In the early 1980’s, when physicians were being asked to examine children for signs of possible sexual abuse, there was little information regarding the details of normal genital anatomy and appearance of the perianal tissues in children. Perceived irregularities in the appearance of the hymen, such as a relatively narrow rim of hymen, or mounds and indentations on the hymen edge were thought to be due to abuse.^1,2

The first paper describing a study of anal findings in children selected for non-abuse was published by ^{McCann, et al in 1989}.³ In the same edition of the journal (Child Abuse & Neglect), a paper by Hobbs and Wynn in England4 described and showed photographs of anal findings in a group of children who gave a disclosure of anal abuse. The McCann paper described findings such as anal dilation which were commonly seen in the non-abused group but were considered by Hobbs and Wynne to be signs of anal penetration in the children disclosing abuse.

There was some disagreement about the significance of anal dilation, especially comparing the interpretation of the finding by physicians in Great Britain, who tended to give greater weight to that sign. In 1989, Hobbs and Wynne published the results of a retrospective chart review of records of 337 children referred to their center for examination due to suspected sexual abuse.⁴ They noted the description of “anal signs” included erythema, swelling, laxity, fissures, venous congestion, anal dilation, hematoma and bruising. They reported that 42% of the children had one or more anal finding, and 60% of the subset of children less than 6 years of age (115 subjects) had one or more of the signs.

The first extensive study and description of genital findings in female children selected for non-abuse was published by ^{McCann and associates in 1990}.⁵ Variations such as somewhat narrow rim of hymen, mounds on the hymen edge, exposure of intravaginal ridges and “enlarged” hymen opening were commonly found. The results of these studies of children selected for non-abuse made it possible to develop a list of examination findings that could be roughly categorized as “normal”, “non-specific”, evidence of injury, and presence of a sexually transmitted infection.

Using information from the studies of non-abused children, along with clinical experience, in 1992 one of the first “classification system⁶ was proposed as a way to summarize what was known at the time regarding the interpretation of examination findings with respect to sexual abuse. There were two sections in the classification table; classification of anogenital findings and “overall assessment of the likelihood of sexual abuse.” The overall assessment table included important descriptive information provided by the child (child’s description of abuse events), results of testing for sexually transmissible infections, as well as forensic evidence of sexual contact such as identification of semen or sperm in specimens collected from the child’s body.

The sections in the classification of findings table were 1) Normal, 2) Nonspecific, 3) Suspicious for abuse, 4) Suggestive of Abuse/Penetration, and 5) Clear evidence of penetrating genital or anal trauma. “Normal” findings were those that had been described in studies of non-abused children. Findings with causes other than sexual abuse, such as from constipation, genital irritation, vaginitis, or dermatologic conditions like lichen sclerosus were in the “non-specific” section. Findings like anal gaping with stool present, anal fissures, venous pooling in the perianal tissues and flattened anal folds were also nonspecific for abuse.

Findings were classified as “suspicious” for abuse when they were sometimes found in cases of sexual abuse but could potentially have other causes. For example, bruising, abrasions or lacerations of the labia can be caused by accidental falls, and genital warts (Condyloma acuminate) can be spread by non-sexual as well as sexual contact. Based on data from the McCann, et al study of genital findings⁵ in non-abused girls, width of the hymen opening larger than 2 standard deviations above the mean is listed as a suspicious finding. From the McCann et al study of anal findings in non-abused children,³ immediate anal dilation of 15 mm or more, with no stool visible in the rectal vault was also listed as a suspicious finding.

Clear evidence of penetrating trauma, to the genital or anal tissues included missing segments of the hymen, hymen transections, perianal lacerations deep to the external anal sphincter, and scar of the posterior fourchette with a loss of hymen tissue between 5 and 7 o’clock. Evidence of genital contact includes positive tests for Neisseria gonorrheae when samples were taken from the pharynx, rectum or vagina, confirmed diagnosis of syphilis or HIV not acquired at birth, and recovery of semen or sperm from the genital or anal area. All of these findings are still considered as being caused by sexual contact and/or blunt force trauma to the genital or anal tissues.

This proposed approach to classifying genital and anal findings in cases of suspected child sexual abuse was a suggestion, but it was unclear how widely the published classification system was being used by other medical professionals who performed sexual abuse examinations. In 1990, a 68 item, 8-page survey was mailed to members of the Section on Child Abuse and Neglect of the American Academy of Pediatrics (270 members), and to members of the North American Society for Pediatric and Adolescent Gynecology (475 members) to determine how participants classified a list of normal and abnormal genital and anal findings with respect to suspicion for abuse. The results of the survey were published in 1993.⁷

The level of agreement with the classification of listed findings as abnormal was 89% (hymen transection) to 99% (presence of sperm). There were significant differences in how some of the listed findings were interpreted, based on the experience level of the survey participant. One of the findings listed was “condyloma in a child less than 2 years of age” as abnormal. Survey participants who had examined fewer than 200 children for possible sexual abuse were significantly more likely to rate this finding as abnormal, compared to those who had examined more than 500 children (47% vs. 8%, p = 0.000). The participants with more experience examining children recognized that condyloma can be transmitted at birth, as well as from abusive sexual contact.

Among participants who examined more than 5 children per month for sexual abuse, the cases with the highest level of agreement with the expert raters included those showing normal hymen variations (79% to 89%). A photo showing anal dilation with stool visible in the rectal vault was rated as normal by the expert raters, but by fewer than 50% of the participants who examined fewer than 5 children per month for suspected abuse. A photo showing anal dilation with no stool visible was rated abnormal by 85% of the expert raters, but by 63% to 100% of the participants.

The problem with trying to interpret the meaning of anal dilation during examination, whether or not stool is visualized, is the fact that chronic constipation, as well as other underlying conditions can cause anal dilation. Neurologic conditions can cause a decrease in anal tone, as can sedation. McCann, et al⁸ found some degree of anal dilation in 74% of the pediatric autopsies he reviewed, and in 32%, the rectal ampulla was visible.

In 2001, and updated version of the table listing categories of medical and laboratory findings in suspected sexual abuse was published.⁹ In this paper, anal dilation was listed as “concerning for abuse or trauma”, and defined as: “marked, immediate dilation of the anus, with no stool visible in the rectal vault, when the child is examined in the prone knee-chest position, provided there is no history of encopresis, chronic constipation, neurological deficits, or sedation.”

Anal dilation when stool is present in the rectal ampulla has generally been interpreted as a non-specific finding for abuse.¹⁰ In 2007 a group of specialists in child abuse pediatrics published a set of guidelines for medical care of children who may have been sexually abused.¹¹In that paper, the finding of anal dilation less than 2 centimeters in a child with or without stool in the rectum was considered a finding “commonly caused by other medical conditions”. In the absence of any predisposing condition, anal dilation of greater than 2 centimeters was considered an “indeterminate” finding, meaning there was insufficient or conflicting data from research studies^3,12,13 to determine the possible significance of the finding.

A retrospective review of medical records of 1115 children referred for suspected sexual abuse to one sexual abuse evaluation clinic was conducted to examine any correlation between a disclosure of anal abuse and the presence of anal findings.¹⁴In this study, specialty pediatricians blindly reviewed photographs of children who had been examined at the center and described the findings that were visible on the photographs. The forensic report was then reviewed to determine whether or not the child had given a disclosure of anal abuse, had a diagnosis of a sexually transmitted infection at the time of the examination, or if the abuse had been witnessed by an adult.

Based on the above criteria, 198 children (17.8%) were classified as belonging to the anal penetration group. There was a significant positive association with several findings and the designation as probable anal penetration. Anal soiling (p = 0.046), anal fissure(s) (p = 0.000), anal laceration (p = 0.000), and total anal dilation (p = 0.000) were significantly more common the anal abuse group, compared to the children without probable anal penetration. Interestingly, total anal dilation was significantly associated with anal penetration in girls, but not in boys, in children examined in the prone knee-chest position, compared to not, and in children without anal symptoms.

^{Hobbs and Wright published a study in 2014}, describing and comparing anal findings in children referred for sexual abuse with findings in children referred for physical abuse or neglect and found significant differences in percentage of children with one or more anal signs.¹⁵ They described finding anal dilation in 22% of children evaluated for sexual abuse (abuse), compared to none of the children referred for physical abuse or neglect (controls). For venous congestion, less than 1% of the children in the control group had that finding, compared to 36% of the abuse group. Anal fissures were significantly more common in the abuse group, compared to the control group (14% v. 1%).

As shown in Table 1, in 1992⁶, anal dilation of > 15 mm, with no stool present was classified as “suspicious” for abuse. In 2001,¹⁰ it was called “concerning” for abuse, and was defined as “marked dilation of the anus with no stool present, and no predisposing conditions”. By 2007¹¹ (Table 2), the finding was listed separately depending on how much anal dilation occurred. It was listed as a “finding commonly caused by other medical conditions” when there was “Partial or complete anal dilation less than 2 cm with or without stool visible”.

However, “marked, immediate anal dilation to a diameter of 2 cm or more, in the absence of predisposing factors” was listed as an “indeterminate” finding for abuse.

Table 1 Comparison of classification of medical findings in the evaluation of child sexual abuse from 1992 to 2001. 

Table 2 Changes in the classification of medical findings in the evaluation of child sexual abuse from 2007 to 2013 

In the 2018 update of the interpretation of findings table¹⁶ the amount of dilation (less or more than 2 cm) was removed, and instead, was categorized according to whether or not the child had a predisposing condition, such as constipation, encopresis, sedation, anesthesia, or a neuromuscular condition. In the presence of one or more of these conditions, anal dilation was listed in the “commonly caused by medical conditions other than sexual abuse”, and in the “no expert consensus regarding the degree of significance with respect to abuse” section if there were no pre-disposing conditions.

In order to assess the level of expert consensus on the interpretation of findings in suspected child sexual abuse, in 2017 a survey was sent to 491 members of the Ray E. Helfer Society, a specialty society of physicians involved in the evaluation of child sexual abuse. Responses were received from 90 physicians who indicated they regularly provide examinations for children with suspected sexual abuse. The survey respondents indicated if they agreed with how items were listed in the interpretation of findings table, published in the 2016 updated guidelines paper.¹⁷

There was 89% to 100% agreement on how the findings were categorized except for 4 specific findings. In the list of “no expert consensus” findings there was less than 89% agreement on the findings of “complete anal dilation in the absence of predisposing factors”, 68% agreement for “deep hymen notch in the posterior hymen rim”, and 83% agreement for “confirmed HSV-1 or HSV-2 in genital or anal areas in a child with no other indicators of abuse. There was 81% agreement for listing Trichomonas vaginalis infection in the “infections transmitted by sexual contact” section, but 95% to 100% agreement on findings of acute injury, and for the presence of a scar in the perianal area or posterior fourchette.