The British Journal of Radiology, 82 (2009), 363370

The use of gonad shielding in paediatric hip and pelvis radiographs

S L FAWCETT,

BA, BM BCh, MRCS

and S J BARTER,

MBBS, MRCP, FRCR

Addenbrookes Hospital, University of Cambridge Teaching Hospitals NHS Foundation Trust, Hills Road, Cambridge CB2
2QQ, UK

ABSTRACT. The problem of inaccurate placement of gonad shields in children has been
highlighted by several publications nationally and internationally over the past decade
and more. Here, we review the literature and present the results of a regional audit
designed to assess the use and accuracy of placement of gonad shields for hip and
pelvis radiographs in children. 100 consecutive anteroposterior hip or pelvis
radiographs in patients under the age of 16 years were reviewed in each of 9 centres.
We also included the most recent and all previously available relevant radiographs. A
total of 2405 radiographs were reviewed with regard to the presence of a shield and to
the accuracy of any shield placement with respect to gonad protection and
visualization of orthopaedic landmarks. It is recommended that gonad shields are used
in all follow-up paediatric pelvis radiographs. Our results show they were only used in
70% of such cases. When placed, only 38% of all shields were considered to be
positioned accurately. For cases where shielding was indicated, an accurately placed
shield was present in just 26% of radiographs. Formal written departmental guidelines
for shield use were only available in two centres. We conclude that clear guidelines
need to be formulated which, together with shield redesign, improved training and
audit, should increase effective gonad protection for children.

Since the 1950s, the contribution of pelvis and hip

radiography to significant radiation exposure to the
gonads has been of concern [1]. In the mid 1950s, devices
for protection of the gonads and methods for their
placement were described in this journal for both male
and female patients. It was suggested that they be applied,
in every infant or child, whether cooperative or not [2, 3].
There is a lack of national guidelines in the UK and, in our
region, a lack of local guidelines for the use of gonadal
shields (GSs) for paediatric radiography of the hip and
pelvis. Recently, the practice of omitting shields for initial
radiographs at the presenting event to avoid potentially
obscuring essential landmarks has been widely accepted;
however, application of GSs for all subsequent exposures
is advised [4]. Published audits from the UK on the use of
GSs in children identify appropriate use and positioning
in only ,25% of cases [58]. Here, we report the results of
a region-wide multicentre audit of the use and accuracy of
placement of GSs in hip and pelvis radiographs in patients
under the age of 16 years.

Methodology
Radiologists at nine hospitals within the Eastern
Deanery were invited and agreed to participate in this
audit. The contributing hospitals included two
University Teaching Hospitals and seven District
Address correspondence to: S L Fawcett, Addenbrookes Hospital,
University of Cambridge Teaching Hospitals NHS Foundation
Trust, Hills Road, Cambridge CB2 2QQ, UK. E-mail: sjfawcett@
doctors.org.uk

The British Journal of Radiology, May 2009

Received 4 February 2008

Revised 11 August 2008
Accepted 3 September
2008
DOI: 10.1259/bjr/86609718
2009 The British Institute of
Radiology

General Hospitals but no dedicated childrens hospitals.

In each centre, a list of cases was generated from the
Radiology Information System or similar IT system.
These systems were used to search for 100 consecutive
hip and pelvis radiographs taken of patients under the
age of 16 years. The most recent and all previously
available anteroposterior (AP) hip and pelvis radiographs for each patient were included in the study. Each
radiograph was reviewed with regard to the presence
and accuracy of the placement of GSs. Images taken with
an image intensifier were excluded. A total of 2405
exposures in 900 patients were included.
For each radiograph, the presence or absence of a GS
was noted. If present, the accuracy of the GS placement
was assessed by a radiologist with regard to adequate
gonad protection and obscuration of orthopaedic landmarks (Figures 14).
Placement was considered inaccurate if any of the
landmarks illustrated were obscured. For boys, placement
was also regarded as incorrect if part, or all, of one or both
testicles was visible beyond the limits of the lead. For girls,
exposure of more than a 1 cm wide rim of pelvic contents
within the pelvic basin was also deemed inadequate.
The x2 test was applied to determine statistically
significant differences between groups, with p,0.05
being considered significant.

Results
In total, a GS was placed in 48% (1161/2405) of cases.
The frequency of GS placement in follow-up exposures
363

S L Fawcett and S J Barter

each of the nine centres are given in Figure 5. The gender

differences in frequency of application and accuracy
were analysed by the x2 test. For application frequency,
x2524.9 with p,0.0001. For placement accuracy, x2554.0
with p,0.0001. Both of these gender differences, therefore, reach statistical significance.
A re-audit in one of our nine centres (Centre 2)
demonstrated an increase in placement accuracy, from
27% (20/74) in Round 1 to 44% (31/71) in Round 2. This
largely reflected an improvement in the accuracy of GS
placement in boys, with there actually being a small
decrease in accuracy in girls from 15% (6/39) to 12.5%
(6/16). There was no overall change in placement
frequency between the two rounds, which again
reflected an improvement in boys but a decline for girls
from 41% (39/96) to 25% (16/65).

Review of the literature and discussion

Figure 1. Ideal positioning of a gonad shield in a female

patient.

was 70% (903/1294). When present, the GS was

accurately placed in 38% (438/1161) of cases.
The results for accuracy and frequency of GS placement for each centre are given in Tables 13. In cases
where gonad protection was indicated, an accurately
placed shield was only present in 26% (333/1294) of
cases.
Both the use and accuracy of placement of a GS was
lower in girls than in boys. GSs were placed in 43% (550/
1268) of girls, of which 26% (145/550) were placed
accurately. GSs were placed in 54% (611/1137) of boys,
of which 48% (293/611) were accurate. The results for

Figure 2. Ideal positioning of a gonad shield in a male

patient.

364

In 1956, Osborn and Appendix [1] estimated that, in

males, 70% of the contribution to significant radiation
from diagnostic radiology was received during examination of the hip, pelvis and femurs. This led to Ardran and
Kemp [2] proposing in 1957 that every effort should be
made to reduce the dose; they went on to suggest the
application, by the radiographer, of a 1 mm lead shield
to protect the gonads of infants and young boys. They
also described a method for older patients, and for use in
departments staffed by female radiographers, that would
avoid embarrassment to patients and staff. The device
described was made of Perspex, covered in 2 mm thick
lead and intended for placement over the top of a gown
or sheet. The use of the midline and of the symphysis
pubis and greater trochanter as palpable landmarks was
detailed. They advocated its use to protect the male
gonad in all cases where practicable.
In 1958, Abram et al [3] pointed out that the problem of
protecting the ovaries was a more complex one and the
very varied position of the ovaries within the pelvis,
especially in children, was recognized [912]. Abram et
al [3] described a device intended for use in female
patients in every case in which radiography of the
structure lying outside the pelvic girdle, as opposed to its
contents, was required in their childrens hospital
setting. They stated that its use was applicable to every
infant and child whether cooperative or not. They found
that just two sizes of shield were needed to cover the
range from a newborn to those aged 16 years. The
midline and anterior superior iliac spines were used as
landmarks but with experience, they claimed, correct
positioning could be judged by eye.
Interestingly, it can be seen from the original description that the shield design extends above the line of the
anterior superior iliac spines to just below the umbilicus,
also protecting the sacroiliac joints and lower lumbar
region [3].
The limited data available on the variable position of
the ovaries do indeed seem to suggest that a parasacral
position of the ovaries, especially in children, is not
unusual [912].
It is of note, therefore, that the most common method
of protection in our region the Kings Lynn shield
does not provide protection to the sacral region. Hence it
The British Journal of Radiology, May 2009

Gonad shielding in paediatric hip and pelvis radiographs

Figure 3. Important diagnostic

orthopaedic landmarks. (1)
Shentons line; (2) Perkins lines; (3)
acetabular angle. The left hip is
normal. The acetabular angle is less
than 30 . The right hip demonstrates
the appearance of congenital dislocation. Shentons line is disrupted
and the femoral head ossification
centre can be seen in the outer
quadrant created by Perkins lines.

might be argued that a malpositioned shield, placed over

the sacroiliac joints, is actually likely to be more effective
than if it was positioned, as intended, just above the
pubic symphysis. If this is the case, our effective rate of
shielding might be higher than we have calculated based
on current standards. Conversely, we may be failing to
protect the ovaries in those children who have, what is
believed to be, an ideally placed GS. Some examples of
shield positioning are given in Figures 611.
The data on ovarian position used by the International
Commission on Radiological Protection (ICRP) comprises a study of mainly adult females in which the
ovarian position was inferred primarily from hysterosalpingography and a small set of paediatric data
involving just 13 children from 1 day to 12 years of age
[9, 10, 13]. In the latter, a German study, the ovarian
position was ascertained surgically in children with
known pelvic pathology [9]. The results of two subsequent studies [11, 12] had not been published at the
time when the ICRP used the information from the
former publications. The aim of these later publications,
one using CT and the other ultrasound, was to localise
the ovaries for intentional therapeutic radiation exposure. Both studies investigated ovarian position in adults
only. The variability of ovarian position in children is
perhaps uncertain and we are currently undertaking a
study to address this question.
We considered this information on ovary position
when defining our criteria for accurate placement of a GS
in girls. We set strict criteria allowing no more than a
1 cm rim of the pelvic contents to be exposed. The
criteria set by other published audits varied. For
example, two of the publications regarded the ovaries
as being located adjacent to the ischial spines [7, 8].
Another regarded two-thirds coverage of the pelvic basin

in the transverse plane as adequate [5]. One audit did not

describe the correct position in more detail other than to
state that, according to their protocol, the shield should
be placed above the symphysis pubis [6]. Regardless of
these varied definitions, the overall results are comparable. Despite our strict criteria for girls, malpositioning
was rarely a subtle matter. In our experience, when
deemed incorrect, the shields were frequently very
clearly inaccurate, leaving little room for debate.
Our data demonstrate a statistically significant difference in the application rate and the accuracy of
placement between girls and boys. All nine centres in
our study used shields less often for girls and were more
likely to place them incorrectly. This difference was also
reported and found to be statistically significant by
Kenny and Hill [8] in their sample of 346 radiographs.
Although the same trend was observed by McCarty et al
[5], the difference did not prove to be statistically
significant in their smaller sample.
In the absence of cryptorchidism, there can be little
doubt about the testicle position in boys. As discussed,
the same is not true for the female gonad and it seems
logical that application accuracy would be lower in girls
when relying on surrogate surface landmarks. In addition, the relatively inflexible design of the GS in current
use can be difficult to balance in a stable position,
especially on a relatively protuberant lower abdomen,
even if there is confidence about the location of the
surface landmarks. A mobile infant or uncooperative
child increases this difficulty.
Kenny and Hill [8] stated that, although it was easy to
explain the observed gender difference in application
accuracy, it was more difficult to explain the difference
in application frequency. We believe that the apparent
relative reluctance to use a shield in girls reflects a

Figure 4. Trethowans line. In the

normal right hip, the line passing
along the superior aspect of the
femoral neck is seen to pass through
the femoral head. In the abnormal
left hip, a slipped upper femoral
epiphysis means that the same line
remains superior to the femoral
head. This is known as Trethowans
sign.

The British Journal of Radiology, May 2009

365

S L Fawcett and S J Barter

Table 1. Breakdown of frequency of all GS placements for all nine centres
Centre

Total number of exposures

Total number of exposures with GS in situ

% of exposures with GS

1
2
3
4
5
6
7
8
9
Total

360
209
195
222
379
120
539
249
132
2405

209
74
72
68
173
44
370
64
87
1161

58
35
37
31
46
37
69
26
66
48

GS, gonad shield.

general acknowledgment by radiographers that accuracy

is a particular problem for the female patient. Shields are
therefore more often omitted for fear of obscuring the
landmarks of diagnostic importance. It is interesting,
therefore, that our results suggested that this was not the
main problem encountered. Only 11% (80/723) of GSs
deemed to be positioned inaccurately were classified as
such on the basis of landmark obscuration.
Despite the fact that the testicle position in boys is
relatively clear, we report that under 50% of GSs in boys
successfully protect the gonads without obscuring landmarks. Perhaps the current unisex design of GSs
provides simplicity at the expense of protection.
Shields tailored to girls or boys individually might
improve protection accuracy.
Since Ardran and Abram suggested use of shielding in
every practicable case in the 1950s, it is now generally
accepted that the omission of GSs for the initial
presenting episode for hip or pelvis radiography in
children is appropriate [24]. This reflects the risk of
obscuring essential orthopaedic landmarks and the
consequent need for a repeat exposure and increased
radiation burden. The use of shields for all but the initial
event is advised. In seven of the nine centres, it was not
possible to locate a written departmental policy on the
use of GSs in children. In all seven of these centres, there
was general informal acceptance that GSs should be
omitted in first-time exposures and applied for all
follow-up investigations. Only two of the nine centres
in our study had a written policy that could be produced.
Of interest, one of these two hospitals was the only
centre to have a different approach to the application of
GSs in boys. This policy stated that a GS should be
applied to all hip and pelvis radiographs in boys

regardless of whether they were initial or follow-up

exposures. Once again, this reflected a belief that
accuracy of placement was more problematic in girls,
with the consequent risk of obscuring landmarks being
too high. This policy also perhaps implies that GS
placement for boys was not a significant problem. In that
particular hospital, 50% of GSs in boys were inaccurate.
It would appear, therefore, that the acceptance of the
concept of omission of GSs for first-time exposures is a
reflection of the difficulties posed for accurate application in the first place. This was certainly the justification
for advocating shield omission by Liakos et al [4] in 2001.
Omission is regarded as preferable to the unintentional
obscuration of the very structures that need to be
imaged, with the consequent need for a repeat exposure.
It could be argued that the design faults of the current
shields are therefore dictating policies for their use.
There would perhaps be no need to ever intentionally
omit a GS and consciously irradiate the gonads if the
shield design allowed a reliably accurate application.
Testicular shields with a wrap around design, constructed using 1 mm of lead, are commercially available.
Their ability to reduce the dose during pelvic CT has
been investigated in both adults and phantoms and
shown to achieve a 7793% dose reduction from indirect
radiation [14, 15]. Up to a 10-fold reduction in dose is
also reported with the use of a shield enclosing the
testicles during radiotherapy treatment [16]. Similarly,
trough-shaped designs have been reported for use
during radiographic examinations of the hip [17].
Alternative methods for shielding the ovary have not
been so well reported.
Without a change in shield design, re-audit has been
shown to improve the effective use of shielding, with the

Table 2. Breakdown of accuracy of GS placement, when present, for all nine centres
Centre

No. of radiographs with GS in situ

No. of correctly positioned GSs

% accurate placement

1
2
3
4
5
6
7
8
9
Total

209
74
72
68
173
44
370
64
87
1161

45
20
32
27
48
19
190
25
32
438

22
27
44
35
28
43
51
39
37
38

GS, gonad shield.

366

The British Journal of Radiology, May 2009

Gonad shielding in paediatric hip and pelvis radiographs

Table 3. Breakdown of frequency and accuracy of GS application in follow-up exposures for all nine centres
Centre

No. of FU
exposures

1
222
2
102
3
85
4
108
5
210
6
NA
7
374
8
150
9
43
Total for all centres 1294

No.of FU exposures
with GS

% GS placement in FU
exposures

No.of FU exposures
with correctly placed
GS

% FU exposures with
correctly positioned GS

162
46
49
58
173
NA
323
55
37
903

73
45
58
54
59
NA
86
42
84
70

38
13
21
26
36
NA
164
19
16
333

17
13
25
24
17
NA
44
15
37
26

GS, gonad shield; NA, data not available; FU, follow-up.

accuracy reaching 68% after Round 2 and as high as

94.3% after Round 3 [5]. Our personal experience did not
reproduce this degree of success. A Round 2 audit was
performed in one of our nine centres. The results of the
first audit were presented to the department during a
clinical governance meeting. The presentation included
clear illustrations of the common errors. The anatomy
and theory of shield application was reviewed. The
results of the re-audit the following year showed an
overall increase in placement frequency but this included
an increase in the first-time exposure group, for whom
application was not indicated. Overall, there was an
increase in application accuracy from 22% in Round 1 to
44% in Round 2. The overall observed improvement in
both the frequency and accuracy of GS placement
occurred despite a disappointing reduction in both of
these categories in girls. In departments with a significant turnover of staff, re-audit alone is perhaps
unlikely to produce a sustainable improvement in
performance.
The risk of gonad irradiation has recently been
reviewed by the ICRP, with a consequent reduction in
the tissue-weighting factor for the gonads from 0.2 to
0.08 [18, 19]. The Commission states that there continues
to be no direct evidence that exposure of patients to
radiation leads to excess heritable disease in their
offspring. It judges, however, that there is compelling
evidence for this effect in experimental animals, and
therefore prudently continues to include the risk in its
system of radiological protection.

Much of the data considered for the 2007 ICRP

recommendations came from the reports reviewed by
the United Nations Scientific Committee on the Effects of
Atomic Radiation (UNSCEAR) 2001 [20]. One of the key
publications considered here reported that ovarian
cancer deaths increased with increasing doses of radiation in a study of 86 572 atomic bomb survivors [21].
However, the UNSCEAR 2001 report argues that, on the
basis of atomic bomb survivor and mouse genetic data,
the risk of heritable disease has been previously overestimated. The most significant change from the ICRP
1990 recommendations is a six- to eight-fold reduction in
the nominal risk coefficient for heritable effects. The
revised estimate of genetic risk is the main reason behind
the reduction in tissue-weighting factor for the gonads.
The ICRPs present estimate of genetic risk up to the
second generation of ,0.2% per Gy is essentially the
same as that cited by UNSCEAR 2001 [19, 20].
The atomic bomb survivor data estimated the excess
lifetime risk per Sievert for solid cancers for those
exposed at age 30 years to be 0.10 and 0.14 for males
and females, respectively. Those exposed at age 50 years
were thought to have about one-third of these risks.
Projection of lifetime risks for those exposed at age
10 years was regarded as more uncertain [21]. Under a
reasonable set of assumptions, estimates for this group
range from about 1.0 to 1.8 times the estimates for those
exposed at age 30 years. The ICRP also acknowledges
that there are significant differences in risk between
males and females and also with respect to an indivi-

Figure 5. Accuracy of all gonad

shield placement by gender and
centre.

The British Journal of Radiology, May 2009

367

S L Fawcett and S J Barter

Figure 6. The shield is too small and too high in this female
patient.

Figure 8. The shield is upside down, too high and too big in
this female patient.

duals age at the time of exposure. However, the

Commission continues to recommend that nominal risk
coefficients be applied to whole populations and not to
individuals in the belief that this policy provides a
general system of protection that is simple and robust
[19].
The dose received by the gonad during a pelvic
radiograph has been investigated. The entrance surface
dose has been shown to range from 0.262.89 mGy per
exposure; the average effective dose has been estimated
to be between 0.3 mSv and 1.4 mSv, increasing with both
age and male gender [22]. Whether modern digital
radiographical techniques have had an impact to reduce
effective dose is beyond the scope of this discussion but
could be used as an additional argument for intentional

Figure 7. The shield is upside down and too low in this

female patient.
368

shield omission in light of the reduced ICRP tissueweighting factors.

As mentioned, Osborn and Appendix [1] in 1956
estimated that, in males, 70% of the contribution to
significant radiation from diagnostic radiology was
received during examinations of the hip, pelvis and
femur. With the increasing use of CT, this figure will
now be out of date. The dose to the gonads during
unshielded abdominal CT is significantly higher than
plain radiography, reported to be ,2.4 mSv [15]. This
might also be used as an argument for not being too
concerned with protection during plain radiographic
examinations but rather turning our attentions to dose
reduction during CT in order to have the greater impact
on dose to the population as a whole. However, owing to
the cumulative risk, the use of GS is particularly
important for children with orthopaedic problems who
might be subject to serial investigations throughout their
childhood. On an individual basis, the contribution of

Figure 9. The shield is upside down, too big, too high and
obscuring landmarks in this male patient.
The British Journal of Radiology, May 2009

Gonad shielding in paediatric hip and pelvis radiographs

Figure 10. The shield is placed in a configuration as if for a

female patient and obscures essential landmarks. This
patient is male.

plain radiographs to their cumulative dose remains

significant. The impact that accurately placed GSs can
have on dose reduction remains important for this
patient group. It must not be forgotten that, for these
patients in particular, it is also as important to avoid
unnecessary investigations in the first place.
Despite the consequent reduction in tissue-weighting
factors, the Commission emphasised that this provides
no justification for allowing controllable gonadal exposures to increase in magnitude. As described, tissueweighting factors are sex- and age-averaged, representing mean values for humans. They are thus assumed to
be valid for both sexes and all age groups. Young
children are considered to be a particularly radiosensitive sub-population and any unnecessary radiation
exposure to the gonads may be considered unacceptable
even in view of the reduction in mean tissue-weighting
factor. The ICRP would therefore almost certainly not

approve a move to abandon the use of GSs, at least in

girls, altogether.
Data on the use of GSs for other radiographic
investigations, such as spinal examinations for female
scoliosis patients, are also available [23].The Medical and
Dental Guidance notes on GSs are not limited to
examination of the pelvis and are also not exclusive to
children. Their use is also advised for individuals of
reproductive capacity and examinations likely to give a
significant gonadal dose, unless these shields interfere
with the examination [24]. GS use for non-pelvic radiography, for adult radiography and for routine CT is
beyond the scope of this article but many of the same
principles apply.
Unfortunately, it appears that the effective use of
shielding during examinations of the hip and pelvis in
our region is inadequate. Our results, and the published
results of others from the UK and the USA, suggest that
this may be a nationwide and even an international
problem. It might be argued that there is selection bias
for publication of poor results, as they add fuel to a
controversial debate. Our sample of nine different
hospitals failed to identify a centre that was performing
to a standard that might be regarded as acceptable, at
least in terms of accuracy. A truly national audit would
be needed to confirm our suspicion that this is indeed a
problem across the UK.

Conclusions
We have shown GSs to be used ineffectively in nine
centres across our region. We believe that this echoes the
experience of other published studies on this topic and
suggests a nationally poor performance. Local guidelines
are often not clear and, if available, are rarely followed.
When used, GSs are incorrectly placed in an alarmingly
high number of cases. Performance is substandard and
practice has failed to improve despite repeated publications highlighting this issue.
Current methods for shielding in our region are
proving inadequate. Improved radiographer training in
use of the current shielding methods will be essential if
guidelines are to be followed with effect. Alternative
methods of shielding, such as shadow shields and
shaped contact shields, have advantages in that they
can be easier to use but can raise the problem of patient
embarrassment. Is it time that shields were redesigned?
In view of the recent reduction in gonadal tissueweighting factors, it might be argued that it is justifiable
to omit gonad shielding at least in female patients.
In light of the ICRP recommendations, it would seem
that a multidisciplinary review of GS design and the
indications for their application, perhaps at a national
level for the UK and especially for the paediatric
population, is overdue. If GSs continue to be recommended, it is time for clear guidelines to be set.

Acknowledgments
Figure 11. This is a good example of the intended position
for the shield in boys.

The British Journal of Radiology, May 2009

Kathryn Lawrence, Francis Hampson, Edmund

Godfrey, Andrew West, Laura Watson, Avnesh Thakor
and Herbert Alberto Vargas for their valued contribu369

S L Fawcett and S J Barter

tions to the data collection. Karen Goldstone (Radiation

Protection Advisor, Addenbrookes Hospital) for her
help with the historical aspects. Mick Cafferkey (Senior
Illustrator, Medical Photography and Illustration,
Addenbrookes Hospital) for his illustrations. Chris
Palmer (Medical Statistician, Director of Centre for
Applied Medical Statistics, University of Cambridge)
for his assistance with the statistical analyses.

References
1. Osborn SB, Appendix K. The hazards to man of nuclear and
allied radiation. London, UK: Medical Research Council, H
M Stationery Office, 1956.
2. Ardran GM, Kemp FH. Protection of the male gonad in
diagnostic procedures. Br J Radiol 1957;30:280.
3. Abram E, Wilkinson DM, Hodson CJ. Gonadal protection
from X radiation for the female. Br J Radiol 1958;31:3356.
4. Liakos P, Schoenecker PL, Lyons D, Gordon JE. Evaluation of
efficacy of pelvis shielding in preadolescent girls. J Pediatr
Orthop 2001;21:4335.
5. McCarty M, Waugh R, McCallum H, Montgomery RJ,
Aszkenasy OM. Paediatric pelvic imaging: Improvement in
gonad shield placement by multidisciplinary audit. Pediatr
Radiol 2001;31:6469.
6. Sikand M, Stinchcombe S, Livesley PJ. Study on the use of
gonadal protection shields during paediatric pelvic X-rays.
Ann R Coll Surg Engl 2003;85:4225.
7. Wainwright AM. Shielding reproductive organs of orthopaedic patients during pelvic radiography. Ann R Coll Surg
Engl 2000;82:31821.
8. Kenny N, Hill J. Gonad protection in young orthopaedic
patients. BMJ 1992;304:14113.
9. Fochem VK, Pape R. Problematick des ovarialschtzes bei
rontgenaufnahmen des beckens. Fortschr Geb Roentgenstr
Nuklearmed 1962;97:78593.
10. DAngio GJ, Tefft M. Radiation therapy with gynecologic
cancers. Ann N Y Acad Sci 1967;142:67594.
11. Counsell R, Bain G, Williams MV, Dixon AK. Artificial
radiation menopause: where are the ovaries? Clin Oncol
1996;8:2503.

370

12. Featherstone C, Harnett AN, Brunt AM. Ultrasound

localization of the ovaries for radiation induced ovarian
ablation. Clin Oncol 1999;11:3937.
13. Protection of the patient in diagnostic radiology. Ann ICRP
1982;9:29.
14. Price R, Halson S, Sampson M. Dose reduction during CT
scanning in an anthropomorphic phantom. Br J Radiol
1999;72:48994.
15. Hohl C, Mahnken AH, Klotz E, Das M, Stargardt A,
Muhlenbruch G, et al. Radiation dose reduction to the male
gonads during MDCT: the effectiveness of a lead shield.
AJR Am J Roentgenol 2005;184:12830.
16. Fraass BA, Kinsella TJ, Harrington FS, Glatstein E.
Peripheral dose to the testes: the design and clinical use
of a practical and effective gonadal shield. Int J Radiat
Oncol Biol Phys 1985;11:60915.
17. Krepler P, Vana N, Havranek C. Dosimetric studies in the
radiological examination of the hips in young infants with a
special fenestration method of gonad protection. Pediatr
Radiol 1977;5:2315.
18. Recommendations of the International Commission on
Radiological Protection. Ann ICRP 1990;21:86.
19. Recommendations of the International Commission on
Radiological Protection. Ann ICRP 2007;37:63.
20. UNSCEAR 2001. United Nations Scientific Committee on
the Effects of Atomic Radiation. Heritable Effects of
Radiation, 2001 Report to the General Assembly with
Scientific Annex. New York, NY: United Nations, 2001.
21. Pierce DA, Shimizu Y, Preston DL, Vaeth M, Mabuchi K.
Studies of the mortality of atomic bomb survivors. Report
12, part 1. Cancer: 19501980. Radiation Research
1996;146:127.
22. Almen A, Mattsson S. The radiation dose to children from
X-ray examinations of the pelvis and the urinary tract. Br J
Radiol 1995;68:60413.
23. Palmer SH, Starritt HC, Paterson M. Radiation protection of
the ovaries in young scoliosis patients. Eur Spine J
1998;7:27881.
24. Medical and Dental Guidance Notes a good practice
guide on all aspects of ionizing radiation protection in the
clinical environment. York, UK: Institute of Physics and
Engineering in Medicine (IPEM) 2002; 3.47.

The British Journal of Radiology, May 2009