Forensic Science International 238 (2014) 22–25

Contents lists available at ScienceDirect

Forensic Science International

journal homepage: www.elsevier.com/locate/forsciint

Reflex anal dilatation: An observational study on non-abused children

Francesca Sfriso a,*, Susanna Masiero a, Veronica Mardegan a, Silvia Bressan a, Anna Aprile b
a
Pediatric Emergency Department, University of Padova, via Giustiniani 3, Padova, 35128, Italy
b
Department of Environmental Medicine and Public Health – Legal Medicine, University of Padova, via Falloppio 50, Padova, 35121, Italy

A R T I C L E I N F O A B S T R A C T

Article history: Objectives: Reflex anal dilatation (RAD) is considered as a possible sign of anal abuse,however studies
Received 26 June 2013 evaluating its prevalence in non-abused children are limited. The aim of this study was to evaluate the
Received in revised form 14 January 2014 prevalence of RAD in a convenience sample of children with no suspicion of abuse admitted to a Pediatric
Accepted 11 February 2014
Emergency Department (PED).
Available online 20 February 2014
Methods: Prospective observational study including children admitted to the PED of Padova, Italy,
between January and June 2011. Patients with no suspicion of abuse and for whom ano-genital
examination was part of their medical evaluation were included. Children were excluded if in critical
clinical conditions or if any suspicion of abuse arose during medical evaluation. Presence/absence of RAD
and of factors favoring its appearance were recorded for each patient.
Results: Two-hundred and thirty children (median age of 12 months, interquartile range 5–35 months)
were finally included. A positive RAD was reported in 14 (6.1%, CI 95% 3.4–10). Only 3 patients (1.3%, CI
95% 0.3–3.7) showed a positive RAD in the absence of any predisposing factor.
Conclusions: RAD is an infrequent sign in non-abused children and it is particularly rare in the absence of
any predisposing factor. Case-control studies are necessary to better clarify its diagnostic relevance.
ß 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction In 2008 the Royal College of Paediatrics and Child Health

Sexual abuse in childhood is a widespread phenomenon: data
published by the World Health Organization in 2006 estimates that
150 million girls and 73 million boys under 18 have experienced
sexual intercourse or other forms of sexual violence [1]. According
to the study carried out by Hobbs and Wynne, child anal abuse
represents 29% of sexual abuse in girls and 83% in boys, and the
majority of this kind of maltreatment involves children aged
between 0 and 5 years old [2]. Unfortunately, signs with high
specificity are few and rarely found also in abused patients, so the
diagnosis is often challenging.
Reflex anal dilatation (RAD) is currently considered as a possible
sign of anal abuse. It is a dynamic phenomenon of relaxation of both
the external and the internal anal sphincters leading to such
dilatation that the examiner may see into the rectum [3]. Some
relevant studies [2,4,5] documented its presence in victims of
buggery, the first one being a landmark paper by Chris Hobbs and
Jane Wynne published in 1989, finding a prevalence of RAD of
approximately 40% in children with ascertained sexual abuse.
* Corresponding author. Tel.: +39 498 213505; fax: +39 498 218487.
E-mail address: fsfriso@yahoo.it (F. Sfriso).
affirmed that ‘‘Evidence indicates that reflex anal dilatation is
associated with a disclosure of anal abuse and has been reported in
sexually abused children [...]’’ while ‘‘It has not usually been
reported in children selected for non-abuse’’ [6]. In the same
document they affirmed that ‘‘If RAD is seen, sexual abuse should
always be considered in the context of the history, medical
assessment and other anogenital signs’’. These statements were
confirmed in the update published in 2011 [7].
In her paper, JA Adams classified RAD with an antero-posterior
diameter of more than 2 cm as an ‘‘Indeterminate sign’’ of anal
abuse. It means that, even if there are insufficient or conflicting
data from research studies and no expert consensus on it, the sign
may support a child’s disclosure and may induce to contact child
protective services in some cases. The author underlined that
‘‘medical providers who examine children’’ with suspected sexual
abuse ‘‘need to be aware of published research on findings in non-
abused children [. . .]’’ [8].
Finally, a retrospective blinded study published in 2013 by
some of the main experts on child abuse, Myhre and JA Adams,
demonstrated that ‘‘total anal dilatation was significantly associ-
ated with anal penetration (p = 0.000)’’, as showed by the bivariate
comparison of its prevalence between children with and without
probable anal penetration. The results of the study suggested that

http://dx.doi.org/10.1016/j.forsciint.2014.02.010
0379-0738/ß 2014 Elsevier Ireland Ltd. All rights reserved.
 F. Sfriso et al. / Forensic Science International 238 (2014) 22–25
‘‘total anal dilatation is potentially an important finding’’ even if ‘‘it
is too early to upgrade the finding into highly suggestive for anal
abuse’’. These statements underline the persistent interest for the
sign in the diagnosis of anal abuse in children [9].
Only few studies evaluating the prevalence of RAD in
nonabused children have been carried out and presented relevant
limitations such as the lack of a precise definition of a positive RAD,
a small sample size, a limited age range of included children or the
selection of patients presenting specific clinical conditions, like
constipation or other favoring factors [10–16].
The aim of the present study was to evaluate the prevalence of
RAD in a convenience sample of children with no suspicion of
abuse admitted to a Pediatric Emergency Department (PED).
2. Patients and methods
This is an observational prospective study including a conve-
nience sample of children admitted to the PED of Padova, Italy,
between January and June 2011. Children presenting the following
criteria were included in the study: (1) no suspicion of sexual
abuse; (2) children for whom ano-genital examination was part of
their routine medical evaluation. Exclusion criteria were: (1)
previous referral to the hospital for allegation or suspicion of abuse
and patients presenting ano-genital findings considered as
significantly suspicious for abuse according to JA Adams’ paper
of 2011 (we excluded patients presenting one or more signs
reported among ‘‘Findings diagnostic of trauma and/or sexual
contact’’ in JA Adams’ table, unless a clear, timely and plausible
description of accidental injury was provided) [8] and/or suspi-
cious data collected from the medical history; (2) positive RAD and
subsequent suspicion of abuse at follow-up evaluation (3) critical
clinical conditions at presentation.
Information collected for each patient included demographic
data (sex, age, nationality) as well as clinical data (the presence/
absence of RAD, the presence/absence of favoring factors for the
appearance of RAD according to the literature). Physicians working
in the PED were trained to evoke RAD by gently parting the
buttocks with both hands and simply observing the anal region for
30 s, with patient laying in the prone knee-chest position, in the
left-lateral position or supine lifting his legs up, indifferently.
Training of medical staff was carried out through: (1) three
dedicated teaching sessions; (2) teaching material (including
photographic material) available in the PED; (3) availability of 2
study investigators for supervision of all patients presenting a
positive RAD and for any other case when deemed necessary by the
physician in charge of the patient. Each of the physicians involved
in the data collection observed the study investigators for the first
10–15 visits in order to uniform the method of detection and
recognition of the sign. RAD was considered as positive when both
the external and the internal anal sphincters dilated showing the
rectal mucosa within 30 s. Constipation, which is the most
frequent favoring factor [14,15], was defined as the lack of
defecation for 2 and 3 days or the evacuation of hard faeces or as
the presence of palpable stools in the abdomen. To check the
presence of these findings parents were asked about their child’s
bowel habits and an accurate abdominal examination (both
superficial and deep palpation of the abdomen) was performed
by the physician in charge of the patient. The presence of other
predisposing conditions (encopresis, sedation, anesthesia [8],
spinal diseases [16], inflammatory bowel diseases [17], digital
rectal examination, use of suppositories or enemas [2]) was also
recorded. Enemas, suppositories and digital rectal examinations
were considered as favoring factors only when performed in the
24 h prior to medical evaluation. All data collected, both
demographic and clinical, were recorded on the PED patient’s
chart and the presence or the absence of RAD was recorded in the
23
ano-genital part of the physical examination. The researchers then
checked all the patients’ charts and looked for those reporting
information about RAD. All charts were reviewed in order to
determine which cases met the inclusion criteria or exclusion
criteria for the study. Then all the data were entered into an
electronic database.
Children with positive RAD were followed-up by contacting
their Primary Care Physician at 3 months from discharge to verify if
any suspicion of abuse has emerged during that period. Monitoring
of return visits to the PED was also performed in order to identify
possible subsequent elements of suspicion for abuse.
3. Statistical analysis
Categorical variables were expressed as percentages and 95%
confidence intervals (CI) were reported for main results. Quantita-
tive variables were expressed as median and interquartile range
(IQR), due to non-normal data distribution. Comparison of
categorical variables was performed by means of x2 test, while
Mann–Whitney Test was used for comparing quantitative vari-
ables. Parameters displaying p  0.05 were considered statistically
significant.
Statistical analyses were conducted using the statistical
program MedCalc 11.1.
All the procedures were performed in compliance with
institutional guidelines and the Internal Review Committee has
approved them. The privacy rights of human subjects was observed.
4. Results
The number of patients screened for RAD was 230. Each of them
corresponded to the inclusion criteria and did not present any
characteristic to be excluded from the sample. So the study finally
included 230 patients, 117 girls (51%) and 113 boys (49%), aged
between 8 days and 12.6 years (median age 12 months, IQR 5–35
months). Eighty-eight percent of the sample were children
younger than 5 years old. One hundred and seventy-four children
(76%) were Italian, while 56 (24%) had a different nationality (24
African, 22 from East Europe, 9 Asian and 1 from South America).
Chief complaints at presentation of included children are reported
in Table 1. None of the 230 children presented any exclusion
criteria, therefore they were all included in the sample.
RAD was found in 14 patients, which represented 6.1% of the
population (CI 95%, 3.4–10%). All of these 14 cases were reassessed
by one of the study investigators and each of them was confirmed to
present a positive RAD. Of these, 10 (71%) were girls and 4 (29%) were
boys, with no significant difference in distribution compared to
children with negative RAD (p = 0.19). The median age of children
with positive RAD was 36 months (IQR 15–84 months), which
resulted significantly higher compared to the group with negative
RAD (median age 11 months, IQR 5–32 months, p = 0.002).
Table 1
Chief complaints of included children.
Chief complaints N %
Constipation and/or encopresis 40 17.4
Abdominal pain 31 13.5
Bloody stools 13 5.6
Anal or perianal itching/pain (suspected parasitosis) 6 2.6
Dysuria 18 7.8
Ascertained accidental genital trauma 4 1.7
Genital pain/swelling (suspected dermatitis) 13 5.6
Othera 105 45.7
Total 230 100
a
Include children wearing diaper for whom ano-genital evaluation was carried
out as part of the routine complete medical evaluation.
24 F. Sfriso et al. / Forensic Science International 238 (2014) 22–25
Table 2
Predisposing factors in children with positive RAD.
Predisposing factors N %
Constipation 6 43
Suppositories 1 7.1
Constipation + enema 3 21.4
Constipation + encopresis 1 7.1
None 3 21.4
Total 14 100
Eleven patients out of 14 (78.6%) showed at least one
predisposing factor, as reported in Table 2.
Only 3 children, 1.3% (CI 95% 0.3–3.7%) of the total examined
population, showed RAD in the absence of any predisposing
condition.
Of the 216 children in whom RAD did not appear 68 (31%) had a
positive history for one or more predisposing factors: 44 (20%) for
constipation, 2 (0.9%) for use of enemas, 9 (4.2%) for use of
suppositories, 11 (5.1%) for constipation and enemas, 1 (0.5%) for
constipation and suppositories and 1 (0.5%) for digital rectal
examination.
Prevalence of RAD among the 79 patients with any predisposing
factor was seven-fold higher than in children with no predisposing
factor (14% (CI 95% 7.2–23.5%), vs 2% (CI 95% 0.4–5.3%), p = 0.0009).
Children with any predisposing condition were significantly older
than children with no predisposing factors (median age 20 months,
IQR 6.25–50.75 vs 10 months, IQR 5–22, p = 0.0039).
We separately analyzed children aged 0–4 and 5–12 years. The
prevalence of RAD resulted higher in the younger age group
compared with older children (14.3% vs 5%), but the difference was
not statistically significant (p = 0.07). Table 3 shows the distribu-
tion of predisposing factors for RAD in both age groups.’’ For the age
group of children younger than 5 years there was a significant
association between presence of RAD and presence of predisposing
factors (p = 0.01). No significant association was found for the older
age-group (p = 0.1).
Children with RAD did not show any element of suspicion for
abuse during the follow-up we performed: 6 of them had one or
more subsequent visits to the PED for conditions not suspicious for
abuse and none of them showed any evidence of abuse at physical
examination. Furthermore no elements of suspicion emerged for
any child with positive RAD from the telephone interview
performed with their Primary Care Physicians at 3 months after
the first discharge.
5. Discussion
Our study is the first to report the prevalence of RAD in a
convenience sample of children with no suspicion of abuse
admitted to the PED, using a precise definition of the sign,
according to the most recent guidelines for child abuse. Only few
other studies, published more than ten years ago, evaluated the
presence of RAD in children with no suspicion of abuse. However
those studies presented some limits which warrant consideration:
Table 3
Distribution of predisposing factors per age group.
0–4 years
Predisposing factors
Yes No
Positive RAD 7 (10.9%) 3 (2.2%)
Negative RAD 57 (89.1%) 135 (97.8%)
Total 64 (100%) 138 (100%)
the one led by Myhre, Berntzen and Bratlid [13] examined a
restricted age-range population of children aged between 5 and 7
years old; the one published by McCann [12] did not consider the
frequency of constipation or other favoring factors in its
population, while reporting incomplete data on the presence/
absence of stools in the rectal ampulla; Stanton and Sunderland
[10] did not use precise criteria to define what they considered as a
positive RAD; and the study by Berenson [11] considered a small
sample including only girls between 0 and 18 months of life. Three
other studies evaluated the frequency of the sign in non-abused
patients, but they limited their analysis to children selected
for specific conditions, such as constipation [14,15] or spinal
disease [16].
The present study found a prevalence of RAD of 6.1% in children
admitted to a PED for conditions requiring an accurate ano-genital
inspection and with no suspicion of abuse. This percentage is
inferior to the prevalence obtained by McCann (14.7% within 30 s)
and by Stanton and Sunderland (14%), slightly higher than the
frequency documented by Myhre (0.7% in children examined in the
left lateral position and 4.7% in those examined in the knee-chest
position) and higher than the one showed by Berenson (0%). These
differences may be explained by the different inclusion criteria and
definition used in the studies. According to our results RAD was
absent in 94% of children with no suspicion of abuse. Its frequency
was seven-fold higher in children presenting predisposing factors
compared to those with no predisposing factors. Our data confirm
previous results reporting association of RAD with constipation
[14,15] and supports data showed by the Myhre et al. study
published in 2013 [9] affirming that anal dilatation is ‘‘significantly
more common in children with probable anal penetration
compared to children without probable anal penetration’’ consid-
ering only those ‘‘without anal symptoms such as constipation,
diarrhea or encopresis (12% vs 3.7%)’’. RAD was elicited in only 3
patients with no favoring factors (1.3% of the overall population
and 2% of the population with no predisposing conditions).
Children with positive RAD were significantly older than those
with negative RAD. In our sample the median age of children with
one or more favoring factors was significantly higher than children
without any of them. This finding may be likely explained by the
inclusion criteria of our study. Most of the infants and young
children, who represent the great majority of our sample,
underwent ano-genital inspection as part of their routine complete
examination only because diaper-wearing, while older children
were included according to their chief complaints, of which
constipation was the most common one.
The prevalence of RAD in children younger than 5 years was
nearly triple compared with older patients (14.3% vs 5%). This
difference was not statistically significant and this may be due to
the small sample size of older children. However the different
prevalence of RAD per age-group seems to have clinical signifi-
cance and be related to the presence of anal symptoms/
predisposing factors. Our further analysis showed that there
was a significant association between the presence of RAD and
predisposing factors in the younger age group. A similar finding
was not observed for older children, but once again the low
5–12 years Total
Predisposing factors
Yes No
4 (26.7%) 0 14
11 (73.3%) 13 (100%) 216
15 (100%) 13 (100%) 230
 F. Sfriso et al. / Forensic Science International 238 (2014) 22–25
numbers in this age-group likely affect the results and do not allow
for definitive conclusions. A larger sample with a more homoge-
neous age distribution will provide better data for a meaningful
age-group analysis.
This study presents some limitations that need to be addressed.
First, even if specific criteria to exclude abuse were used, absolute
certainty can not be reached. However, the use of precise criteria to
verify the absence of maltreatment and the follow-up we
performed much reduced the possibility to include abused
children in the sample. It is important to underline that our PED
has a high experience in child maltreatment. A study led in 2008
showed that between January 2003 and June 2008 one hundred
cases of suspect abuse came to our PED, with a median number of
18.4 cases per year [18]. Furthermore, since 2004 a Medical Doctor
of our team who is trained in this field works as point of reference
for every case of suspicion of abuse admitted to the PED. She works
in close contact with the Crisis Unit for the Battered Child, a
multidisciplinary group which deals with the diagnosis of abuse
and which takes charge of the victims. Second, inter-rater
reliability for RAD evocation was not assessed, as photographic
documentation was not available at the request of the Internal
Review Board. Nevertheless the medical staff training as well as the
availability of teaching material in the PED and of 2 study
investigators in case of need was provided for this study. Patients
with positive RAD were revaluated by the study investigators to
ascertain the presence of the sign. Furthermore the use of a clear
and accepted definition, the easiness of evocation and recognition
of RAD minimized the risk of unreliable findings. The size of
anterior-posterior diameter of positive RAD was not collected,
because the definition of the sign does not include it. The Royal
College of Paediatrics and Child Health affirmed that ‘‘Precise
measurements of the diameter of the dilated anal sphincter is not
possible, practical or feasible and is therefore not recommended’’
[6,7]. It follows that we considered as positive also RAD with a
diameter smaller than 20 mm (which is the size deemed to be
more significant for abuse in the literature [8]), as other Authors
did [13]. Another limitation of our study was not having the
examination position standardized. According to two studies
conducted by Myhre [9,13], RAD was more frequent in children
examined in the KCP as in those in LLP. However, in the study
conducted in 2001 [13], 7 patients out of the 13 examined in KCP
and presenting RAD complained constipation and/or showed
stools in the rectum. Considering only children without anal
symptoms, the difference between LLP and KCP is not statistically
relevant. Actually, even if the lack of a standardized position for the
examination is a limit of our study, data collected do not lose their
significance.
A further limitation is due to age distribution: 88% of the sample
were aged between 0 and 5 years. This bias may be explained by
the study inclusion criteria, i.e. inclusion of diaper-wearing
children, for whom accurate ano-genital inspection was part of
their routine medical evaluation. According to good clinical
practice, children wearing diaper must be completely examined
(including the ano-genital area) independently of presenting
conditions, thus determining a greater prevalence of younger
children in our population. Nevertheless, as the majority of
buggery cases concern children aged between 0 and 5 years [2], the
25
results of our study could be more directly compared to data on the
prevalence of RAD in anally abused children. Finally, the inclusion
of children requiring ano-genital examination as part of their
routine medical evaluation, also determined a higher percentage of
constipation in our sample compared to the prevalence docu-
mented in world pediatric population by a recent review [19]. This
may result in an overestimation of the prevalence of RAD in non-
abused children.
6. Conclusions
RAD is infrequently found in children with no suspicion of
abuse and it is rarely present in patients who do not show factors
favoring its appearance. Well-designed case control studies
matching confirmed anal abused with non-abused children are
necessary to better evaluate the diagnostic reliability of RAD as a
specific sign for anal abuse in children, especially in the absence of
favoring factors.
References
[1] P.S. Pinheiro, Global estimates of health consequences due to violence against
children., Background Paper to the UN Secretary-General’s Study on Violence
against Children, World Health Organization, Geneva, 2006.
[2] J. Hobbs, M. Wynne, Sexual abuse of English boys and girls: the importance of anal
examination, Child Abuse Negl. 13 (1989) 195–210.
[3] J. Hobbs, M. Wynne, Predicting sexual abuse and neglect, in: K. Browne, H. Hanks,
P. Stratton (Eds.), Early Prediction and Prevention of Child Abuse. A Handbook,
John Wiley & Sons, Chichester, 2002.
[4] M. Bruni, Anal findings in sexual abuse of children (a descriptive study), J. Forensic
Sci. 48 (2003) 1343–1346.
[5] C.J. Hobbs, J.M. Wynne, Buggery in childhood–a common syndrome of child abuse,
Lancet 2 (1986) 792–796.
[6] Royal College of Paediatrics and Child Health, The physical signs of child sexual
abuse an evidence-based review and guidance for best practice London 2008.
[7] Royal College of Paediatrics and Child Health, The physical signs of child sexual
abuse: an evidence-based review and guidance for best practice. Interim state-
ment, 2011.
[8] J.A. Adams, Medical evaluation of suspected child sexual abuse: 2011 update, J.
Child Sex Abuse 20 (2011) 588–605.
[9] A.K. Myhre, J.A. Adams, M. Kaufhold, J.L. Davis, P. Suresh, C.L. Kuelbs, Anal findings
in children with and without probable anal penetration: a retrospective study of
1115 children referred for suspected sexual abuse, Child Abuse Negl. 37 (July (7))
(2013) 465–474. http://dx.doi.org/10.1016/j.chiabu.2013.03.011, Epub 2013
Apr 22.
[10] A. Stanton, R. Sunderland, Prevalence of reflex anal dilatation in 200 children, BMJ
298 (1989) 802–803.
[11] A.B. Berenson, A. Somma-Garcia, S. Barnett, Perianal findings in infants 18 months
of age or younger, Pediatrics 91 (1993) 838–840.
[12] J. McCann, J. Voris, M. Simon, R. Wells, Perianal findings in prepubertal children
selected for nonabuse: a descriptive study, Child Abuse Negl. 13 (1989) 179–193.
[13] A.K. Myhre, K. Brentzen, D. Bratlid, Perianal anatomy in non-abused preschool
children, Acta Paediatr. 90 (2001) 1321–1328.
[14] G.S. Clayden, Reflex anal dilatation associated with severe chronic constipation in
children, Arch. Dis. Child. 63 (1988) 832–836.
[15] U. Agnarsson, C. Warde, G. McCarthy, N. Evans, Perianal appearances associated
with constipation, Arch. Dis. Child. 65 (1990) 1231–1234.
[16] N.W. Read, W.M. Sun, Reflex anal dilatation: effect of parting the buttocks on anal
function in normal subjects and patients with anorectal and spinal disease, Gut 32
(1991) 670–673.
[17] World Health Organization, Genital trauma, Managing child abuse, a handbook
for Medical Officers, World Health Organization Regional Office for South-East
Asia, New Delhi, 2004.
[18] V. Mardegan , S. Masiero , L. Da Dalt, Il bambino maltrattato: cinque anni di
esperienza al Pronto Soccorso Pediatrico di Padova results. (unpublished).
[19] M. Van den Berg, M.A. Benninga, C. Di Lorenzo, Epidemiology of childhood
constipation: a systematic review, Am. J. Gastroenterol. 101 (10) (2006) 2401–
2408.
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.