livest. Prod. Sci. 76:45-58.

Le Dividich, J., J. Rooke, and R. Rosario-Ludovino. 1994. Utilization of colostral energy by the newborn pig.
J. Anim. Sci. 72:2082-2089.
Le Dividich, J., J. Marion, and F. Thomas. 2007. Energy and nitrogen utilization of sow colostrum and milk
by the piglet. Can. J. Anim. Sci. 87:571-577.
Le Dividich, J., J. Noblet, J. Herpin, J. Van Milgen, and N. Quiniou. 1998. Thermoregulation. Pages 229-263
in Progress in Pig Science. J. Wiseman, M. A. Varley, and J. P. Chadwick, eds. Nottingham University
Press, Nottingham, UK.
Le Dividich, J., J. A. Rooke, and P. Herpin. 2005. Nutritional and immunological importance of colostrum
for the new-born pig. J. Agric. Sci. 143:469-485.
Le Dividich, J., and B. Sève. 2000. Effects of underfeeding during the weaning period on growth, metabolism,
and hormonal adjustments in the piglet. Domest. Anim. Endocr. 19:63-74.
Li, D. F., J. L. Nelssen, P. G. Reddy, F. Blecha, J. D. Hancock, G. L. Allee, R. D. Goodband, and R. D. Klemm.
1990. Transient hypersensitivity to soybean meal in the early-weaned pig. J. Anim. Sci. 68:1790-1799.
Li, D. F., J. L. Nelssen, P. G. Reddy, F. Blecha, R. D. Klemm, D. W. Giesting, J. D. Hancock, G. L. Allee,
and R. D. Goodband 1991. Measuring suitability of soybean products for early-weaned pigs with
immunological criteria. J. Anim. Sci. 69:3299-3307.
Li, X. L., B. Dong, D. F Li, and J. D. Yin. 2010. Mechanisms involved in the growth promotion of weaned
piglets by high-level zinc oxide. J. Anim. Sci. Biotech. 1:59-67.
Lin, C., D. C. Mahan, G. Y. Wu, and Kim, S. W. 2009. Protein digestibility of porcine colostrum by neonatal
pigs. Livest. Sci. 121:182-186.
Lkhagvadorj, S., Q. Long, W. Cai, P. Oliver, C. Couture, R. Barb, G. J. Hausman, D. Nettleton, L. L. Anderson,
J. C. M. Dekkers, and C. K. Tuggle. 2010. Gene expression profiling of the short-term adaptive response
to acute caloric restriction in liver and adipose tissues of pigs differing in feed efficiency. Am. J. Physiol.
Regul. Integr. Comp. Physiol. 298:R494-R507.
Lordelo, M. M., A. M. Gaspar, L. Le Bellego, and J. P. B. Freire. 2008. Isoleucine and valine supplementation
of a low-protein corn-wheat-soybean meal-based diet for piglets: Growth performance and nitrogen
balance. J. Anim. Sci. 86:2936-2941.
Marion, J., and J. Le Dividich. 1999. Utilization of sow milk energy by the piglet. Page 254 in Manipulating
Pig Production VII. P. D. Cranwell, ed. Australas. Pig Sci. Assoc, Werribee, Australia.
Mavromichalis, I., T. M. Parr, V. M. Gabert, and D. H. Baker. 2001. True ileal digestibility of amino acids in
sow’s milk for 7-day-old pigs. J. Anim. Sci. 79:707-713.
Miller, B. G., T. J. Newby, C. R. Stokes, and F. J. Bourne. 1984. Influence of diet on post-weaning malabsorption
and diarrhoea in the pig. Res. Vet. Sci. 36:187-193.
Nollet, H., P. Deprez, E. Van Driessche, and E. Muylle. 1999. Protection of just weaned pigs against infection
with F18+ Escherichia coli by non-immune plasma powder. Vet. Microbiol. 65:37-45.
NRC. 1998. Nutrient Requirement of Swine. 10th rev. ed. Natl. Acad. Press, Washington, DC.
Opapeju, F. O., D. O. Krause, R. L. Payne, M. Rademacher, and C. M. Nyachoti. 2009. Effect of dietary
protein level on growth performance, indicators of enteric health, and gastrointestinal microbial ecology
of weaned pigs induced with postweaning colibacillosis. J. Anim. Sci. 87:2635-2643.
Ou, D. D. Li, C. Yunha, X. Li, J. Yin, S. Qiao, and G. Wu. 2007. Dietary supplementation with zinc oxide
decreases expression of the stem cell factor in the small intestine of weanling pigs. J. Nutr. Biochem.
18:820-826.
Pakkanen, R., and J. Aalto. 1997. Growth factors and antimicrobial factors of bovine colostrum. Int. Dairy
J. 7:285-297.

J.R. Pluske
Playford, R. J., C. E. Macdonald, and W. S. Johnson. 2000. Colostrum and milk-derived peptide growth
factors for the treatment of gastro-intestinal disorders. Am. J. Clin. Nutr. 72:5-14.
Pluske, J.R., D. J. Hampson, and I. H. Williams. 1997. Factors influencing the structure and function of the
small intestine in the weaned pig: a review. Livest. Prod. Sci. 51:215-236.
Pluske, J. R., C. F. Hansen, H. G. Payne, B. P. Mullan, J.-C. Kim, and D. J. Hampson. 2007. Gut health in the
pig. Pages 147-158 in Manipulating Pig Production XI. J. E. Paterson, and J. A. Barker, eds. Australasian
Pig Science Association, Werribee, Australia.
Pluske, J. R., D. J. Kerton, P. D. Cranwell, R. G. Campbell, B. P. Mullan, R. H. King, G. N. Power, S. G.
Pierzynowski, B. Weström, C. Rippe, O. Peulen, and F. R. Dunshea. 2003. Age, sex and weight at
weaning influence the physiological and gastrointestinal development of weanling pigs. Aust. J. Agric.
Res. 54:515-527.
Pluske, J. R., D. W. Pethick, D. E. Hopwood, and D. J. Hampson. 2002. Nutritional influences on some major
enteric bacterial diseases of pigs. Nutr. Res. Rev. 15:333-371.
Pluske, J. R., I. H. Williams, and F. X. Aherne. 1995. Nutrition of the neonatal pig. Pages 187-235 in The
Neonatal Pig: Development and Survival. M. A. Varley, ed. CAB Intl, Wallingford, UK.
Pluske, J. R., I. H. Williams, and F. X. Aherne. 1996a. Maintenance of villous height and crypt depth in piglets
by providing continuous nutrition after weaning. Anim. Sci. 62:131-144.
Pluske, J. R., I. H. Williams, and F. X. Aherne. 1996b. Villous height and crypt depth in piglets in response
to increases in the intake of cows’ milk after weaning. Anim. Sci. 62:145-158.
Poulsen, H. D. 1995. Zinc oxide for weanling piglets. Acta Agric. Scand. Sect. A, Anim. Sci. 45:159-167.
Rooke, J.A., and I. M. Bland. 2002. The acquisition of passive immunity in the new-born piglet. Livest. Prod.
Sci. 78:13-23.
Schweiger, M., M. Steffl, and W. M. Amselgrube. 2003. Differential expression of EGF receptor in the pig
duodenum during the transition phase from maternal milk to solid food. J. Gastroenterol. 38:636-642.
Shimizu, M., and T. Terashima.1982. Appearance of enterotoxigenic Escherichia coli in piglets with diarrhea
in connection with feed changes. Curr. Top. Microbiol. Immunol. 26:467-477.
Stein, H. H., and D. Y. Kil. 2006. Reduced use of antibiotic growth promoters in diets fed to weanling pigs:
Dietary tools. Part 2. Anim. Biotechnol. 17:217-231.
Tokach, M. D., R. D. Goodband, J. L. Nelssen, and L. J. Kats. 1992. Influence of weaning weight and growth
during the first week post-weaning on subsequent pig performance. Pages 15-17 in Report of Progress
667, Kansas University Swine Day. Manhattan, Kansas, USA.
Tuchscherer, M., B. Puppe, A. Tuchscherer, and U. Tiemann. 2000. Early identification of neonates at risk:
traits of newborn piglets with respect to survival. Theriogenology 54:371-388.
Van Barneveld, R. J., and F. R. Dunshea. 2011. Colostrum protein isolate increases gut and whole body
growth and plasma IGF-1 in neonatal pigs. Asian-Aust. J. Anim. Sci. 24: 670-677.
Van Dijk, A.J., H. Everts, M. J. A. Nabuurs, R. J. C. F. Margry, and A. C. Beynen. 2001. Growth performance
of weanling pigs fed spray-dried animal plasma: a review. Livest. Prod. Sci. 68:263-274.
Vente-Spreeuwenberg, M. A. M, and A. C. Beynen. 2003. Diet-mediated modulation of small intestinal
integrity in weaned piglets. Pages 145-198 in Weaning the Pig: Concepts and Consequences. J. R.
Pluske, J. Le Dividich, and M. W. A. Verstegen, eds. Wageningen Academic Publishers, Wageningen,
the Netherlands.
Wellock, I. J., P. D. Fortomaris, J. G. M. Houdijk, and I. Kyriazakis. 2008. Effects of dietary protein supply,
weaning age and experimental enterotoxigenic Escherichia coli infection on newly weaned pigs: health.
Animal 2:834-842.

Xu, R. J. 1996. Development of the newborn GI tract and its relation to colostrum/milk intake: a review.
Reprod. Fert. Dev. 8:35-48.
Xu, R. J., P.T. Sangild, Y. Q. Zhang, and S. H. Zhang. 2002. Bioactive compounds in porcine colostrum
and milk and their effects on intestinal development in neonatal pigs. Pages 169-192 in Biology of the
Intestine in Growing Animals. R. Zabielski, P. C. Gregory, and B. Weström, eds. Elsevier. Amsterdam,
the Netherlands.
Yun, J. H., I. K. Kwon, J. D. Lohakare, J. Y. Choi, J. S. Yong, J. Zheng, W. T. Cho, and B. J. Chae. 2005.
Comparative efficacy of plant and animal protein sources on the growth performance, nutrient
digestibility, morphology and caecal microbiology of early-weaned pigs. Asian-Aust. J. Anim. Sci.
18:1285-1293.

13. Emerging technologies with the potential to improve feed
efficiency in swine
F.R. Dunshea
CRC for an Internationally Competitive Pork Industry; Department of Agriculture and Food
Systems, Melbourne School of Land and Environment, The University of Melbourne, Parkville, Vic
3010, Australia; fdunshea@unimelb.edu.au
Abstract
Although there are metabolic modifiers such as injectable porcine somatotropin available for some
pork industries, producers are very much interested in orally active compounds that may cause
metabolic shifts that result in similar improvements in feed efficiency and body composition.
The purpose of this review is to discuss some of the alternatives that may be used to manipulate
growth, feed efficiency, and body composition in growing pigs. The new technologies include
immunization against gonatotropin releasing factor (GnRF) to enhance the use of boars while the
orally active dietary additives discussed include ractopamine, cysteamine, chromium, betaine,
and dietary neurolepetics. Cysteamine may stimulate somatotropin secretion in pigs and under
some circumstances can increase ADG and decrease back fat. However, responses are variable
and further work is needed to define the dose response curve. Responses to dietary chromium
have also been variable and this may be in part due to inefficient digestion and absorption.
Micro- and nano-particles of chromium may increase efficacy of dietary chromium and improve
insulin sensitivity and growth performance in pigs. Dietary betaine has the potential to reduce
maintenance requirements and in some circumstances this additional energy can be used for
protein deposition. Dietary neuroleptics offer a means of reducing sexual and aggressive activities
in finisher pigs, specifically boars, and improve growth performance. Immunization against
GnRF reduces boar taint, sexual activity, and aggression in boars and improves lifetime feed
efficiency over that of barrows. Responses to dietary feed additives are generally not as powerful
as the responses to porcine somatotropin but they do offer an alternative means to manipulate
growth performance and carcass quality.
Introduction
It has been known for almost 60 years that injection of pigs with pituitary tissue extracts containing
porcine somatotropin (pST) increases protein deposition and decreases fat accretion in growing
pigs (Turman and Andrews, 1955). Advances in biotechnology have now provided a means of
producing pST on a commercial scale and the efficacy of daily injection of recombinantly-derived
pST for improving productive performance of swine is beyond doubt (Campbell et al., 1991).
Porcine somatotropin has now been approved for use in the Australian pig industry as a daily
injectable metabolic modifier (Dunshea et al., 2002; Dunshea, 2005). However, daily injection
with pST is problematic and so alternative injection regimes with less frequent injections have
been investigated and demonstrated to have similar efficacy to daily injections (Dunshea, 2002).
Regardless, pork producers are still interested in orally active compounds that may cause metabolic
shifts that result in similar improvements in body composition, in particular reductions in fat

F.R. Dunshea
depth. The purpose of this review is to discuss some of the emerging technologies and dietary
additives that may be used to manipulate growth, body composition, or meat quality in growing
pigs. The orally active dietary additives discussed include the β-agonist ractopamine, cysteamine,
chromium, lecithin, betaine and dietary narcolpetics. There will also be a brief review of pST since
it’s key presence in the ST/IGF axis makes it an important point of regulation and some of the
alternative technologies focus on the regulation of this axis.
Porcine somatotropin
Somatotropin is a protein hormone secreted from the anterior pituitary gland under the
action of growth hormone-releasing factor (stimulatory; GRF) and somatostatin (inhibitory,
SRF) released from the hypothalamus with quite distinct upstream and downstream points of
regulation (Figure 1; Wray-Cahen et al., 1998). Exogenous porcine somatotropin (pST) treatment
consistently improves average daily gain, feed conversion efficiency, and protein deposition, and
reduces fat deposition in pigs (Chung et al., 1985; Etherton et al., 1987; Evock et al., 1988; Campbell
et al., 1988; 1989; King et al., 2000). Dose-dependent increases in lean deposition and reductions
in feed intake, fat deposition, and carcass fat have been observed (Chung et al., 1985; Evock et
al., 1988; Krick et al., 1992; Suster et al., 2005). It appears that the pST-induced shift in nutrient
partitioning is mediated, in part, through inducing insulin resistance in adipose tissue. Unlike
for bovine ST, which is delivered via an implant, under commercial conditions pST is delivered
as either a daily, every other day or thrice weekly injection (ca. 5 mg/day) administered using a
propane powered applicator (Dunshea, 2002; Dunshea et al., 2002). As well as issues related to
Endocrine hypothalmus
Inhibitory SRIF GRF CRF
Stimulatory Cysteamine Ghrelin
(+/–) Variable Anterior pituitary
ST ACTH
(+/–)
Prostaglandins
(+/–)
Whole body Adrenal gland
IGF-I & IGFBPs Cortisol
(+/–) Immune system
(+/–) Cytokines
Chromium Insulin axis Thyroid
axis
(+/–)
Diet
Most significant
IGF-I effector Bone Adipose Muscle
and an immune
modulator
Figure 1. Somatrophic axis showing key points of regulation of growth and nutrient partitioning (modified
from Wray-Cahen et al., 1998).

13. Emerging technologies with the potential to improve feed efficiency in swine
dissolution and aggregation of the pST molecule (Roberts et al., 2001), a major impediment to
the development of a slow release pST implant is the degree of insulin resistance induced by pST
in pigs. The pST implants investigated to date display first-order kinetics with a declining release
over time (Klindt et al., 1992) and this presents a challenge in developing a prolonged release
system that would maintain an effective circulating dose of pST over time without inducing
severe insulin resistance.
Consequently, without a suitable prolonged release delivery system, current registration of pST
involves daily injection of 5 mg pST per day for up to 30 days over the finishing period (Dunshea et
al., 2002). Injection of pST is via a proprietary liquid petroleum gas driven injection device. While
at first thought daily injection of pST may appear to be an onerous task, experienced operators
can administer in excess of 1000 doses per h. The daily injection procedure has been shown to
be moderately aversive and that the welfare of injected pigs is similar to that of pigs receiving
minimal human contact as occurs in the routine husbandry of growing pigs (Hemsworth et al.,
1996). Hemsworth et al. (1996) found that pigs injected with saline using the injection gun had
similar growth rates, adrenal gland weights, and plasma cortisol to pigs that received minimal
handling. On the other hand, pigs that were negatively handled had compromised growth rate,
larger adrenal glands, and elevated plasma cortisol. Nevertheless, there is much interest in
reducing the frequency of injections of pST to reduce labor costs. Dunshea (2002) compared
daily saline injection, daily injection with pST (5 mg) (D), bidaily injection of pST (10 mg) (2D),
and injection of pST (12 mg) every Monday, Wednesday, and Friday (MWF) for 3 weeks. All
pST regimes caused a reduction in feed intake, feed conversion ratio (FCR), and P2 back fat.
Responses were greatest for the D group with the 2D and MWF treatment regimes being similar.
Using a leaner genotype, Hellams et al. (2000) conducted a dose response study comparing D
with MWF regimes and found that both injection regimes gave similar FCR responses at the
current recommended dose of 35 mg pST/week (Figure 2). Consequently, most producers who
use pST would inject pST less frequently than the daily regime. Overwhelmingly, the greatest
impediment to wider adoption of pST is the mode of delivery and the development of a slow
release delivery system is required. Efforts should continue in this area.
Ractopamine
The β-agonist ractopamine is approved for use in many countries as an in-feed ingredient to
increase lean tissue growth and improve production efficiency in pigs. Treatment of pigs with
β-agonists, particularly ractopamine (RAC), generally has given dose-dependent improvements
in average daily gain (ADG), FCR, and carcass lean content (see Dunshea and Gannon, 1995).
Unlike for pST, feed intake is typically unchanged (Adeola et al., 1990; Gu et al., 1991b; Yen et al.,
1991; Moore et al., 2009) or decreased slightly (Adeola et al., 1990; Mitchell et al., 1991; Watkins,
et al., 1990; Rikard-Bell et al., 2009a) during β-agonist treatment. Other β-agonists that have
improved performance in finisher pigs are salbutamol, cimaterol, clenbuterol, Ro16-8714, BRL-
47672, and L-644,969 although none of these other β-agonists have been approved for use in pig
production (Dunshea and Gannon, 1995).
While there is general agreement that protein and lean tissue deposition are increased during
β-agonist treatment, effects on fat deposition have been equivocal. For example, while RAC

F.R. Dunshea
3.00 Daily
MWF
2.80
2.60
FCR
2.40
2.20
2.00
0 20 40 60 80
Dose of pST (mg/week)
Figure 2. Effect of administration of 35 mg pST per week as either daily injections (daily, 5 mg/day) or given
every Monday (10 mg), Wednesday (10 mg) and Friday (15 mg) (MWF) for 4 weeks on feed conversion ratio
(FCR) (Hellams et al., 2000).
increased protein deposition in boars, gilts, and barrows, there was little effect on fat deposition
(Dunshea et al., 1993a). A review of the literature suggests that ractopamine does not appear to
decrease backfat measured along the midline (Aalhus et al., 1990; Dunshea et al., 1993a;b;1998a;
Gu et al., 1991a; Yen et al., 1991) whereas backfat depths measured off the midline have been
either decreased (Aalhus et al., 1990; Adeola et al., 1990; Dunshea et al., 1993b; Gu et al., 1991a;
Watkins et al., 1990; Yen et al., 1991) or unchanged (Dunshea et al., 1998a; Mitchell et al., 1991;
Sainz et al., 1993; Yen et al., 1991). Since the approval of ractopamine in the USA and elsewhere,
additional data have become available investigating a range of dose and treatment regimes. A
summary of proprietary information from 20 studies conducted in the late 1980s to early 1990s
indicated a reduction in backfat at the 10th rib in relatively fat animals (Schinckel et al., 2001).
In a more recent experiment with leaner pigs, there was a reduction in 10th rib backfat with
ractopamine feeding although the response was not as great (Schinckel et al., 2001). However, a
summary of a number of recent Australian studies with ractopamine using a variety of treatment
regimens in different classes of pigs concluded that although lean meat yield was increased, there
were no effects of ractopamine on backfat measured at either the P2 site or over the leg (Smits
and Cadogan, 2003). β-agonists act directly through β-adrenergic receptors on adipocytes and
influence cellular metabolism via signalling cascades and so should be expected to increase
lipolysis (fat mobilization and hydrolysis) (Liu and Mills, 1989; Dunshea, 1993; Mersmann,
1998). However, ractopamine and other β-agonists do not appear to decrease fat deposition in
pigs because of a combination of rapid down-regulation of adipocyte β-adrenergic receptors
(Dunshea, 1993; Dunshea and King, 1995; Dunshea et al., 1998b; Rikard-Bell et al., 2011) and a
relative insensitivity of porcine adipocytes to β-agonists (Dunshea and D’Souza, 2003).

One of the actions of ST on adipose tissue metabolism is to increase responsiveness to

β-adrenergic receptor stimulation (Sechen et al., 1990; Boisclair et al., 1997). Recent studies in
Australia have attempted to utilize the stimulatory effects of pST administration on β-adrenergic
receptor response to counteract the rapid down-regulation of adipocyte β-adrenergic receptors
that occurs during ractopamine treatment (Rikard-Bell et al., 2009b; Van Barneveld et al., 2009).
For example, Van Barneveld et al. (2009) was able to show that either daily or bi-weekly treatment
of finisher pigs with pST over the last 2 weeks of a 4 week treatment with ractopamine could
further enhance growth performance over that of ractopamine alone (Figure 3). Thus, FCR was
decreased from 3.3 to 3.0 with ractopamine alone and to 2.6 with the combination of ractopamine
and pST over the last 2 weeks. The FCR for pigs given four injections of pST over the last 2 weeks
was not significantly different than those given daily injections.
Interestingly, Israel et al. (2001) reported that there were polymorphisms in the human β2-
adrenergic receptors genotype (on codon 27) that resulted in marked variation in bronchodilator
response to albuterol. Interestingly, some of these polymorphisms were related to body mass
index in sedentary females (Barr et al., 2001). If similar polymorphisms exist in β-adrenergic
receptors genotypes in pigs it may be possible to select for a pigs that have either an increased
responsiveness to endogenous catecholamines or to exogenous β-agonists such as ractopamine.
Cysteamine
The release of endogenous pST from the pituitary is under the regulation of a number of
secretagogues and inhibitors such as somatostatin. A number of researchers have attempted to
increase pST secretion by immunizing against somatostatin but results have been inconsistent
possibly due to low antibody titres (Dubreuil et al., 1989; McCauley et al., 1995). Previous
studies have found that the sulfhydryl compound cysteamine hydrochloride (CSH) increases
somatotropin secretion in rats (Tannenbaum et al., 1990), sheep (McLeod et al., 1995a,b) and
fish (Xiao and Lin, 2003) presumably through inhibition of somatostatin secretion (Wakabayashi
Feed:gain
3.4
3.2
3.0
2.8
2.6
2.4
Control Ractopamine Ractopamine + Ractopamine +
daily pST pST 2×/week
Figure 3. Effect of dietary ractopamine treatment for 4 weeks with alone or with two treatment regimes of
pST for the last 2 weeks (Van Barneveld et al., 2009).

F.R. Dunshea
et al., 1985). There have been a few recent studies that have investigated the effects of CSH on
growth performance in pigs. Zhou et al. (2005) fed CSH (0.6 g/kg diet) to finisher pigs for 5 weeks
and reported an increased ADG (+4.8%) and decreased back fat (-20%). Dunshea (2007) found
that dietary CSH (0.7 g/kg diet) increased ADG (+12%) and decreased FCR (-9.4%) over the first
2 weeks of supplementation. While the response was not as pronounced over the entire 5 weeks,
dietary CSH supplementation resulted in a modest increase in ADG (+7.4%). As a result, final
liveweight (+2.0 kg) and carcass weight (+1.6 kg) were increased while back fat was decreased
(-1.0 mm) by CSH. Yang et al. (2005) found that daily gain was increased (+13.7%) by a low dose
(0.03 g/kg feed) but not a higher dose (0.05 g/kg feed) of CSH although these doses are very low
compared to other studies. Back fat at the P2 site was reduced in a dose dependent manner by
dietary CSH supplementation (15.7, 12.7, and 10.4 mm for pigs fed 0, 0.03, and 0.05 g/kg CSH).
Finally, Liu et al. (2009) found that dietary CSH (0.07 g/kg) increased ADG (+19%), feed intake
(+15%), and nitrogen retention (+63%) but did not change FCR. In a recent study conducted
within the Pork CRC dietary CSH (0.7 mg/kg) had no effect on growth performance but did
decrease P2 back fat (-1.9 mm) (David Miller, personal communication). Cysteamine had no
effect on serum somatotropin but did decrease gherelin after 3 weeks of treatment. The inhibitory
effect of CSH on gherelin may be a feedback mechanism and may partially explain the apparent
transitory nature of the growth response (Dunshea, 2007). An area of research that needs to be
investigated is the dose response to CSH, as there appears to be 10-fold range of doses being
investigated and it may be that down-regulation is more rapid at higher doses.
Chromium
Chromium is an essential mineral element in both humans and animals (Mertz, 1969;1993) with
trivalent Cr (Cr3+) being the most stable form occurring in nature. Chromium functions as a
cofactor for the hormone insulin and enhances the ability of insulin to regulate glucose, protein,
and fat metabolism (Amoikon et al., 1995). However, Cr3+ is normally poorly absorbed and
utilized with the digestibility of inorganic and organic forms of Cr being 0.5-2% and 10-25%,
respectively (Mertz, 1969; Underwood, 1977). Currently, there is no recommendation for dietary
chromium (Cr) supplementation (NRC, 1998) for pigs and since most pig diets are primarily
formulated from ingredients from plant origin, which are usually low in Cr content (Giri et al.,
1990), it is possible that many pig diets could be low in Cr.
It has been suggested that dietary chromium may increase insulin sensitivity in pigs (Steele et al.,
1977; Matthews et al., 2001b) and over the last three decades it has been investigated as a potential
means of manipulating fat deposition in humans and farm animals. While chromium chloride
has been successful in improving the growth rate of turkeys (Steele and Rosbrough, 1979) or
protein deficient rats (Roginski and Mertz, 1969), effects in pigs have been more equivocal (Page
et al, 1993; Mooney and Cromwell, 1995; 1997; Zhang et al., 2011). Page et al. (1993) reported the
data from three experiments conducted to determine the efficacy of Cr chloride, Cr picolinate,
and picolinate alone in grower-finisher pigs. While there were no effects of Cr or picolinate on
growth rate or FCR there were dose dependent effects of Cr picolinate on loin eye area and backfat
depth. Others have also seen little or no effect of Cr picolinate on ADG rate or FCR (Evock-Clover
et al., 1993; Mooney and Cromwell, 1995; Kim et al., 2009). Zhang et al. (2011) saw no effect of
dietary Cr picolinate in grower pigs but a small increase in ADG in finisher pigs. Lean tissue

deposition or loin eye area was increased and fat deposition or backfat decreased by Cr picolinate
in some studies (Page et al., 1993; Jackson et al., 2009) but not in others (Evock-Clover et al., 1993;
Matthews et al., 2001c). Recently, Sales and Jancik (2011) conducted a meta-analyses of 31 studies
with dietary Cr in pigs and concluded that overall there were small but significant increases in
ADG (P=0.023), feed efficiency (P<0.001), and carcass lean (P<0.001) and reductions in carcass
fat (P=0.009). However, it was noted that the effects within studies were small and at least for
ADG and feed efficiency the 95% confidence intervals for the effect in most studies included 0.
At least part of the variation in responses to Cr may be related to the low and variable digestion,
absorption, and availability of Cr and there is potential to improve this by producing nano- or
micro-sized Cr. For example, in rodents, the efficiency of uptake of 100 nm size particles by the
intestinal tissue was 15-250 fold higher compared to large size microparticles (Desai et al., 1996).
Indeed, the digestibility of nano- Cr picolinate was approximately 3 fold higher than native Cr
picolinate in rats (Lien et al., 2009). Recently, we conducted a meta-analysis of the 7 studies
conducted with nano Cr over the last 6 years and found that ADG and FCR were improved
by 7 and 4%, respectively (Hung et al., 2010). Hung et al. (2009) investigated the interactions
between dietary fat (2 vs. 6% fat) and form of dietary Cr picolinate (400 μg/kg of either normal,
micro- and nano-sized particles) for 8 weeks in finisher gilts. Nano- and micro-Cr was processed
by the dry polish method in a dry cryo-nanonization grinding system integrated with a size
separator. Briefly, the particle size of the raw Cr picolinate was reduced by a grinder and then
passed through appropriate sized sieve end plates to collect nano- and micro-size particles of Cr
picolinate. No solvent was used and the temperature was controlled under 40 °C during the milling
process. Over the first 21 days of the experiment there was a significant effect of Cr on ADG (944
vs. 1011 g/d, respectively, P=0.021) but not of Cr size (P=0.17). Dietary Cr increased carcass weight
(Figure 4) and muscle depth with responses being greatest for nCr. Also, dietary Cr decreased P2
back fat with the greatest response seen in pigs fed nCr and a high fat diet. Furthermore, dietary Cr
decrease plasma insulin (7.0 vs. 5.1 mU/ml, P=0.04) without changing plasma glucose (3.6 vs. 3.5
mmol/l, P=0.60), and increased muscle PI3K and Akt mRNA expression (Hung et al., 2011a). These
metabolite and mRNA expression data indicate an improvement in insulin sensitivity in muscle
72
70
Carcass weight (kg)
68
66
64
62
60
Control CrPic micro CrPic nano CrPic
Figure 4. Effect of different forms of dietary chromium picolinate on carcass weight of finisher gilts (from
Hung et al., 2009).

F.R. Dunshea
through enhanced signalling downstream of the insulin receptor. A more recent study conducted
under commercial conditions with very lean gilts found an increase in final liveweight (94.0 vs. 95.4
kg, P=0.09) but no change in back fat (8.0 vs. 8.0 mm, P=0.94) in pigs fed a diet containing 400 μg/
kg of micro-Cr picolinate (Hung et al., 2011b). These data suggest that dietary Cr picolinate can
increase ADG and improve carcass traits in a lean genotype with responses being most pronounced
in pigs fed a high fat diet and micro-Cr and nano-Cr. Further studies should be conducted to
determine the dose response curve to determine whether the responses to nano- and micro-Cr
picolinate might occur at lower inclusion rates and hence are more efficacious than native CrPic.
Betaine
Betaine is an active methyl donor with a lipotropic effect (Barak et al., 1993). Betaine also acts
as a methyl donor which has been investigated as a partial replacement for choline in pig and
poultry diets since it is less expensive and less corrosive than choline. When incorporated into pig
diets, betaine has been reported to improve growth performance by reducing the maintenance
energy requirement of the animal, perhaps by reducing the need for ion pumping involved with
maintaining intracellular osmolarity (Schrama et al., 2003). In addition, dietary betaine has been
reported to increase protein deposition and carcass leanness (Matthews et al., 2001a; Fernandez-
Figares et al., 2002; Suster et al., 2004) and to decrease back-fat (Cadogan et al., 1993). There is
evidence that betaine has a more pronounced effect when dietary energy is limiting (Suster et
al., 2004) and so it offers a means of improving growth performance and meat quality through
ensuring the provision of additional energy. The additional energy may improve or maintain
performance and meat quality in conditions where performance may be limited by energy intake
such as in genetically improved pigs housed under commercial conditions and in pigs treated
with pST or ractopamine (Suster et al., 2004; Dunshea et al., 2009). Sales (2011) conducted a
meta-analyses of 19 studies with dietary betaine in pigs and concluded that overall while there
was no effect on ADG, dietary betaine improved feed efficiency and carcass lean.
Dietary neuroleptics
In Australia, the pork industry has traditionally not castrated male pigs in order to take advantage
of the improved carcass leanness and feed efficiency associated with entire males. Apart from
the risk of boar taint, another disadvantage of entire male pig production is the likelihood of
aggressive and sexual activity among group housed boars (Cronin et al., 2003; Rhydmer et al.,
2006, 2010) which may prevent them from performing to their potential. Cronin et al. (2003)
found that immunization against gonadotrophin releasing factor (GnRF) to reduce boar taint
also decreased sexual and aggressive activities to levels observed in physical castrates. In addition,
dietary tryptophan can reduce aggression in boars mixed during lairage (Pethick et al., 1997)
while dietary magnesium reduces plasma catecholamine concentrations and the incidence of
pale soft exudative pork in negatively-handled pigs (D’Souza et al., 1998). Potassium bromide
is a dietary neuroleptic that has been shown to decrease sexual and aggressive activities without
altering growth rate in growing bulls (Genicot et al., 1991).
These findings led McCauley et al. (2003a) to conduct a study to examine the effects of dietary
magnesium (5 g magnesium proteinate/kg, Mg), bromide (140 mg sodium bromide/kg, Br),

and tryptophan (5 g tryptophan/kg, Trp) on growth performance of entire male pigs. The
growth performance was compared to that of control entire males, physical castrates, and boars
immunized against GnRF (see next section). The immunized boars were given the first dose of
an anti-GnRF vaccine (Improvac™, Pfizer Animal Health, Parkville) at this time. Over the period
between 17 and 19 weeks and 19 and 22 weeks, there was no significant effect of sex or dietary
treatment on ADG (Table 1). Over the entire period between 17 to 22 weeks, the immunized
boars had a greater ADG than all other classes of pigs. The ADG of physical castrates tended to be
higher than the entire boars (P=0.087). Feed intake of the control and Trp boars was less (P<0.05)
than that of the other groups over the first two weeks after transfer into the finisher shed. For the
Trp pigs this was most pronounced during the first week, whereas for the control boars it was
most pronounced during the second week. Feed intake of the physical castrates was higher than
any of the other groups over the first 2 weeks of the finishing period. There was no effect of any
of the dietary additives on feed intake of entire boars over the latter part of the finishing phase
between 19 and 22 weeks. Entire males immunized against GnRF increased their feed intake after
the first week post secondary immunization to a similar level as the surgical castrates. There was
no effect of sex or dietary additive on CR over the first 2 weeks of the finishing period. However,
over the latter part of the finishing phase the FCR of the physical castrate pigs was 21% higher
than that of the control entire males. Over the entire finishing period the FCR of the Trp boars
was 10% lower than the control entire males, whereas the FCR of the surgical castrates was 17%
higher than that of the control entire males. The FCR of the boars immunized against GnRF or
consuming diets containing either Br or Mg were not different from the control entire males. Final
live weight of the castrates and boars immunized against GnRF were greater (P<0.05) than the
control entire males and the boars consuming Trp with the others intermediate. Carcass weight was
significantly increased over the control entire males in the castrates, in the immunized boars and
in the boars fed diets containing Br. In the surgical castrates and Br boars, this was a result of both
greater live weight as well as increased dressing rate whereas for the immunized boars the increased
carcass weight was the result of increased live weight but unchanged dressing rate. Dressing out rate
was also increased by dietary Mg and Trp. Surgical castrates had higher P2 (+5 mm) fat than the
Table 1. Effect of sex and dietary additives over the finisher phase between 17 and 22 weeks of age on final
weight and carcass characteristics at slaughter (after McCauley et al. 2003a and unpublished)1.
Boars Anti- Castrate LSD2 P-value

GnRF
Control Mg Br Trp
Final weight (kg) 93.7 95.2 97.9 94.1 99.8 99.3 4.66 0.044
Carcass weight (kg) 69.0 71.3 74.1 71.0 73.7 76.8 4.62 0.053
Dressing (g/kg) 750.9 761.0 761.4 760.1 754.8 773.0 8.70 0.009
P2 back fat (mm) 10.6 11.0 11.1 10.3 11.7 15.6 1.36 <0.001
1 Anti-GnRF injections were given at 13 and 17 weeks of age.
2 Least significant difference (P=0.05) between treatment groups.

F.R. Dunshea
control boars whereas there was no significant effect of any of the dietary additives or immunization
on either P2 fat.
A subsequent study (McCauley et al., 2004), investigated the effects of Br and Trp, alone and
in combination, in pigs that were light, average, or heavy for age. Again, immunization against
GnRF was used as a positive control. Over the period from 17 to 22 weeks, the immunized
males grew more quickly than all other classes of pigs. In particular, the immunized males grew
19% (+153 g/d) faster than the control entire males. However there was an interaction between
treatment and weight such that the growth response was greatest in the medium weight class of
pigs treated with Improvac and least in the light pigs. While there were no significant individual
effects of either bromide (+1.8%) or tryptophan (+2.2%) treatments on daily gain, pigs treated
with both compounds grew significantly faster (10.3%, P<0.05) than controls. Importantly, there
was an interaction between treatment and weight class such that this effect was most pronounced
in the heavy pigs where pigs from all treatment groups grew faster than the control entire males.
Immunization against GnRF
Traditionally, the major reasons for castration were to control the reproductive state of
contemporary females, to take advantage of the propensity for castrate animals to fatten, to
eliminate male pheromones and to reduce the incidence of aggressive behaviors often observed
in entire animals. However, castration results in significant reductions in feed efficiency and
excess deposition of fat (Dunshea and D’Souza, 2003). An alternative method of inhibiting sexual
development and aggressive behaviours in the late finisher phase and reducing the accumulation
of boar taint compounds in pork is immunization against GnRF. This results in a reduction
in plasma gonadotropins, testosterone, and boar taint compounds (Dunshea et al., 2001). The
reduction in testosterone as a result of immunization against GnRF also has a profound effect upon
behaviour. Cronin et al. (2003) found that there was a reduction in both aggressive and sexual
activities in immunized boars who exhibited similar activities as barrows. As a consequence, the
immunized pigs increased the amount of time they spent eating and feed intake rose. There are
now a number of studies conducted with anti-GnRF immunization across the globe and these
have been incorporated into a series of meta-analyses. These analyses of 16 studies that, compared
to non-immunized boars, immunization against GnRF increases feed intake and ADG with only
small increases in FCR over the finisher phase. Final carcass weight is increased over that of
entire males, as is back fat (Dunshea et al., 2001, 2011, unpublished results). There are now also
sufficient data comparing the growth performance and carcass characteristics of immunized
boars with those of barrows to conduct comparative meta-analyses. These analyses of up to 11
studies show that during the finisher phase, immunization against GnRF increases feed intake
and ADG and decreases FCR compared to contemporary barrows (Table 2). Final slaughter
measurements, reflecting growth both before (entire male phase) and after second immunization
show increased live weight, decreased back fat and decreased dressing %, with no effect on carcass
weight (F.R. Dunshea, unpublished results). It should be borne in mind that the improvements in
FCR are cumulative across the entire grower/finisher period (Hennessy et al., 2009) and this may
be a driver for adoption of this technology more widely. Also, immunization against GnRF has
additive effects when used with pST (McCauley et al., 2003b; Oliver et al., 2003) and ractopamine
(Moore et al., 2009; Rikard-Bell et al., 2009a).

Table 2 Average fixed effects of immunization against GnRF and physical castration (immunized males
– barrows) from meta-analyses of data from studies (n=11) with group-housed pigs1,2,3 (F.R. Dunshea,
unpublished results).
Effect SED 95% CI P-value # studies
ADG (g/d) 149 14.7 (120, 178) <0.001 10

ADFI (g/d) 107 42.3 (24, 190) 0.011 9
FCR -0.35 0.031 (-0.41, -0.29) <0.001 8
Live weight (kg) 2.21 0.604 (1.03, 3.40) <0.001 11
Carcass weight (kg) 0.45 0.541 (-0.61, 1.51) 0.41 10
Dressing (%) -1.63 0.148 (-1.92, -1.34) <0.001 8
Back fat (mm) -2.59 0.297 (-3.17, -1.28) <0.001 9
1 Analyses only included data from studies where animals slaughtered between 4 and 5 weeks after the secondary
vaccination (mean final live weight 110 kg).

2 Difference was determined over the period between the second vaccination and slaughter or equivalent time in
barrows.
3 Sources include: Mullan and D’Souza (2000); Dunshea et al. (2000, 2001); Schmoll et al. (2009); Zamaratskaia et al.
(2008); Pauly et al. (2009); Fabrega et al. (2010); Morales et al. (2010). Some sources include more than one study.
Conclusions
Dietary additives offer a means to manipulate growth performance and carcass quality. However,
responses are often variable and inconsistent, so further research is needed to clearly determine
under which circumstances they are most effective. While the responses to dietary feed additives
are generally not as powerful as the responses to porcine somatotropin, they do offer an alternative
means to manipulate growth performance and carcass quality. Research should continue to try
to develop a delivery system for pST.
References
Aalhus, J. L., S. D. M. Jones, A. L. Schaefer, A. K. W. Tong, W. M. Robertson, J. K. Merrill, and A. C. Murray.

1990. The effect of ractopamine on performance, carcass composition and meat quality of finishing pigs.
Can. J. Anim. Sci. 70: 943-952.
Adeola, O., E. A. Darko, P. He, and L. G. Young. 1990. Manipulation of porcine carcass composition by
ractopamine. J. Anim. Sci. 68: 3633-3641.
Amoikon, E. K., J. M. Fernandez, L. L. Southern, D. L. Thompson, Jr., T. L. Ward, and B. M. Olcott. 1995.
Effect of chromium tripicolinate on growth, glucose tolerance, insulin sensitivity, plasma metabolites,
and growth hormone in pigs. J. Anim. Sci. 73: 1123-1130.
Barak, A. J., H. C. Beckenhauer, M. Junnila, and D. L. Tuma. 1993. Dietary betaine promotes generation of
hepatic S-adenosylmethionine and protects liver from ethanol-induced fatty infiltration. Alcohol. Clin.
Exp. Res. 17: 552-553.

F.R. Dunshea
Barr, R. G., D. M. Cooper, F. E. Speizer, J. M. Drazen, and C. A. Camargo, Jr. 2001. Beta(2)-adrenoceptor
polymorphism and body mass index are associated with adult-onset asthma in sedentary but not active
women. Chest 120: 1474-1479.
Boisclair, Y. R., K. B. Johnston, D. E. Bauman, B. A. Crooker, F. R. Dunshea, and A. W. Bell. 1997. Paradoxical
increases of circulating nonesterified fatty acids in somatotropin treated cattle undergoing mild
disturbances. Dom. Anim. Endo. 14: 251-262.
Cadogan, D. J., R. G. Campbell, D. Harrisonand A. C. Edwards. 1993. The effects of betaine on the growth
performance and carcass characteristics of female pigs. pp. 219 in Manipulating Pig Production IV. E.
S. Batterham, ed. Australasian Pig Science Association: Werribee, Australia.
Campbell, R. G., N. C. Steele, T. J. Caperna, J. P. McMurtry, M. B. Solomon, and A. D. Mitchell. 1988.
Interrelationships between energy intake and endogenous porcine growth hormone administration on
the performance, body composition and protein and energy metabolism of growing pigs weighing 25
to 55 kg liveweight. J. Anim. Sci. 66: 1643-1655.
Campbell, R. G., N. C. Steele, T. J. Caperna, J. P. McMurtry, M. B. Solomon, and A. D. Mitchell. 1989.
Interrelationships between sex and exogenous growth hormone administration on performance, body
composition and protein and fat accretion of growing pigs. J. Anim. Sci. 67: 177-186.
Campbell, R. G., R. J. Johnson, M. R. Taverner, and R. H. King. 1991. Interrelationships between exogenous
porcine somatotropin (PST) administration and dietary protein and energy intake on protein deposition
capacity and energy metabolism of pigs. J. Anim. Sci. 69: 1522-1531.
Chung, C. S., T. D. Etherton, and J. P. Wiggins. 1985. Stimulation of swine growth by porcine growth
hormone. J. Anim. Sci. 60: 118-130.
Cronin, G. M., F. R. Dunshea, K. L. Butler, I. McCauley, J. L. Barnett, and P. H. Hemsworth. 2003. The effects
of immuno- and surgical-castration on the behaviour and consequently growth of group-housed, male
finisher pigs. App. Anim. Behav. Sci. 81: 111-126.
Desai, M. P., V. Labhasetwar, G. L. Amidon, and R. J. Levy. 1996. Gastrointestinal uptake of biodegradable
microparticles: Effect of particle size. Pharm. Res. 13: 1838-1845.
D’Souza, D. N., R. D. Warner, B. J. Leury, and F. R. Dunshea. 1998. The effect of dietary magnesium aspartate
supplementation on pork quality. J. Anim. Sci. 76: 104-109.
Dubreuil, P., G. Pelletier, D. Petitclerc, H. Lapierre, P. Gaudreau, and P. Brazeau. 1989. Effects of active
immunisation against somatostatin on serum growth hormone concentration in grower pigs: Influence
of fasting and repetitive somatocrinin injections. Endocrinology 1125: 1378-1384.
Dunshea, F. R. 1993. Effect of metabolism modifiers on lipid metabolism in the pig. J. Anim. Sci. 71: 1966-
1977.
Dunshea, F. R. 2002. Metabolic and production responses to different porcine somatotropin injection
regimes in pigs. Aust. J. Agric. Res. 53: 785-793.
Dunshea, F. R. 2005. Sex and porcine somatotropin impact on variation in growth performance and back
fat thickness. Aust. J. Exp. Agric. 45: 677-682.
Dunshea, F. R. 2007. Porcine somatotropin and cysteamine hydrochloride improve growth performance
and reduce back fat in finisher gilts. Aust. J. Exp. Agric. 47: 796-800.
Dunshea, F. R. and N. J. Gannon. 1995. Nutritional and other factors affecting efficacy of β-agonists in pigs.
Recent Advances in Animal Nutrition in Australia 10: 46-52.
Dunshea, F. R. and R. H. King. 1995. Responses to homeostatic signals in ractopamine-treated pigs. Br. J.
Nutr. 73: 809-818.
Dunshea, F. R. and D’Souza, D. N. 2003. Fat metabolism in the pig. In ‘Manipulating Pig Production IX’, pp
127-150, ed j. Paterson. Australasian Pig Science Association: Werribee.

Dunshea, F. R., R. H. King, and R. G. Campbell. 1993a. Interrelationships between dietary protein and
ractopamine on protein and lipid deposition in finishing gilts. J. Anim. Sci. 71: 2931-2941.
Dunshea, F. R., R. H. King, R. G. Campbell, R. D. Sainz, and Y. S. Kim. 1993b. Interrelationships between sex
and ractopamine on protein and lipid deposition in rapidly growing pigs. J. Anim. Sci. 71: 2919-2930.
Dunshea, F. R., R. H. King, P. J. Eason, and R. G. Campbell. 1998a. Interrelationships between dietary
ractopamine, energy intake, and sex in pigs. Aust. J. Agric. Res. 49: 565-574.
Dunshea, F. R., B. J. Leury, and R. H. King. 1998b. Lipolytic responses to catecholamines in ractopamine-
treated pigs. Aust. J. Agric. Res. 49: 875-881.
Dunshea, F. R., C. Colantoni, K. Howard, P. Jackson, K. A. Long, S. Lopaticki, E. A. Nugent, J. A. Simons, J.
Walker, and D. P. Hennessy. 2001. Vaccination of entire boars with Improvac® eliminates boar taint and
increases growth performance. J. Anim. Sci. 79: 2524-2535.
Dunshea, F. R., M. L.Cox, M. R. Borg, M. N. Sillence, and D. R. Harris. 2002. Porcine somatotropin (pST)
administered using a commercial delivery system improves growth performance of rapidly-growing,
group-housed finisher pigs. Aust. J. Agric. Res. 53: 287-293.
Dunshea, F. R., D. J. Cadogan, and G. G. Partridge. 2009. Dietary betaine and ractopamine combine to
increase lean tissue deposition and betaine on body composition and tissue distribution in finisher pigs,
particularly gilts. Anim. Prod. Sci. 49: 65-70.
Dunshea, F. R., G. M. Cronin, J. L. Barnett, P. H. Hemsworth, D. P. Hennessy, R. G. Campbell, B. Luxford,
R. J. Smits, A. J. Tilbrook, R. H. King, and I. McCauley. 2011. Immunisation against gonadotrophin
releasing hormone (GnRH) increases growth and reduces variability in group-housed boars. Anim.
Prod. Sci. 51: 695-701.
Etherton, T. D., J. P. Wiggins, C. M. Evock, C. S. Chung, J. F. Rebhun, P. E. Walton, and N. C. Steele.
1987. Stimulation of pig growth performance by porcine growth hormone: Determination of the dose-
response relationship. J. Anim. Sci. 64: 433-443.
Evock, C. M., T. D. Etherton, C. S. Chung, and R. E. Ivy. 1988. Pituitary porcine growth hormone (pGH)
and a recombinant pGH analog stimualte pig growth performance in a similar manner. J. Anim. Sci.
66: 1928-1941.
Evock-Clover, C. M., M. M. Polansky, R. A. Anderson, and N. C. Steele. 1993. Dietary chromium
supplementation with or without somatotropin treatment alters serum hormones and metabolites in
growing pigs without affecting growth performance. J. Nutr. 123: 1504-1512.
Fernandez-Figares, I., D. Wray-Cahen, N. C. Steele, R. G. Campbell, D. D. Hall, E. Virtanen, and T. J.
Caperna. 2002. Effect of dietary betaine on nutrient utilization and partitioning in the young growing
feed-restricted pig. J. Anim. Sci. 80: 421-428.
Genicot, B., F. Mouligneau, and P. Lekeux. 1991. Efficiency of a sedative on the behaviour and the
performances of Belgian white and blue double-muscled cattle during fattening. J. Vet. Med. Series A
38: 668-675.
Giri, J., K. Usha, and T. Sunitha. 1990. Evaluation of the selenium and chromium content of plant foods.
Plant Foods Hum. Nutr. 40: 49-59.
Gu, Y., A. P. Schinckel, J. C. Forrest, C. H. Kuel, and L. E. Watkins. 1991a. Effects of ractopamine, genotype,
and growth phase on finishing performance and carcass value in swine: I. Growth performance and
carcass merit. J. Anim. Sci. 69: 2685-2693.
Gu, Y., A. P. Schinckel, J. C. Forrest, C. H. Kuel, and L. E. Watkins. 1991b. Effects of ractopamine, genotype,
and growth phase on finishing performance and carcass value in swine: II. Estimation of lean growth
rate and lean feed efficiency. J. Anim. Sci. 69: 2694-2702.

F.R. Dunshea
Hellams, S. J., W. L. Bryden, D. Harris, and F. R. Dunshea. 2000. Strategic use of growth hormone and pig
performance. Asian Austr. J. Anim. Sci. 13D: 10.
Hemsworth, P. H., J. L. Barnett, and R. G. Campbell. 1996. A study of the relative aversiveness of a new daily
injection procedure for pigs. App. Anim. Behav. Sci. 49: 389-401.
Hennessy, D., C. Ma, Z. Liu, Q. Wu, and H. Yang. 2009. The growth performance of male pigs vaccinated
with the boar taint vaccine, Improvac and the effects on boar taint assessment. Proceedings of the 49th
International Congress of Meat Science and Technology PE 1.26.
Hung, T. Y., B. J. Leury, T. F. Lien, and F. R. Dunshea. 2010. Potential of nano-chromium to improve body
composition and performance of farm animals. In ‘14th Asian-Australasian Association for Animal
Production,’ Taiwan, pp. 108-112.
Hung, T. Y., B. J. Leury, M. A. Sabin, T. F. Lien, and F. R. Dunshea. 2009. Nano- and micro-size chromium
picolinate increase carcases weight and muscle and decreases fat in finisher pigs. p. 183 in Manipulating
Pig Production XII. R. J. Van Barneveld, ed. Australasian Pig Science Association, Werribee, Australia.
Hung, T. Y., B. J. Leury, M. A. Sabin, T. F. Lien, and F. R. Dunshea. 2011a. Skeletal muscle gene expression
in response to dietary nano-chromium and fat supplementation in finisher gilts. In ‘Manipulating Pig
Production XIII’, ed R.j. van Barneveld. Australasian Pig Science Association: Werribee, pp 201.
Hung, T. Y., C. L. Collins, B. J. Leury, M. A. Sabin, and F. R. Dunshea. 2011b. Effect of nano-sized chromium
picolinate on growth and carcase traits in finisher gilts during summer. In ‘Manipulating Pig Production
XIII’, ed R.j. van Barneveld. Australasian Pig Science Association: Werribee, pp 61.
Israel, E., J. M. Drazen, S. B. Liggett, H. A. Boushey, R. M. Cherniack, V. M. Chinchilli, D. M. Cooper, J. V.
Fahy, J. E. Fish, J. G. Ford, M. Kraft, S. Kunselman, S. C. Lazarus, R. F. Lemanske Jr., R. J. Martin, D. E.
McLean, S. P. Peters, E. K. Silverman, C. A. Sorkness, S. J. Szefler, S. T. Weiss, and C. N. Yandava. 2001.
Effect of polymorphism of the beta(2)-adrenergic receptor on response to regular use of albuterol in
asthma. Int. Arch. Allergy Immun. 124: 183-186.
Jackson, A. R., S. Powell, S. L. Johnston, J. O. Matthews, T. D. Bidner, F. R. Valdez, and L. L. Southern. 2009.
The effect of chromium as chromium propionate on growth performance, carcass traits, meat quality,
and the fatty acid profile of fat from pigs fed no supplemented dietary fat, choice white grease, or tallow.
J. Anim. Sci. 87: 4032-4041.
Kim, B. G., M. D. Lindemann, and G. L. Cromwell. 2009. The effects of dietary chromium(III) picolinate on
growth performance, blood measurements, and respiratory rate in pigs kept in high and low ambient
temperature. J. Anim. Sci. 87: 1695-1704.
King, R. H., R. G. Campbell, R. J. Smits, W. C. Morley, K. Ronnfeldt, K. Butler, and F. R. Dunshea. 2000.
Interrelationships between dietary lysine, sex and porcine somatotropin administration on growth
performance and protein deposition in pigs between 80 and 120kg live weight. J. Anim. Sci. 78: 2639-
2851.
Klindt, J., F. C. Buonomo, and J. T. Yen,. 1992. Administration of porcine somatotropin by sustained-release
implant: growth and endocrine responses in genetically lean and obese barrows and gilts. J. Anim. Sci.
70: 3721-3733.
Krick, B. J., K. R. Roneker, R. D. Boyd, D. H. Beermann, P. J. David, and D. J. Meisinger. 1992. Influence of
genotype and sex on the response of growing pigs to recombinant porcine somatotropin. J. Anim. Sci.
70: 3024-3034.
Lien, T. F., H. S. Yeh, F. Y. Lu, and C. M. Fu. 2009. Nanoparticles of chromium picolinate enhance chromium
digestibility and absorption. J. Sci. Food Agric. 89: 1164-1167.
Liu, C. Y., and S. E Mills. 1989. Determination of the affinity of ractopamine and clenbuterol for the beta-
adrenoceptor of the porcine adipocyte. J. Anim. Sci. 67: 2937-2942.

Liu, G. M., Z. S. Wang, D. Wu, A. G. Zhou, and G. L. Liu. 2009. Effects of dietary cysteamine supplementation
on growth performance and whole-body protein turnover in finishing pigs. Livestock Sci. 122: 86-89.
Matthews, J. O., L. L. Southern, T. D. Bidner, and M. A. Persica. 2001a. Effects of betaine, pen space, and
slaughter handling method on growth performance, carcass traits, and pork quality of finishing barrows.
J. Anim. Sci. 79: 967-974.
Matthews, J. O., L. L. Southern, J. M. Fernandez, J. E. Pontif, T. D. Bidner, and R. L. Odgaard. 2001b. Effect
of chromium picolinate and chromium propionate on glucose and insulin kinetics of growing barrows
and on growth and carcass traits of growing-finishing barrows. J. Anim. Sci. 79: 2172-2178.
Matthews, J. O., L. L. Southern, A. D. Higbie, M. A. Persica, and T. D. Bidner. 2001c. Effects of betaine on
growth, carcass characteristics, pork quality, and plasma metabolites of finishing pigs. J. Anim. Sci. 79,
722-728.
McCauley, I., A. Billinghurst, P. O. Morgan, and S. L. Westbrook. 1995. Manipulation of endogenous
hormones to increase growth of pigs. pp. 52-61 in Manipulating Pig Production V. D. P. Hennessy, and
P. D. Cranwell, eds. Australasian Pig Science Association: Werribee, Australia.
McCauley, I., G. M. Cronin, R. H. King, P. H. Hemsworth, J. L. Barnett, B. Luxford, R. J. Smits, D. P. Hennessy,
R. G. Campbell, and F. R. Dunshea. 2003a. Dietary narcoleptics can improve the efficiency of growth in
group-housed boars. p. 83 in Manipulating Pig Production IX. J Paterson, ed. Australasian Pig Science
Association, Werribee, Australia.
McCauley, I., M. Watt, D. Suster, D. J. Kerton, W. T. Oliver, R. J. Harrell, and F. R. Dunshea. 2003b. An
immunocastration vaccine (Improvac®) and porcine somatotropin (Reporcin®) have synergistic effects
upon growth performance in both boars and gilts. Aust. J. Agric. Res. 54: 11-20.
McCauley, I., G. M. Cronin, R. H. King, P. H. Hemsworth, J. L. Barnett, B. Luxford, R. J. Smits, D. P. Hennessy,
R. G. Campbell, and F. R. Dunshea. 2004. Dietary narcoleptics and immunocastration improve growth
in group-housed boars. Asia Pac. J. Clin. Nutr. 13 (Suppl.): S89.
McLeod, K. R., D. L. Harmon., K. K. Schillo, S. M. Hileman, and G. E. Mitchell, Jr. 1995a. Effects of
cysteamine on pulsatile growth hormone release and plasma insulin concentrations in sheep. Comp.
Biochem. Physiol. B. Biochem. Mol. Biol. 112: 523-533.
McLeod, K. R., D. L. Harmon, K. K. Schillo, and G. E. Mitchell, Jr. 1995b. Cysteamine-induced depletion of
somatostatin in sheep: time course of depletion and changes in plasma metabolites, insulin, and growth
hormone. J. Anim. Sci. 73: 77-87.
Mersmann, H. J. 1998. Overview of the effects of beta-adrenergic receptor agonists on animal growth
including mechanisms of action. J. Anim. Sci. 76: 160-172.
Mertz, W. 1969. Chromium occurrence and function in biological systems. Physiol. Rev. 49: 163-180.
Mertz, W. 1993. Chromium in human nutrition: A review. Journal of NutritionJ. Nutri. 123: 626-633.
Mitchell, A. D., M. B. Solomon, and N. C. Steele. 1991. Influence of level of dietary protein or energy on
effects of ractopamine in finishing swine. J. Anim. Sci. 69: 4487-4495.
Mooney, K. W., and G. L. Cromwell. 1995. Effects of dietary chromium picolinate supplementation on
growth, carcass characteristics, and accretion rates of carcass tissues in growing-finishing swine. J. Anim.
Sci. 73: 3351-3357.
Moore, K. L., F. R. Dunshea, B. P. Mullan, D. P. Hennessy, and D. N. D’Souza. 2009. Ractopamine
supplementation increases lean deposition in entire and immunocastrated male pigs. Animal Production
ScienceAnim. Prod. Sci. 49: 1113-1119.
National Research Council. 1998. ‘Nutrient requirements of swine.’ National Academy Press: Washington,
DC.

F.R. Dunshea
Oliver, W. T., I. McCauley, R. J. Harrell, D. Suster, and F. R. Dunshea. 2003. A GnRF vaccine (Improvac®)
and porcine somatotropin have synergistic and additive effects on growth performance in group-housed
boars and gilts, respectively. J. Anim. Sci. 81: 1959-1966.
Page, T. G., L. L. Southern, T. L. Ward, and D. L. Thompson, Jr. 1993. Effect of chromium picolinate on
growth and serum and carcass traits of growing-finishing pigs. J. Anim. Sci. 71: 656-662.
Pethick, D. W., R. D. Warner, D. N. D’Souza, and F. R. Dunshea. 1997. Nutritional manipulation of meat
quality. pp. 91-99 in Manipulating Pig Production VI. P. D. Cranwell, ed. Australasian Pig Science
Association: Werribee, Australia.
Rikard-Bell, C., M. A. Curtis, R. J. van Barneveld, B. P. Mullan, A. C. Edwards, N. J. Gannon, D. J. Henman,
P. E. Hughes, and F. R. Dunshea. 2009a. Ractopamine hydrochloride improves growth performance
and carcass composition in immunocastrated boars, intact boars, and gilts. J. Anim. Sci. 87: 3536-3543.
Rikard-Bell, C. V., J. R. Pluske, R. J. van Barneveld, B. P. Mullan, A. C. Edwards, N. J. Gannon, D. J. Henman,
and F. R. Dunshea. 2009b. Combining a ractopamine feeding regime and porcine somatotropin has
additive effects on finisher pig peformance. In ‘Manipulating Pig Production XII’, ed R.j. van Barneveld.
Australasian Pig Science Association: Werribee, pp 70.
Rikard-Bell, C. V., G. Nattrass, J. R. Pluske, R. J. van Barneveld, B. P. Mullan, A. C. Edwards, N. J. Gannon,
D. J. Henman, and F. R. Dunshea. 2011. Ractopamine effects the β-1 and β-2 adrenergic receptor gene
expression in fat and muscle tissue of boars and gilts. In ‘Manipulating Pig Production XIII’, ed R.j. van
Barneveld. Australasian Pig Science Association: Werribee, pp 90.
Roberts, C. J., Q. Ji, L. Zhang, and R. T. Darrington. 2001. Dissolution behavior of porcine somatotropin with
simultaneous gel formation and lysine Schiff-base hydrolysis. J. Contr. Release 77: 107-116.
Roginski, E. E., and W. Mertz. 1969. Effects of chromium 3+ supplementation on glucose and amino acid
metabolism in rats fed a low protein diet. J. Nutri. 97: 525-530.
Rydhmer, L., K. Lundstrom, and K. Andersson. 2010. lmmunocastration reduces aggressive and sexual
behaviour in male pigs. Animal 4: 965-972.
Rydhmer, L., G. Zamaratskaia, H. K. Andersson, B. Algers, R. Guillemet, and K. Lundstrom. 2006.
Aggressive and sexual behaviour of growing and finishing pigs reared in groups, without castration.
Acta Agriculturae Scandinavica Section A-Animal Science 56: 109-119.
Sainz, R. D., Y. S. Kim, F. R. Dunshea, and R. G. Campbell. 1993. Temporal changes in growth enhancement
by ractopamine in pigs: Performance aspects. Aust. J. Agric. Res. 44: 1449-1455.
Sales, J. 2011. A meta-analysis of the effects of dietary betaine supplementation on finishing performance
and carcass characteristics of pigs. Anim. Feed Sci. Tech. 165: 68-78.
Sales, J., and F. Jancik. 2011. Effects of dietary chromium supplementation on performance, carcass
characteristics, and meat quality of growing-finishing swine: a meta-analysis. J. Anim. Sci. 89: 4054-
4067.
Schinckel, A. P., B. T. Richert, C. T. Herr, M. E. Einstein, and D. C. Kendall. 2001. Effects of ractopamine
on swine growth, carcass composition and quality. In Second International Virtual Conference on Pork
Quality.
Schrama, J. W., M. J. Heetkamp, P. H. Simmins, and W. J. Gerrits. 2003. Dietary betaine supplementation
affects energy metabolism of pigs. J. Anim. Sci. 81: 1202-1209.
Sechen, S. J., F. R. Dunshea, and D. E. Bauman. 1990. Somatotropin in lactating cows: effect on response to
epinephrine and insulin. Amer. J. Physiol. 258: E582-E588.
Smits, R. J., and D. J. Cadogan. 2003. The use of ractopamine as the commercial product, Paylean®, for the
Australian pig industry. Recent Advances in Animal Nutrition in Australia 14: 143-150.

Steele, N. C., and R. W. Rosebrough. 1979. Trivalent chromium and nicotinic acid supplementation for the
turkey poult. Poult. Sci. 58: 983-984.
Steele, N. C., T. G. Althen, and L. T. Frobish. 1977. Biological activity of glucose tolerance factor in swine.
J. Anim. Sci. 45: 1341-1345.
Suster, D., B. J. Leury, R. H. King, M. Mottram, and F. R. Dunshea. 2004. Interrelationships between porcine
somatotropin (pST), betaine, and energy level on body composition and tissue distribution of finisher
boars. Aust. J. Agric. Res. 55: 983-990.
Suster, D., B. J. Leury, R. Hewitt, D. J. Kerton, and F. R. Dunshea. 2005. Porcine somatotropin (pST) alters
body composition and the distribution of fat and lean tissue in the finisher gilt. Austr. J. Exper. Agri.
45: 683-690.
Tannenbaum, G. S., G. F. McCarthy, P. Zeitler, and A. Beaudet. 1990. Cysteamine-induced enhancement of
growth hormone-releasing factor (GRF) immunoreactivity in arcuate neurons: morphological evidence
for putative somatostatin/GRF interactions within hypothalamus. Endocrinology 127: 2551-2560.
Turman, E. J., and F. N. Andrews. 1955. Some effects of purified anterior pituitary growth hormone on
swine. J. Anim. Sci. 14: 7-18.
Underwood, E. J. 1977. Chromium. pp. 258-270 in Trace elements in human and animal nutrition. E. J.
Underwood, ed. Academic Press: New York, NY, USA.
Van Barneveld, R. J., R. J. E. Hewitt, A. Cook, C. V. Rikard-Bell, J. R. Pluske, B. P. Mullan, A. C. Edwards, N.
J. Gannon, D. J. Henman, and F. R. Dunshea. 2009. The synergistic effects of ractopamine and porcine
somatotropin on finisher performance. In ‘Manipulating Pig Production XII’, ed R.j. van Barneveld.
Australasian Pig Science Association: Werribee, pp 50.
Wakabayashi, I., Y. Tonegawa, T. Shibasaki, T. Ihara, M. Hattori, and N. Ling. 1985. Effect of dopamine,
bombesin and cysteamine hydrochloride on plasma growth hormone response to synthetic growth
hormone-releasing factor in rats. Life Science 36: 1437-1443.
Watkins, L. E., D. J. Jones, D. H. Mowrey, D. B. Anderson, and E. L. Veenhuizen. 1990. The effect of various
levels of ractopamine hydrochloride on the performance and carcass characteristics of finishing swine.
J. Anim. Sci. 68(11): 3588-3595.
Wray-Cahen, D., H. V. Nguyen, D. G. Burrin, P. R. Beckett, M. L. Fiorotto, P. J. Reeds, T. J. Wester, and T.
A. Davis. 1998. Response of skeletal muscle protein synthesis to insulin in suckling pigs decreases with
development. Am. J. Physiol. 275: E602-609.
Xiao, D., and H. R. Lin. 2003. Cysteamine-a somatostatin-inhibiting agent-induced growth hormone
secretion and growth acceleration in juvenile grass carp (Ctenopharyngodon idellus). Gen. Comp.
Endo. 134: 285-295.
Yang, C-B., A-K. Li, Y-L. Yin, R-L Huang, T. J. Li, L-L. Li, Y-P. Liao, Z-Y. Deng, J. Zhang, B. Wang, Y-G.
Zhang, X. Yang, J. Peng, and M. Z. Fan. 2005. Effects of dietary supplementation of cysteamine on
growth performance, carcass quality, serum hormones and gastric ulcer in finishing pigs. J. Sci. Food
Agri. 85: 1947-1952.
Yen, J. T., J. A. Nienaber, J. Klindt, and J. D. Crouse. 1991. Effect of ractopamine on growth, carcass traits,
and fasting heat production of U.S. contemporary crossbred and Chinese Meishan. J. Anim. Sci. 69:
4810-4822.
Zhang, H., B. Dong, M. Zhang, and J. Yang. 2011. Effect of chromium picolinate supplementation on growth
performance and meat characteristics of swine. Biol. Trace Element Res. 141: 159-169.
Zhou, J., E. X-H. Wei, D. Xia, J-X. Xu, T-S. Lu, N. Chen, R-Q. Zhao. 2005. Effect of cysteamine on fat
deposition in finishing pigs. Chin. J. Agric. Biotech. 2: 15.

011B - Feed Efficiency in Swine PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

011B - Feed Efficiency in Swine PDF

Uploaded by

Copyright:

Available Formats

Feed efficiency

edited by: John F. Patience

This work is subject to copyright. All rights are

The publisher is not responsible for possible

1. Herd management factors that influence whole herd feed efficiency 15

2. Feeding and barn management strategies that maximize feed efficiency 41

Feed efficiency in swine 7

4. Amino acid nutrition and feed efficiency 81

5. The influence of dietary energy on feed efficiency in grow-finish swine 101

6. Feed processing to maximize feed efficiency 131

8 Feed efficiency in swine

8. Pig breeding for improved feed efficiency 167

9. Effect of climatic environment on feed efficiency in swine 183

10. Fueling the immune response: what’s the cost? 211

11. Influence of health on feed efficiency 225

Feed efficiency in swine 9

Relationships between dietary protein source and post-weaning diarrhea 247

10 Feed efficiency in swine

J.F. Patience, editor

Feed efficiency in swine 13

Feed efficiency in swine 15

Are we measuring feed conversion correctly?

Item DDGS (%) SEM P-value

Start weight (kg) 5.86 5.86 5.81 0.05 NS

16 Feed efficiency in swine

How do we measure ourselves over time?

Feed efficiency in swine 17

Table 2. Finishing closeout-comparison based on 2007 data (Goodband et al., 2008)1.

Item Grower 1 Grower 2

Weight in (kg) 25.9 20.4

1 Closeouts are adjusted for initial and final weight only.

Adjusted F:G = ( Observed F:G + (Standardized sw – Actual sw) ×

F:G = Feed:Gain ratio; sw = start weight; fw = final weight; el = energy level

18 Feed efficiency in swine

Observed Feed:Gain = 3.00

FTF adjusted for mortality (live)

Feed efficiency in swine 19

Item Grower 1 Grower 2

Weight in (kg) 25.9 20.4

1 Closeouts are adjusted for initial and final weight only.

2 Adjustment is based on mortality impact on G:F.

Is it time to shift our thinking?

20 Feed efficiency in swine

PIC energy eff. PIC FCR

Figure 2. Caloric efficiency vs. feed efficiency expressed on a carcass basis.1

NRC ME; 76% yield factor.

Figure 3. Impact of sow productivity on sow feed per wean pig.1

Feed efficiency in swine 21

Sow feed per wean kilograms produced

Whole herd carcass FCR

22 Feed efficiency in swine

Sow replacement rate

Feed efficiency in swine 23

Treatment Preweaning Weaner Grower Finisher

Progeny of gilts1 14.4% 5.6% 3.9% 6.7%

a percentage of the number of pigs at the start of each growth phase.

Scenario 1 18,720 2,349 21,767 15,420 127.2 1,170.6 3.86

24 Feed efficiency in swine

Scenario 1 18,720 2,349 21,767 15,420 127.2 1,170.6 3.86

Feed efficiency in swine 25

Impact of birth weight on feed efficiency

26 Feed efficiency in swine

Figure 8. Impact of pigs/sow/year on birth weight (Source: the Maschhoffs).

Feed efficiency in swine 7

8 Feed efficiency in swine

Feed efficiency in swine 9

10 Feed efficiency in swine

Feed efficiency in swine 13

Feed efficiency in swine 15

16 Feed efficiency in swine

Feed efficiency in swine 17

Adjusted F:G = ( Observed F:G + (Standardized sw – Actual sw) ×

18 Feed efficiency in swine

Feed efficiency in swine 19

20 Feed efficiency in swine

Feed efficiency in swine 21

22 Feed efficiency in swine

Feed efficiency in swine 23

24 Feed efficiency in swine

Feed efficiency in swine 25

26 Feed efficiency in swine

Feed efficiency in swine 27

28 Feed efficiency in swine

Feed efficiency in swine 29

30 Feed efficiency in swine

Feed efficiency in swine 31

32 Feed efficiency in swine

Feed efficiency in swine 33

34 Feed efficiency in swine

Feed efficiency in swine 35

36 Feed efficiency in swine

Feed efficiency in swine 37

38 Feed efficiency in swine

Feed efficiency in swine 39

Feed efficiency in swine 41

42 Feed efficiency in swine

Feed efficiency in swine 43

44 Feed efficiency in swine

Feed efficiency in swine 45

46 Feed efficiency in swine

Feed efficiency in swine 47

48 Feed efficiency in swine

Feed efficiency in swine 49

50 Feed efficiency in swine