558641

research-article2014
CRE0010.1177/0269215514558641Clinical RehabilitationGuerra et al.

CLINICAL
Article REHABILITATION

Clinical Rehabilitation

Early mobilization of patients 2015, Vol. 29(9) 844854

The Author(s) 2014
Reprints and permissions:
who have had a hip or knee joint sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0269215514558641

replacement reduces length of stay cre.sagepub.com

in hospital: a systematic review

Mark L Guerra1, Parminder J Singh2 and

Nicholas F Taylor3,4

Abstract
Objective: To systematically review the effect of early mobilization after hip or knee joint replacement
surgery on length of stay in an acute hospital.
Methods: Randomized controlled trials were selected from electronic databases based on inclusion
criterion requiring an experimental group mobilizing (sitting out of bed/walking) earlier than a comparison
group post joint replacement surgery of the hip or knee in an acute hospital. Clinically homogeneous data
were analyzed with meta-analysis.
Results: Five randomized controlled trials (totaling 622 participants) were included for review. A meta-
analysis of 5 trials found a reduced length of stay of 1.8 days (95% confidence interval 1.1 to 2.6) in favor
of the experimental group. In 4 of the 5 trials the experimental group first sat out of bed within 24 hours
post operatively. In 4 of the 5 trials the experimental group first walked within 48 hours post operatively.
Individual trials reported benefits in range of motion, muscle strength and health-related quality of life in
favor of the experimental group. There were no differences in discharge destinations, incidence of negative
outcomes or adverse events attributable to early mobilization when compared to the comparison groups.
Conclusion: Early mobilization post hip or knee joint replacement surgery can result in a reduced length
of stay of about 1.8 days. Trials that reported these positive results showed that early mobilization can
be achieved within 24 hours of operation. This positive gain was achieved without an increase in negative
outcomes.

Keywords
Early mobilization, knee joint replacement, hip joint replacement, length of stay

Received: 17 December 2013; accepted: 12 October 2014

1Physiotherapy Department, Eastern Health, Melbourne, Corresponding author:

Victoria, Australia
2Orthopaedic Consultant, Eastern Health, Melbourne,
Victoria, Australia
3Allied Health Clinical Research Office, Eastern Health, Australia
4Department of Physiotherapy, La Trobe University, Australia
Mark L Guerra, Physiotherapy Department, Eastern Health,
Maroondah Hospital, Davey Drive, Ringwood East, Victoria,
3135, Australia.
Email: mark.guerra@easternhealth.org.au
Guerra et al. 845
Introduction
Osteoarthritis is one of the most common chronic
diseases. Worldwide the number of total hip and
knee replacements being performed is rising stead-
ily and continues to be a common surgical tech-
nique for the treatment of osteoarthritis.1-3
Once medically and surgically stable, a focus of
the acute management of the orthopedic inpatient
population is discharge planning and being able to
safely discharge patients from hospital in the
shortest time.4,5 This focus on early safe discharge
is driven by an imperative from the health service
to have a more efficient service and from the
patient view enables them to return to living inde-
pendently in the community as soon as possible.
Early mobilization, defined as getting out of bed
and/or walking as close to the time of surgery as
possible, can reduce the risks associated with bed
rest such as deep vein thrombosis, pulmonary
embolus, chest infection and urinary retention.6
Early mobilization has been also shown to improve
indicators of functional status and cardiovascular
outcomes.7 Conversely, possible negative effects
of early mobilization such as low hemodynamics
with an increased risk of collapse, increased pain
levels, increased sickness and nausea levels must
be considered. The possible benefits and risks of
early mobilization highlight the clinician dilemma
of how quickly patients can be safely mobilized
after surgery.
In the absence of an existing systematic review
documenting the effect of early mobilization the
primary aim of this systematic review was to find
out if mobilizing a patient early has benefits
in reducing length of stay after primary joint
replacement surgery for knee and hip osteoarthri-
tis. We hypothesized that the earlier the patient
gets out of bed and walking after their surgery the
earlier they can be discharged from acute care.
The secondary aim was to evaluate the effects
of early mobilization on clinical outcomes and
adverse events.
Method
This systematic review protocol was registered
with PROSPERO (CRD42012003034).
Without time or date restrictions, we applied our
search strategy to the electronic databases of
Medline, Cinahl and Embase and searched up until
September 2014. We used the concepts of popula-
tion, intervention and design to create our
search terms (Table 1). The concept of population
was limited to primary joint replacement of the hip
or knee and synonyms. The concept of intervention
was built from that of early mobilization. We lim-
ited our search to randomized controlled trials in an
attempt to extract the highest quality evidence.
Search terms, including key words and MeSH
terms were entered separately for the population,
intervention and design concepts and synonyms
combined with the OR operator. Finally we com-
bined these results for the population, intervention
and design with the AND operator. Database
searching was supplemented by searching refer-
ence lists and citation tracking of included trials.
Trials were eligible if there was a primary total
joint replacement of the hip or knee, or a primary
knee unicompartmental joint hemiarthroplasty
(Table 2). Trials were eligible if their design was a
randomized controlled trial published in English
where the experimental group was mobilized (get-
ting out of bed/first walking) sooner than the com-
parison group regardless of what day this occurred
on. Eligibility criteria were applied to title and
abstract by two reviewers independently. After
retrieving full text copies of articles for further
consideration, the eligibility were re-applied. At
all stages differences of opinion were resolved
through discussion and consensus was achieved
without the need to involve a third reviewer.
Data extraction was based on the Cochrane
reviews guide and then modified into a final version
for our purposes.8 Data were extracted on patient
demographics: sex, age, diagnosis, and surgery
type. Data were also extracted on trial characteris-
tics: description of physiotherapy treatment, when
the patient first walked or sat out of bed, and any
co-interventions. Finally data were extracted on
study outcomes: primary outcome of length of stay
and secondary outcomes of: impairments, activity
limitations or participation restrictions and adverse
events. The primary reviewer completed data
extraction and a second reviewer checked the accu-
racy of data extraction.
846 Clinical Rehabilitation 29(9)
Table 1. Search Strategy.
Population AND
hip replacement, OR
knee replacement, OR
primary hip replacement, OR
primary knee replacement, OR
primary hip arthroplasty, OR
primary knee arthroplasty, OR
hip arthroplasty, OR
knee arthroplasty
Table 2. Inclusion and exclusion criteria.
Population
Intervention
Comparison/control
Outcome
Design
Other
JRS: joint replacement surgery.
Intervention AND Design
true early ambulation, OR random* controlled trial
time to first walk*, OR
early ambulation, OR
early mobili*, OR
early mobili* protocol, OR
enhanced total hip replacement recovery, OR
enhanced total knee replacement recovery, OR
enhanced recovery pathway, OR
rapid recovery, OR
rapid recovery pathway, OR
early amb*, OR
accelerated rehab* protocol*, OR
fast track rehab*, OR
first walk* after knee arthroplasty, OR
first walk* after hip replacement, OR
first walk* after hip arthroplasty, OR
first walk* after knee replacement
Inclusion Exclusion
Total joint replacement surgery (JRS) (Hip JRS (e.g.: hemiarthroplasty) where
or knee); JRS - unicompartmental (knee) only one surface is replaced; hip/knee
fractures that do not progress to JRS;
revision JRS
Early mobilization: experimental group Delayed mobilization: as defined:
either walked OR first sat out of bed prior experimental group NOT first
to comparison group walked OR sitting out of bed prior to
comparison group
Delayed mobilization: defined as NOT first
walked OR sitting out of bed until day 4
(or later) post operation
Length of stay: other measures relating
(but not limited) to: impairment; pain
scores; activity; length of stay; other
quantitative or qualitative scales/measures;
discharge destination; and adverse events
Randomized controlled trial Other designs e.g.: systematic
reviews; thesis; book chapters;
commentaries; opinions
English language Not in English - all or part
Guerra et al. 847
Two researchers independently applied a vali-
dated scale (PEDro) to rate the risk of bias of the
trials.9 The 11 items are based upon the Delphi list.10
Each item is scored yes or no with a maximum egy, 31 potentially relevant full-text articles were
score of ten as criterion one is not scored. The
PEDro score has sufficient inter-rater reliability for
clinical trials.11 A trial with a score of six or more
was considered to be of high quality.12,13
Meta-analysis was performed with a random
effects model using inverse variance analysis with
the software Review Manager (RevMan) Version
5.1.4 for the primary outcome of length of stay.
Statistical heterogeneity was assessed with I2 sta-
tistic with values less than 25% considered low
levels of heterogeneity.14 If means or standard
deviations were not reported they were calculated
from raw data, or estimated.8,15
The Grading of Recommendations Assessment,
Development and Evaluation (GRADE) approach
was applied to the meta-analysis to determine the
quality of evidence across the included trials.16
When using this approach, all randomized con-
trolled trials are considered high quality, or at low
risk of bias. The level is then downgraded to
moderate and again to low or very low if limita-
tions are present.
We downgraded the evidence if any of the fol-
lowing occurred: imprecision of results (wide con-
fidence interval); unexplained heterogeneity or
inconsistent results (we applied a figure of 25%
for the I value whereby if the figure for I value
was too high (above 25%) we would downgrade
the quality of evidence by one level); and limita-
tions in the design and implementation of availa-
ble studies suggesting likelihood of bias indicated
by a score of <6 on the PEDro scale for the major-
ity of trials in the meta-analysis. In the case of a
high level of statistical heterogeneity represented
by I figure > 25% we performed a sensitivity
analysis, removing any trials of lesser quality from
the analysis.
Secondary clinical outcomes such as range of
motion and muscle strength and adverse events
were analysed descriptively and between-group
differences reported. Secondary outcomes did not
undergo meta-analysis due to clinical heterogene-
ity in these.
Results
Of the 1300 articles identified by the search strat-
retrieved to evaluate eligibility (Figure 1). Twenty-
six of these papers were excluded (see supplemen-
tary material Appendix A and B) giving a final
library of 5 articles identified for review.17-21
Patient characteristics are described in Table 3.
The average age of patients in the 5 trials ranged
from 63 to 68 years. Overall, there were 404
women and 218 men included in the 5 trials.
Almost all patients (>99%, 618/622) had a diagno-
sis of osteoarthritis. Across the 5 trials there were
114 total hip arthroplasties, 300 total knee arthro-
plasties, and 45 unicompartmental hemiarthroplas-
ties leaving 163 patients which had either a total
hip arthroplasties or total knee arthroplasties.17
While not clearly reported, it is assumed that all
knee trials had at least tibio-femoral (both medial
and lateral) osteoarthritis with the exception of
one study which included only antero-medial tibio-
femoral compartment osteoarthritis.19
There was heterogeneity in type of surgery
performed. One trial20 included all total hip
replacement, total knee replacement and medial
unicompartmental knee joint replacement surgery.
One trial17 included both total hip replacement and
total knee replacement joint replacement surgery.
One trial18 included only total hip replacement
joint replacement surgery. One trial19 included only
medial unicompartmental knee joint replacement
surgery. One trial21 included only total knee
replacement joint replacement surgery.
Four of the 5 experimental groups were mobi-
lized (either sitting out of bed or walking) on the
day of surgery (the fifth on day 1 post operation).
The comparison groups took approximately until
day 2 post operation to start mobilization. All
experimental groups had defined goal orientated
post operative care plans. This was commonly
referred to as a pathway. The comparison groups
had either: a relaxed version of the experimental
pathway.18,20 or reactive treatment17 or the same
care plan as for the experimental group but delayed
by 48-72 hours,21 or the same care without the
urgency of the experimental group.19
848 Clinical Rehabilitation 29(9)
Trials idenfied through database
searching: n=1600
(Cinahl: 945 Medline: 563
Embase: 92)
Duplicates removed (n = 303)
Titles and abstracts screened
(n = 1300)
Potenally relevant full-text trials
retrieved to evaluate eligibility
(n = 31)
Trials included in the review
(n = 5)
Figure 1. Yield of studies.
Most trials reported offering both groups some
form of pre-operative information sessions. The
purpose was to inform the patient of the planned
peri-operative pathway, discharge plans and post
discharge pathway. Most trials had daily physio-
therapy sessions for both groups. Overall, there did
not appear to be any significant difference in the
type of interventions between the groups apart
from the timing of mobilization. There did not
appear to be any additional co-interventions
Addional trials idenfied through
reference scanning and citaon tracking
(n = 3)
Trials excluded based on tle and abstract
(n = 1269)
Papers excluded aer evaluaon
of full-text (n = 26)
- All paents mobilized at same
me (n= 16)
- Doctoral thesis/review (n=4)
- Non RCT study design/using data
from an included RCT (n=3)
- Commentary/arcle based about
an included RCT (n=3 )
afforded to the experimental groups compared to
the comparison groups.
There were 4 high quality trials18-21 and one
trial of lesser quality17 with an average PEDro
score of 7.6 across the 5 studies (supplementary
material Table 4). Participant and therapist blind-
ing was not possible as each patient and thera-
pist knew their expected peri-operative journey/
expectations so that the maximum expected
PEDro score was 8.
 Guerra et al.

Table 3. Trial Characteristics.

Trial Diagnosis Participants - Participants-control Intervention- Intervention- Type of surgery - Type of surgery
experimental experimental control experimental control
Dowsey etal., 199917 OA n=92 n=71 Dedicated Reactive Either THR/A or Either THR/A
(Australia) Women: Women: rehab treatment TKR/A (not stated or TKR/A (not
n=107/163(overall) n=107/163(overall) pathway which) stated which)
Age: Age:
64.2 (0.4) 68.23 (0.4)
Labraca etal., 201121 OA n=138 n=135 Rehab onset Rehab onset TKR only TKR only
(Spain) women: Women: within first within first R TKR: n=89 L R TKR: n=79 L
n=101/138 n=110/135 24 hours 48-72 hours TKR: n= 49 TKR: n= 56
Age: 65.5(4.8) Age: 66.2(5.0) post op post op
Larsen etal., 200820 OA n=45 n=42 Accelerated Standard Either THR, Either THR,
(Denmark) Women: Women: protocol protocol TKR or TKR or
n=25/45 n=19/42 unicompartmental unicompartmental
Age: 64.0(0.8) Age: 66.0(9.2) knee knee
hemiarthroplasty hemiarthroplasty
THR: n=28 TKR: THR: n=28 TKR:
n= 15 UKR: n=2 n= 12 UKR: n=2
Reilly etal., 200519 OA n=21 n=20 Accelerated Standard All All
(UK) Women: Women: protocol(Aim protocol unicompartmental unicompartmental
n=17/41(overall) n=17/41(overall) for discharge knee knee
Age: 63 Age: 63 at 24 hours) hemiarthroplasty hemiarthroplasty
Whitney and Parkman OA n=31 n=27 Enhanced standard post All THR/A. All THR/A.
200418 (USA) Women: 12/31 Women: 13/27 standard post THR pathway Cemented: n= 19 Cemented: n= 17
Age: 66(8) Age: 65.5(9.4) THR pathway Uncemented: Uncemented:
n=12 n=10

OA: Osteoarthritis; TKR/A: Total Knee Replacement/Arthroplasty; THR/A: Total Hip Replacement/Arthroplasty; Rehab: rehabilitation; R: right; L: left.
849
850 Clinical Rehabilitation 29(9)

Experimental Control Mean Difference Mean Difference

Study or Subgroup Mean [days] SD [days] Total Mean [days] SD [days] Total Weight IV, Random, 95% CI [days] IV, Random, 95% CI [days]
Dowsey 1999 7.1 0.4 92 8.6 0.4 71 23.1% -1.50 [-1.62, -1.38]
Labraca 2011 6.4 1.2 138 8.5 2.6 135 21.2% -2.10 [-2.58, -1.62]
Larsen 2008 4.9 2.4 45 7.8 2.1 45 17.0% -2.90 [-3.83, -1.97]
Reilly 2005 1.5 0.7 21 4.3 1.3 20 19.8% -2.80 [-3.44, -2.16]
Whitney 2004 5.4 1.6 31 5.4 1.2 27 19.0% 0.00 [-0.72, 0.72]

Total (95% CI) 327 298 100.0% -1.84 [-2.57, -1.10]

Heterogeneity: Tau = 0.61; Chi = 45.83, df = 4 (P < 0.00001); I = 91%
-4 -2 0 2 4
Test for overall effect: Z = 4.88 (P < 0.00001) Favours experimental Favours control

Figure 2. Length of stay.

Synthesis of results manipulation. The most adverse events within a

Meta-analysis of the 5 trials with 622 participants
provided moderate quality evidence that early
mobilization, compared to control, was effective in
reducing length of stay in an acute healthcare facil-
ity by 1.8 days (95% CI 1.1 to 2.6) (Figure 2). The
level of evidence was downgraded from high
to moderate as the I value was over 25%.
Completion of a sensitivity analysis with exclusion
of one lesser quality trial17 resulted in a similar esti-
mate of the effect of length of stay (mean differ-
ence 1.9 days (95% CI 0.7 to 3.2), but the I2 value
remained high at 92%.
Impairment outcomes (range of motion, gait,
balance and muscle strength) showed greater gains
in favour of the experimental group when
compared to the comparison group in two trials
(Table 5).19,21 Activity outcomes (Oxford knee
assessment score and visual analogue scale) and
health-related quality of life also showed gains in
favour of the experimental group when compared
to the comparison group in three trials.19-21 A
patient satisfaction scale showed no between-group
differences when patients were asked about their
experiences of admission and post operation recov-
ery phase.22
Where reported, there was no difference in the
discharge destinations between the experimental
and control groups.
Where reported there was no significant differ-
ences for adverse events between the experimental
and control groups. Adverse events included post
operative joint infection, vascular compromise
(namely deep vein thrombosis of the calf), and
stiffness in knee range of motion requiring
trial was 11% of the experimental group (n=10/92
patients) and 28% of the comparison group
(n=28/71 patients)17. The comparison group in the
Dowsey trial had more calf deep vein thromboses
(3 compared to 1), more wound infections (4 com-
pared to 1), reduced range of motion (7 events
compared to zero) than the experimental group.
Some of these required re-admission to the acute
healthcare facility.
Discussion
The results of this systematic review provide mod-
erate quality evidence that early mobilization in
the first 24 hours after hip or knee joint replace-
ment surgery was effective in reducing length of
stay in an acute healthcare facility.
Length of stay was reduced on average by 1.8
days. This figure is large when put into context of
average length of stay of patients after joint
replacement surgery. For example in 2012/13 the
state of Victoria, Australia had and average length
of stay for total knee replacements of 6.7 days and
of 8.2 days for total hip replacements potentially
representing a saving of approximately 20% in
bed-day terms.23
Importantly, the benefit in reduced length of stay
was not achieved at the expense of other outcomes.
There were no observed differences in discharge
destination and in adverse outcomes between the
groups. In fact, a number of the included trials
reported other benefits from mobilizing earlier after
lower limb joint replacement surgery such as
improved lower limb range of motion and muscle
strength, and improved health-related quality of life.
Table 5. Outcome Measures and Results.

Trial Primary outcome measure: Secondary outcome measures Results Results

Length of stay days Experimental Comparison
Guerra et al.

mean(SD)
Dowsey et al 1999 Experimental: 7.1(0.4) Time to SOOB 1.9 days 3.4 days
(Australia) Comparison: 8.6 (0.4) First walked 2.2 days 3.6 days
Complications n=10 n=20
Match to planned DC n=64/92 n=43/71
destination
Labraca et al 2011 Experimental: 6.4(1.2) First SOOB Day 0 post op Day 2 post op
(Spain) Comparison: 8.5(2.6) First walked Day 1 post op Day 2 post op
Complications Nil reported Nil reported
Joint ROM More flexion 16.3(11.4) than control
More extension 2.1(3.2) than control
Muscle strength* Increased quadriceps by 0.98 (0.54)
Pain Increased hamstring by 1.05 (0.72)
Autonomy Reduced pain by 2.36(2.47) VAS
Fewer sessions until medical
discharge by 5.0(2.3) than control
Gait Higher scores for gait (P<0.047) and
Balance balance (P<0.045) than control

Larsen et al 2008 Experimental: 4.9(2.4) Health related QOL at 3 Improved 0.4(0.3) Improved 0.3(0.3)
(Denmark) Comparison: 7.8(2.1) months Day 0 post op Day 1 post op
First SOOB Day 0 post op Day 1 post op
First walked 2 2
Complications
Reilly et al 2005 Experimental: 1.5(0.7) Pain (VAS) at 6 months 43.5(9.5) 44.7(9.5)
(U.K.) Comparison: 4.3(1.3) Knee flexion () 124.7(5.5) 119.8(6.8)
Knee extension () -3.1(3.5) -2.3(3.5)
Patient satisfaction (n very 19/21 18/20
satisfied) n=2 n=1
Complications Day 0 post op Day 1 post op
First SOOB Day 0 post op Day 1 post op
First walked 43.7(3.7) 42.2(7.1)
Change in Oxford Knee
851

Assessment
852 Clinical Rehabilitation 29(9)

In addition to the health service and clinical

Abbreviations: SOOB: sit out of bed; ROM: range of motion; VAS: Visual analogue scale; QOL: Quality of life. *: strength measured by numerical scale 0-5 where 0 means nil
benefits, patients who mobilized early and had
accelerated discharge following elective ortho-
pedic surgery expressed satisfaction with their

Day 1 post op
Day 2 post op
Nil reported
Comparison
management.22
Results

The results of this systematic review suggest that

delaying mobilization on an otherwise functionally
ready patient does affect length of stay as well as
potentially exposing the patient to risks of morbid-
ity associated with bed rest. Concerns of mobilizing
a patient too early because of impaired hemody-
namics or other patient compromise may be mis-
placed. In practice clinicians should be encouraged
by these findings to aim at early mobilization, in
particular the first 24 hours, post primary hip or
knee joint arthroplasty for osteoarthritis. We com-
bined and collectively analyzed all included hip and
Day 0 post op
Day 1 post op
Experimental

Nil reported

knee joint replacement surgery types in our review

Results

as we considered that the issues regarding early

mobilization were similar. Ultimately, the exact
time to first mobilize a patient will be a clinical
judgment call made by clinicians at the bedside
Secondary outcome measures

with all information available to them.

To assist the patient and clinician, peri and post
operative analgesic techniques are ever advancing.
Differing modes of delivery/infiltration and/or
types of analgesic agents aim to promote an
Complications

increased ability for the patient to mobilize earlier

First walked
First SOOB

(for example by reducing pain associated with

movement or reducing post operative nausea)
which may further increase the likelihood of achiev-
ing safe and effective early mobilization.24-27
Primary outcome measure:

Our findings are similar to other evidence from

other research designs not included in our review. An
Experimental: 5.4(1.6)
Comparison: 5.4(1.2)
Length of stay days

observational design found initiating physiotherapy

in total hip arthroplasty on day of surgery reduced
length of stay by 0.2 days.28 Most (67%) patients
were discharged in 3 days or less compared to initiat-
mean(SD)

ing physiotherapy on day 1 after surgery (where only

57% were discharged in 3 days or less).28 Importantly,
activity and 5 normal activity

a shortened length of stay did not delay the achieve-

Table 5.(Continued)

ment of functional milestones.

Our review has shown the benefit of early
Parkman 2004

mobilization on length of stay but did not aim to

Whitney and

determine how much therapy gives the best result. In

this review both experimental and comparison
(USA)
Trial

groups typically had daily therapy. Renkawitz

Guerra et al. 853
showed a flattening of benefits at day 8 post opera-
tion in a pathway of further optimisation of an
existing accelerated clinical pathway protocol after
total knee replacement (which included ultra-early/
doubled physiotherapy and use of motor-driven
continuous passive motion machine units).6
Lenssen showed that two daily sessions received
by an experimental group and one by a comparison
group following total knee arthroplasty resulted
in no between-group difference in the primary
outcome measure of range of motion.29 Perhaps
early mobilization within the first 24 post operative
hours and one therapy session per day, as was typi-
cal in the trials included in this systematic review,
could be considered as a benchmark.
Concerns about the effect of accelerated
discharge on increasing burden on the primary care
sector may arise. Richards compared early dis-
charge with a hospital at home scheme and found
no increased burden on the primary care sector.30
In our review only two trials documented that their
patients were discharged directly home, with
neither study reporting an increased burden.17,19
These results may suggest that early discharge to
home may not increase burden reporting on the
demands for general practitioner services.
The strengths of our review are that having
included only randomized controlled trials (4 of 5 that
were high quality studies on the PEDro scale) we can
be confident of the result being less subject to bias.
This systematic review adhered to the Preferred
Reporting Items for Systematic reviews and Meta-
Analyses (PRISMA) guidelines.31 We used the
GRADE approach to grade the strength of evidence.
Our review had some limitations. Limiting
our inclusion criteria to English language may
introduce a selection bias. Some of the included
trials had small sample sizes. One study included 273
participants, or 44% of the total participants analysed
across this systematic review.21 These participants
had total knee replacements only. The number of
total hip replacements included for analysis was esti-
mated at 32% perhaps suggesting the results of our
review have a greater applicability to patients with
total knee replacement. The I value of 91% was high
indicating a high degree of statistical heterogeneity
in our meta-analysis. One contribution to the large
variation could be the variation in the types of sur-
gery included between each trial (no two trials com-
pared the same surgery types). The small number of
trials and variability in surgery types did not allow us
to perform meaningful sub-group analysis. However,
the issues for the clinician to determine early mobili-
zation are similar for hip or knee joint replacement
surgery. In our sensitivity analysis removal of the one
trial of lesser quality did not change the conclusion
that earlier mobilization reduced length of stay, but
statistical heterogeneity remained high.
In conclusion, getting patients walking and/or
sitting out of bed early (within 1 day of hip or knee
joint replacement surgery) reduces length of stay
and may improve clinical outcomes without
increasing the rate of adverse events.
Clinical messages
After hip or knee joint replacement,
mobilizing a patient within 24 hours is
associated with a reduced length of stay.
Early mobilization after hip or knee joint
replacement has been shown to lead to
improvement in outcome measures with-
out an increase in adverse events or other
poor outcomes.
Conflict of interest
All authors declare that they do not have any potential
conflict of interest.
Funding
This research received no specific grant from any funding
agency in the public, commercial, or not-for-profit sectors.
References
1. Ravi B, Croxford R, Reichmann WM, Losina E, Katz
JN and Hawker GA. The changing demographics of
total joint arthroplasty recipients in the United States
and Ontario from 2001 to 2007. Best Pract & Res Clin
Rheumatol 2012; 26: 637647.
2. Singh JA, Vessely MB, Harmsen WS, etal. A popula-
tion-based study of trends in the use of total hip and total
knee arthroplasty, 19692008. Mayo Clin Proc 2010; 85:
898904.
854 Clinical Rehabilitation 29(9)
3. Mota RE, Tarricone R, Ciani O, Bridges JF and
Drummond M. Determinants of demand for total hip and
knee arthroplasty: a systematic literature review. BMC
Health Serv Res 2012; 12: 225.
4. Mahomed NN, Davis AM, Hawker G, etal. Inpatient
compared with home-based rehabilitation following pri-
mary unilateral total hip or knee replacement: a rand-
omized controlled trial. J Bone Joint Surg Am 2008; 90:
16731680.
5. Pashikanti L and Von Ah D. Impact of early mobilization pro-
tocol on the medical-surgical inpatient population: an inte-
grated review of literature. Clin Nurse Spec 2012; 26: 8794.
6. Renkawitz T, Rieder T, Handel M, etal. Comparison of two
accelerated clinical pathwaysafter total knee replacement
how fast can we really go? Clin Rehabil.2010; 24: 230239.
7. Marshall JR and Hawrysio A. Inpatient recovery follow-
ing myocardial infarction and coronary artery bypass graft
surgery. J Cardiovasc Nurs 1988; 2: 112.
8. Higgins JPE and Green SE. Cochrane Handbook for
Systematic Reviews of Interventions In: Higgins JP and
Green S, (eds.). The Cochrane Collaboration, 2011.
9. de Morton NA. The PEDro scale is a valid measure of the
methodological quality of clinical trials: a demographic
study. Aust J Physiother 2009; 55: 129133.
10. Verhagen AP, de Vet HC, de Bie RA, etal. The Delphi list:
a criteria list for quality assessment of randomized clini-
cal trials for conducting systematic reviews developed by
Delphi consensus. J Clin Epidemiol 1998; 51: 12351241.
11. Maher CG, Sherrington C, Herbert RD, Moseley AM and
Elkins M. Reliability of the PEDro scale for rating qual-
ity of randomized controlled trials. Phys Ther 2003; 83:
713721.
12. Hahne AJ, Ford JJ and McMeeken JM. Conservative
management of lumbar disc herniation with associated
radiculopathy: a systematic review. Spine 2010; 35:
E488E504.
13. Maher CG. A systematic review of workplace interven-
tions to prevent low back pain. Aust J Physiother.2000;
46: 259269.
14. Higgins JP, Thompson SG, Deeks JJ and Altman DG.
Measuring inconsistency in meta-analyses. BMJ 2003;
327: 557560.
15. Hozo SP, Djulbegovic B and Hozo I. Estimating the mean
and variance from the median, range, and the size of a
sample. BMC Med Res Methodol 2005; 5: 13.
16. Atkins D, Best D, Briss PA, etal. Grading quality of
evidence and strength of recommendations. BMJ 2004;
328: 1490.
17. Dowsey MM, Kilgour ML, Santamaria NM and Choong
PF. Clinical pathways in hip and knee arthroplasty: a
prospective randomised controlled study. Med J Aust
1999; 170: 5962.
18. Whitney JD and Parkman S. The effect of early postopera-
tive physical activity on tissue oxygen and wound healing.
Biol Res Nurs 2004; 6: 7989.
19. Reilly KA, Beard DJ, Barker KL, Dodd CAF, Price AJ and
Murray DW. Efficacy of an accelerated recovery protocol
for Oxford unicompartmental knee arthroplasty - A ran-
domised controlled trial. The Knee 2005; 12: 351357.
20. Larsen K, Srensen OG, Hansen TB, Thomsen PB and
Sballe K. Accelerated perioperative care and rehabilita-
tion intervention for hip and knee replacement is effective:
a randomized clinical trial involving 87 patients with 3
months of follow-up. Acta Orthop 2008; 79: 149159.
21. Labraca NS, Castro-Snchez AM, Matarn-Pearrocha
GA, Arroyo-Morales M, Mar and Moreno-Lorenzo C.
Benefits of starting rehabilitation within 24 hours of pri-
mary total knee arthroplasty: randomized clinical trial.
Clin Rehabil 2011; 25: 557566.
22. Reilly KA and Barker KL. An exploration of patients
attitudes to a randomised controlled trial of accelerated
discharge following elective orthopaedic surgery (oxford
unicompartmental knee arthroplasty). J Bone Joint Surg
Br 2005; 87-B: 157.
23. Department of Health, Victoria. WIES19 cost weights
table. In: 201213 VCW, (ed.). Victoria, Australia:
Department of Health, Victoria, 2013.
24. Chinachoti T, Lungnateetape A and Raksakietisak M.
Periarticular infiltration of 0.25% bupivacaine on top of
femoral nerve block and intrathecal morphine improves
quality of pain control after total knee arthroplasty: a ran-
domized double-blind placebo controlled clinical trial.
JMed Assoc Thai 2012; 95: 15361542.
25. Dobrydnjov I, Anderberg C, Olsson C, Shapurova O,
Angel K and Bergman S. Intraarticular vs. extraar-
ticular ropivacaine infusion following high-dose local
infiltration analgesia after total knee arthroplasty: a
randomized double-blind study. Acta Orthop 2011; 82:
692698.
26. Kadic L, Niesten E, Heijnen I, etal. The effect of addition
of pregabalin and s-ketamine to local infiltration analgesia
on the knee function outcome after total knee arthroplasty.
Acta Anaesthesiol Belg 2012; 63: 111114.
27. Rikalainen-Salmi R, Forster JG, Makela K, etal. Local
infiltration analgesia with levobupivacaine compared with
intrathecal morphine in total hip arthroplasty patients.
Acta Anaesthesiol Scand 2012; 56: 695705.
28. Juliano K, Edwards D, Spinello D, etal. Initiating physi-
cal therapy on the day of surgery decreases length of stay
without compromising functional outcomes following
total hip arthroplasty. HSS J: The Musculoskel J Hosp
Spec Surg 2011; 7: 1620.
29. Lenssen AF, Crijns YH, Waltje EM, etal. Efficiency of
immediate postoperative inpatient physical therapy fol-
lowing total knee arthroplasty: an RCT. BMC Musculoskel
Disord 2006; 7: 71.
30. Richards SH, Coast J, Gunnell DJ, Peters TJ, Pounsford J
and Darlow MA. Randomised controlled trial comparing
effectiveness and acceptability of an early discharge, hos-
pital at home scheme with acute hospital care. BMJ 1998;
316: 1796801.
31. Moher D, Liberati A, Tetzlaff J and Altman DG. Preferred
reporting items for systematic reviews and meta-analyses:
the PRISMA statement. BMJ 2009; 339: b2535.
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.