Child Abuse

0031-3955/90 $0.00

+ .20

Use of the Colposcope in Childhood Sexual Abuse Examinations

John McCann, MD*

Since the recognition that children can be victims of sexual molestation, the medical community has been searching for methods of detecting physical changes that would help substantiate such an allegation. This search has fostered a multitude of approaches that include the use of various physical examination techniques, the application of dyes to help determine if minor cellular changes are present, the introduction of cameras to record results, and the use of magnification to uncover evidence of microtrauma. 9 , 18, 19, 28 Of the more recent additions, the employment of the colposcope, which incorporates both magnification and photography, has led to some of the most significant advances in the evaluation of the victim of childhood sexual abuse. 35 As with all new techniques, the use of the colposcope has both advantages and disadvantages. On the negative side, these instruments are expensive. The cost can vary from $5000, for a basic model, to more than $15,000 for the most sophisticated instrument equipped with various attachments. This instrument, originally developed for adult gynecologic use, can be difficult to master. This is especially true when dealing with the young child who may not be as cooperative as the older female patient for whom it was designed. On the positive side it provides an excellent light source, a magnified view of the child's anatomy, and, through the use of an attached camera, an opportunity to record the results of an examination. Not all examiners believe the colposcope is necessary in the evaluation of the sexually abused child. In a recent study in which a colposcopic examination was compared with an unaided examination, only 4 out of the 72 children with abnormal findings were overlooked when magnification was not employed. 24 In another report, the colposcope did not increase the *Associate Professor of Clinical Pediatrics, University of California, San Francisco; Department of Pediatrics, Valley Medical Center, Fresno, California

Pediatric Clinics of North America-Vol. 37, No.4, August 1990

863

864

JOHN MCCANN

number of abnormalities detected; however, a review of the photographs did decrease the frequency of suspicious findings. 3 Estimates by examiners as to the amount of additional help the colposcope provided in the assessment of the child for suspected sexual abuse ranged from 10% to 73%.33.35 Despite the differences of opinion, the use of the colposcope in the evaluation of the child suspected of having been sexually molested continues to grow.

COLPOSCOPE The search for better ways of illuminating the genital tract during an examination has led to several innovative approaches. This has included the use of the old-fashioned goose neck lamps, the newer fiber optic examination lights, and the ever-present otoscope. As light sources improved, health care providers began exploring methods of magnifYing these tissues in an effort to determine if minor changes occurred that could help with a diagnosis. Magnification was attained through an assortment of instruments that included magnifYing lenses secured to walls, placed on stands, attached to head bands, or held by hand. Although several of these devices incorporated more than one feature, it was not until the introduction of the colposcope with its camera that it became possible to have all three attributes in one instrument. The colposcope is essentially a binocular system of different strength lenses on a support stand that contains a light source (Fig. 1). Either a

Figure 1. Colposcope with videocamera attached. Note one-way mirror in the background.

USE OF COLPOSCOPE IN CHILDHOOD SEXUAL ABUSE EXAMINATIONS

865

Polaroid or a 35mm camera can be attached. Most colposcopes have a green lens filter to assist in the detection of scars or vascular abnormalities. An ever increasing number of attachments are becoming available. Some models (Leisegang Medical, Inc., Boca Raton, FL) are fitted with a stereoscopic lens that, through the use of a special viewing device, allows the examiner to review the photographs in three dimensions. The training of new examiners has been enhanced by the use of a split-image lens that can be added to some instruments. This feature has permitted the attachment of a teaching head or video camera to the colposcope. When the video system used in an examination room with a one-way mirror, trainees, after receiving permission from both the parent and the child, are able to watch an examination from an adjacent room with minimal disturbance to the patient (Fig. 1). CAMERAS AND PHOTOGRAPHS

The two types of cameras available for use with the colposcope are the Polaroid (Kodak, Rochester, NY) and the 35mm camera. The advantage of the Polaroid camera is that the photograph is immediately available. Unfortunately, the resolution of the Polaroid picture is relatively poor and is inadequate for many of the purposes for which these photographs are used. The image produced by the 35mm camera is excellent, and the film can be developed into either a photograph or slide. Several 35mm cameras can be used with the colposcopes. Although they vary widely in price and sophistication, it is the attachments added to the cameras that make the greatest difference' during the examinations. Recording data backs, automatic winders, and remote shutter releases are all devices that can significantly improve the ease with which the child with a short attention span can be photographed. The data backs can be programmed so an identification number and other information can be imprinted on each photograph. This is invaluable in the later identification of the child. The automatic winder allows the examiner to take several pictures in a row without having to stop and manually advance the film. The remote shutter release can be operated by hand or through the use of a foot pedal. This feature allows the examiner to use both hands during the examination, and by adding an extension cord to the shutter release either an assistant or the child can take the picture. Allowing children to take their own photograph provides them with a sense of control while helping to divert their attention away from the examination. The addition of a ring flash may significantly improve the quality of the photographs. The quality of the photograph is dependent on various factors. Obtaining a photograph at just the right instant and having it in focus can be a challenge, particularly when dealing with the squirming child. Proper positioning of the patient and the colposcope is critical if the image on the photograph is to approximate what was seen during the examination. Having the colposcope at right angles to the perineum is important if all parts of the picture are to be in focus. When lying supine, the plane of the young child's perineum may be tilted as much as 45 degrees from the perpendic-

866

JOHN MCCANN

ular angle. This means that the colposcope may have to be aimed downward 45 degrees to achieve the correct angle. The narrow depth of field at the higher magnifications may cause part of the picture to be out of focus if the proper alignment is not attained. Another factor in attaining quality photographs is in the development of the film. Accurate processing of the film is essential if the precise hue and color tones are to be duplicated. The reproduction of the colors of the tissues can be enhanced by periodically photographing a color bar on the same roll of film with one of the subjects. The inclusion of a spectrum of known colors can help the laboratory achieve the proper tones during the processing. The photographs produced by the colposcope have many uses. They not only provide a permanent document for the medical record, but they can also be used for peer review, second opinions, court testimony, standardization of techniques, clinical research, educational purposes, and later perusal as new observations are discovered. The opportunity to discuss findings and obtain a second opinion without having to subject the child to another examination can be of benefit to both the examiner and the child. The acquisition of another assessment from a colleague is an effective method for relieving some of the burden in these difficult cases. If a dispute arises, the attainment of a second appraisal through the use of these photographs is relatively easy. The importance of this latter capability is underscored by the fact that many courts are beginning to insist on having these children examined with the use of this instrument for this and other reasons. The rendering of a second opinion from a photograph rests on the premise that the image reproduced is a reasonable facsimile of what was found during the examination. This supposition is complicated by the fact that it is now clear that different examination techniques can produce different results. 18 At the present time, no unanimity exists as to which method should be employed during these evaluations. The ability to standardize each of the procedures through the use of the colposcopic photographs is one solution to this problem. By identifYing the examination technique and recording the findings with a camera, it would become possible for examiners to compare results and provide an interpretation of anatomical changes documented by others. Another need for standardization is in the determination of sizes. Through the use of these photographs, it also becomes possible to accurately record the measurements of soft-tissue findings. 18 The one measurement that has received the most attention is that of the hymenal orifice size. Through the use of a built-in measuring device available on some models, or by the construction of one's own scale, examiners can now determine sizes to within 11100 of a millimeter. The fabrication of a measuring device can be accomplished by photographing a metric scale at a predetermined magnification. Because of the narrow depth of field at the higher magnifications, the image of the scale will be the same relative size as the image produced in other photographs taken at the same setting. To maintain a constant focal length, which is critical to this method, examiners may need to determine their own eye-piece setting for the colposcope. It is also

USE OF COLPOSCOPE IN CHILDHOOD SEXUAL ABUSE EXAMINATIONS

867

possible to devise a similar measuring device for use with slides. When using slides, the image must be projected on a screen that is always the same distance from the projector. This can be accomplished through the use of a self-contained projector such as Kodak's Ringmaster. The photographs generated by the colposcope have proved to be of great value to researchers and educators in the field. The ability to collect and analyze objective information gathered from these reproductions has provided a wealth of material to the clinical researcher. The educator's use of slides in presentations and the publication of photographs in medical journals has expedited the dissemination of information to both trainees and other examiners. PHYSICAL EXAMINATION

The appropriate preparation of the child is a key element in obtaining an adequate examination and essential in minimizing any emotional stress experienced by the child. A careful explanation of the procedures to be used can significantly reduce a child's anxiety. The use of stuffed animals or ordinary dolls to demonstrate the different positions provides the child with a background of information that can help facilitate the examination. Allowing the child to touch swabs similar to the ones used to collect specimens and permitting the youngster to look through the colposcope further increases their familiarity with the proceedings. During the examination a maximum state of relaxation can best be achieved by having the child do the talking. This can usually be accomplished by having the younger child count, the school-aged youngster perform simple mathematics problems, and the older preadolescent describe an experience in exquisite detail. Only one person at a time should be allowed to talk to the child during the examination to avoid confusion. Occasionally, all efforts fail, and sedation is required. Although various sedatives have been used in this situation, chloral hydrate is among the safest. Adequate relaxation can usually be achieved with a dose of 75 to 100 mglkg. The child must be watched closely as they awaken because they will be ataxic, and they may fall and hurt themselves. The decision as to whether the parent remains in the room during the evaluation can best be decided with the assistance of both the parent and the child. Various approaches can be used to examine the child. For the apprehensive younger child, the mother's lap may offer the safest refuge. In the older child who will lie down on the examination table, either a frog-leg position or the use of the foot stirrups will provide the examiner with an adequate view of the perineum. The prone knee-chest position is best performed on the examination table because a sway-backed posture is essential in obtaining adequate relaxation of the pubococcygeus muscles. Because older patients may experience some back discomfort during the knee-chest approach, having them kneel on the foot extension of the examination table will usually help to minimize this problem. The most commonly used examination techniques for the female patient are the supine labial separation method and the supine labial traction

868

JOHN MCCANN

technique. In the former approach, the examiner separates the labia by applying pressure downward and laterally on the lower portion of the labia majora. This method is excellent for viewing the perineum, the labia, and the vestibule. Of all the methods, it is the least effective technique for opening the vaginal introitus. 18 To use the labial traction method, the examiner first grasps the labia majora between the thumb and index fingers of each hand. The labia are then gently, but firmly, pulled outward and upward until the edges of hymen are exposed and the vaginal introitus opens. For the prone knee-chest position, the child is placed on her knees with her chest resting on the examination table in a sway-backed position. The head is turned to one side with the forearms resting on either side of the head. The knees are separated 6 to 8 inches (15 to 20 cm) and maintained in 90 degrees of flexion. The examiner's thumbs are then placed beneath the leading edge of the gluteus maximus at the level of the vaginal introitus and the posterior portion of the perineum is lifted, revealing the hymenal orifice. This technique has proved to be the most effective approach for separating the edges of the hymen and opening the vaginal canal. 18 In the knee-chest position, the cervix can be visualized in most girls without the use of a speculum. The perianal portion of the examination can be performed in either the lateral decubitis, the supine, or the prone knee-chest position. The traditional approach of examining children while they are lying on their sides is believed by many to be less embarrassing and therefore less traumatic. The use of the supine technique, in which the hips are flexed on the abdomen, is gaining popularity, particularly in the younger child. The supporters of the knee-chest method argue that this position provides the best view of the perineum, and with proper preparation it is well tolerated by the child. Although each of these techniques has their champions, there currently is no information comparing the efficacy of the three methods. Although magnification permitted clinicians to visualize physical findings that were not seen during unaided examinations, and the colposcopic photographs gave examiners the opportunity of studying the child's anogenital anatomy in detail, it was only recently that comparative studies of nonabused children became available for use in distinguishing the abused from the non-abused child. 9, 16, 17 These newer reports have shown a greater variation in anogenital findings than was previously appreciated. It is apparent from these studies that the soft tissues are influenced by the age of the child, the examination technique, and even the youngster's state of relaxation. Without such instruments as the camera-mounted colposcope to chronicle these findings, examiners would have had to continue to rely on their own experiences to determine normality.

HORMONAL INFLUENCES The influence of maternal hormones on the female infant's genitalia has been recognized for years. This effect can last for two or more years even though these hormones disappear from the tissues after the first few

USE OF COLPOSCOPE IN CHILDHOOD SEXUAL ABUSE EXAMINATIONS

869

weeks of life. 6 The thick redundant nature of the hymenal tissues under the influence of maternal hormones can interfere with the examiner's ability to inspect the edges of this membrane and view the interior of the vaginal canal (Fig. 2A). This same phenomenon recurs as the preadolescent patient's own hormones begin to exert their influence. These changes frequently begin to occur several months before breast buds develop or pubic hair appears. It is in these children that the use of the labial traction method and the knee-chest position have the greatest impact on the results (Figs. 2B, 2C, 3B, and 3C). Both of these methods are excellent techniques for separating the edges of redundant hymens and opening the vaginal introitus. IS When these approaches are combined with the use of the colposcope, the examiner is usually able to detect even the most minute tissue changes. ANOGENITAL FINDINGS IN SEXUALLY ABUSED PREPUBERTAL CHILDREN

One of the principal uses of the colposcope is the documentation of injuries. This instrument has been employed extensively in the recording

Figure 2. A, 2.Case 1. Two year old. Figure A, Case 1. Two year Labial old. Labial by thick separation technique, orificeorifice covered technique, covered by thick separation redundant hymen. From study of nonabused redundant hymen. From study of nonabused children (see (see ref. 9). photograph children ref. Colposcopic 9). Colposcopic photograph X 10.) Case 1. Labial traction (magnification (magnification X B, 10.) B, Case 1. Labial traction method. Hymenal orificeorifice open,open, edge edge smooth. Hymenal smooth. method. Anterior hymenal Hap (arrow a); periurethral Anterior hymenal flap (arrow a); periurethral bandsbands (arrow b). Colposcopic photograph (mag-(mag(arrow b). Colposcopic photograph nification X 10).x C, 1. Knee-chest ap- apC, Case 1. Knee-chest nification 10). Case proach. Anterior hymenal Hap (arrow a) open a) open proach. Anterior hymenal flap (arrow revealing thin, thin, smooth-edged hymen. SeptalSeptal revealing smooth-edged hymen. remnant (arrow b). Intravaginal ridge ridge (arrow c) remnant (arrow b). Intravaginal (arrow c) at 4 o'clock. Colposcopic photograph (magnificaat 4 o'clock. Colposcopic photograph (magnifica~ion X 10).x 10). ~ion

870

JOHN MCCANN

Figure Figure 3.3. A, A, Case Case 2.2. Six Six year year old. old. Labial Labial separation method, redundant redundant hymen hymen with with small small separation method, orifice, orifice, poor poor relaxation. relaxation. (From (From study study of of nonanonabused bused children, children, ref. ref. 9.) 9.) Colposcopic Colposcopic photograph photograph (magnification (magnification xX 10). 10). B, B, Case Case 2.2. Labial Labial traction traction technique, technique, thin, thin, smooth smooth hymenal hymenal edge edge with with wider wider horizontal horizontal diameter. diameter. Anterior Anterior tag tag (arrow (arrow a). a). Intravaginal Intravaginal ridge ridge (arrow (arrow b) b) 44 o'clock. o'clock. ColColposcopic poscopic photograph photograph (magnification (magnification xX 10). 10). C, C, Case Knee- chestposition. position. Anterior Anteriortag tag(arrow (arrow Case2.2. Knee-chest a). a).Hymen Hymenisisthin, thin, smooth smoothedged edgedwith withmound moundatat 10 10o'clock o'clock attached attached toto an an intravaginal intravaginal ridge ridge (ar( arrow row b). b). Colposcopic Colposcopic photograph photograph (magnification (magnification xX 10). 10).

of findings in prepubertal children suspected of having been sexually abused. 22. 25 During this age period, as the effects of the maternal hormones wane and before the child's own estrogens come into existence, the hymen becomes a thin delicate membrane easily subject to damage. Soft-tissue changes thought to be due to trauma are the basis for most reports on sexually molested children in the medical literature. Terms such as attenuation, clefts, notches, bumps, rounded edges, neovascularization, synechiae, and others have evolved since magnified images of this membrane

USE OF COLPOSCOPE IN CHILDHOOD SEXUAL ABUSE EXAMINATIONS

871

have been studied so intensely.9, 22 The major problem that continues to plague medical examiners is the difficulty distinguishing between changes caused by sexual abuse from those that are naturally occurring or from ones that are the result of an accidental injury. Fortunately, in the prepubertal child, major anogenital injuries that result from sexual assault are relatively uncommon. Although the colposcope is helpful in documenting these findings, it is of even greater assistance in detecting minor injuries, such as small lacerations or abrasions. However, many of the other physical changes found in the sexually molested child are nonspecific. Injuries to the young girl's genitalia resulting from a recent episode of fondling may include only erthyema, localized edema, or petechiae. 17. 19 The genital changes brought about by repeated penetration or ongoing sexual misuse are thought to be more specific and are purported by some authors to include erythema and increased vascularity of the perihymenal tissues; increased friability of the posterior fourchette; minor labial adhesions at the posterior fourchette; attenuation of the hymen; and scars and synechiae from the hymen to the vagina. 9, 21, 23 In addition, "bumps" on the hymen between the 3 and 9 o'clock positions and an enlarged hymenal orifice for age was found to be of significance in one study.9 Other authors have concluded that asymmetry of the hymenal orifice; rolled, thickened, rounded, or scalloped hymenal edges; and bands between this membrane and the surrounding tissues are also indicative of sexual abuse. 29, 36 Although there continues to be disagreement among examiners as to the significance of certain findings, all investigators caution that sexual molestation may cause no soft-tissue changes, and the examination may be perfectly normal. Soft-tissue changes of the perianal region attributed to sexual abuse may also be nonspecific. They are said to include erythema, hyperpigmentation, loss of normal skin folds, a prominent anal verge, venous engorgement, skin tags, anal twitching, reflex anal dilatation, fissures, scars, and occasionally lacerations. 12. 14 As in the examination of the girl's genitalia, the medical examiner faces the difficulty that some of these findings are similar to those detected in nonabused or accidentally injured children. This becomes an even greater problem when these changes are discovered in a preverbal child or in an older youngster who is unwilling to describe how an injury occurred.

ANOGENITAL FINDINGS IN NONABUSED PREPUBERTAL CHILDREN A recent study specifically designed to determine the anogenital findings in nonabused prepubertal children looked at many of these same soft-tissue findings. 16. 17 A total of 320 youngsters between the ages of 2 months and 12 years were studied. Approximately one half of the 120 boys and 200 girls were examined and photographed through the use of a colposcope. After screening for the possibility of undetected sexual abuse and the onset of puberty, the data from the remaining 267 subjects were

872

JOHN MCCANN

analyzed to determine what physical changes might be expected in this population. The greatest revelation was the variety of changes observed. The findings were influenced by several variables that included the age of the child, the examination technique used, and the racial heritage. Many other variables including the child's state of relaxation during the examination are still being analyzed. Although no hematomas or lacerations were discovered, several findings were noted that had previously been associated with childhood sexual abuse. The most common genital findings uncovered during these colposcopic examinations included erythema of the vestibule (56%), periurethral bands (50%), labial adhesions (38%), lymphoid follicles on the fossa navicularis (33%), posterior fourchette midline avascular areas (25%), and urethral dilatation with the labial traction technique (14%).16 Hymenal findings included mounds (33%), projections (33%), and "septal remnants" or midline hymenal tags (18%). Intravaginal findings of vaginal ridges (90%) and rugae (88%) were revealed primarily through the labial traction approach. Unusual findings included posterior fourchette friability (4.7%), anterior hymenal clefts (1.2%), and notches of the hymen (6%). One child had an imperforate hymen (1.2%), and four (4.4%) had hymenal septa. A vaginal discharge was detected in two (2.6%) and a foreign body in one (1.3%). The size of the hymenal orifice varied with both the age of the child and the examination technique used. The perianal findings that were encountered with the greatest frequency included erythema (41 %), increased pigmentation (30%), and venous engorgement (52%) after 2 minutes in the knee-chest position. 17 Wedgeshaped smooth areas in the midline (diastasis ani), with or without depressions, were found both anterior and posterior to the anus in 26% of the children. Anal skin tags or folds were discovered anterior to the anus in 11 % of the subjects. Nine percent of the children displayed anal dilatation in the first 30 seconds of the examination. That percentage increased to 49% if they were left in the knee-chest position for 4 minutes or longer. The anus opened and closed intermittently in 62% of the children whose anus dilated. Flattening of the anal verge and rugae occurred during dilatation by the midpoint of the examination in 44% and 34% of the subjects, respectively. Perianal findings that were found infrequently in all subgroups included skin tags or folds (0%) and scars (0%) outside the midline, anal dilatation greater than 20 mm without the presence of stool in the rectal ampulla (1.2%), irregularity of the anal orifice after complete dilatation (3%), and prominence of the anal verge (3%). No abrasions, hematomas, fissures, or hemorrhoids were encountered. Less commonly detected findings within specific subgroups included perianal erythema in girls (32%) as compared with boys (57%), pigmentation in the lighter-skinned white children (22%) when compared with black (53%) and Hispanic (58%) youngsters, and venous congestion at the beginning of the examination (7%) when compared with the same finding after 4 minutes in the knee-chest position (73%). No perianal skin tags or folds were found in the boys. Many of these findings were not detected in the first half of this study

USE OF COLPOSCOPE IN CHILDHOOD SEXUAL ABUSE EXAMINATIONS

873

before a colposcope was employed. The combination of magnification and the availability of the photographs, which could be studied in detail, significantly increased the number of observations recorded. One example was the percentage of children noted to have labial adhesions. In the first half of the project, less than 15% of the youngsters were observed to have these lesions. No labial adhesions smaller than 5 mm were recorded. Following the introduction of the colposcope, the percentage of subjects with labial adhesions increased to 38.9%, with over half of the lesions being smaller than 5 mm. The detection of these minute adhesions did provide an explanation for the midline avascular lesions of the posterior fourchette that had been noted by a number of observers. 4, 14 Upon analysis of the data, it was discovered that there was an association between the avascular lesions on the posterior fourchette and labial adhesions of all sizes. The examiners found that when tension was applied to the tissues, as was done during both the supine labial traction method and in the prone knee-chest position approach, the capillary bed of the labial adhesion emptied, producing the white avascular appearance of these lesions (Fig. 4). Other notable examples of the impact of the colposcope on the results of this study were the discovery of a relatively large number of periurethral bands (51%), lymphoid follicles on the hymen and perihymenal tissues (34%), hymenal projections (38%), hymenal mounds (33%), midline hymenal remnants (21%), and intravaginal longitudinal ridges (90%). The high incidence of these entities in this population of nonabused children has led to a reinterpretation of some of the findings in the child being evaluated for possible sexual abuse.

Figure 4. A 7-year-old child sexually molested by father and uncle. Labial traction method. Narrow hymenal rim (arrow a) with exposed ridges (arrow b) at 3 and 9 o'clock. Anterior column (arrow c). Enlarged hymenal orifice. Colposcopic photographs (magnification X 10).

874

JOHN MCCANN

The small bands of tissue found on either side of the urethral meatus initially were thought to be synechiae that are caused by an injury (see Fig. 2B). It was only after these bands were discovered in the majority of children that their true identity as naturally occurring structures was recognized. Similarly, with the aid of the colposcope, other irregularities of the hymen, such as the projections, mounds, and tags were examined and associations with other anatomic structures explored. The detection of an attachment from projections and mounds on the edge of the hymen to underlying intravaginal ridges provided an explanation as to why these entities became so prominent when this membrane was damaged or attenuated (Figs. 2C, 3B, 3C, and 5). These longitudinal intravaginal ridges were usually covered by the hymen and could not be seen unless this membrane was disrupted or until they were specifically looked for through the use of different examination techniques. 16 In this study of nonabused prepubertal females, intravaginal ridges were found in 90% of the girls. They were only discovered after the supine labial traction technique was employed and the vaginal wall immediately behind the hymen was examined. They were most frequently located between the 2 and 10 o'clock positions. Another anomaly that has led to some controversy among examiners is a tag of tissue that is occasionally found on the posterior rim of the hymen. This is believed by some to be a hymenal tag. 20 Others believe it represents a septal remnant because it may have a counterpart anteriorly that looks like the other half of a septum. 16 Still others have thought that it represents the healing process of an injured hymen. 9 In the study of nonabused children, these tags were discovered in 18% of the girls. 16 They were usually located in the midline and tended to form small mounds of tissue on the edge of the hymen during the supine examination approaches. In the prone knee-chest position they fell forward of their own weight and became small triangular projections with smooth, sharp edges (see Fig. 2C). When viewed under magnification, a narrow band of thickened hymenal tissue was frequently seen extending from this projection onto the fossa navicularis. The combination of the relatively high incidence of these tags in this population of children, along with the normal appearance of the adjacent hymenal tissues led the authors to the conclusion that these were minor congenital anomalies that most likely were remnants of a hymenal septum. 16 Of all the physical findings associated with childhood sexual abuse, the size of the hymenal orifice has received the most attention. Various methods for obtaining an accurate measurement of this orifice have been proposed. 1, 10, IS, 26, 34 The colposcope has recently been employed in this effort. Although some scopes have built-in scales, examiners have easily fashioned their own measuring devices. In the previously quoted study of nonabused children, a high interobserver correlation was found when the colposcopic photographs were used by different examiners to determine the size of this orifice, 16 In a report on the comparison of genital examination techniques, different methods were found to produce different-sized hymenal orifice measurements in the same child. IS In this latter study, the size of both the vertical and horizontal transhymenal spans were the smallest during the

USE OF COLPOSCOPE IN CHILDHOOD SEXUAL ABUSE EXAMINATIONS

875

supine labial separation approach. The prone knee-chest position produced the largest vertical diameters, whereas the supine labial traction method generated the greatest horizontal spans. In one study, it was noted that the size of the orifice varied with the age of the child.1O In another report, an overlap in sizes was observed between adjacent age groups. 10, 16 The effect of body habitus, the child's state of relaxation, and other factors that may influence the size of this orifice are yet to be determined. Although few debate that a large hymenal orifice in a prepubertal child raises the possibility of sexual abuse, many examiners are investigating other parameters in their assessment of the child suspected of having been sexually molested. 9 , 11, 31 One finding recently receiving more attention is the width of the hymenal membrane at its midline attachment along the posterior rim of the vaginal introitus. 9 During actual or attempted vaginal penetration, it is this portion of the hymen that is most likely to be damaged. A narrow or attenuated hymen at this 6 o'clock location, when the child is being examined in the supine position, or the 12 o'clock position when prone, is usually indicative of an injury to this membrane. The additional findings of mounds (bumps), projections, or notches on the edge of the hymen and the exposure of intravaginal ridges further strengthens the possibility that abuse has occurred (Fig. 4). The detection of these findings is usually enhanced by some form of magnification. As helpful as the colposcope is in the genital portion of the examination of the female patient, it is of limited usefulness in the evaluation of the perianal region. It is of value in the documentation of recording soft-tissue injuries, but the recording of erythema, increased pigmentation, persistence of a prominent anal verge, thickened anal skin folds, marked venous congestion, skin tags, and anal dilatation are best chronicled at lower magnifications. Exceptions to this ar~ the small lacerations, superficial abrasions, and other minor skin and mucosal changes that may occur as a result of acute trauma. Through the use of vital stains, such as toluidine blue or Lugor s solution, along with the employment of magnification, even minute cellular changes may be detected. 19, 25

CONGENITAL ANOMALIES Minor congenital anomalies of the anogenital region are normally discovered without the use of magnification. Imperforate hymens, cribriform hymens, septa of the hymen, septal remnants of the hymen, and asymmetry of the hymenal orifice are some of the abnormalities that can be found in this membrane. Occasionally these anomalies create confusion during the evaluation of the child suspected of having been sexually abused. In this situation, a series of magnified photographic images, taken during different examination methods, may help to clarifY the origin of the finding. Several congenital anomalies of the perianal region may also cause the medical examiner some bewilderment. One is a smooth wedge-shaped area in the midline of the anal verge (Fig. 5). These defects, which can be located either anterior or posterior to the anus, were found in 26% of the children in the study of nonabused preadolescents. 17 They were associated

876

JOHN MCCANN

Figure 5. A 9-year-old boy in knee-chest position. Congenital anomaly, diastasis ani (arrow a), with a depression anterior to anus. From a study of nonabused children, ref 10. Colposcopic photograph (magnification x 10).

Figure 6. A 4-year-old girl with perianal skin fold (arrow a) in midline anterior to the anus. Knee-chest position. From a study of nonabused children, ref. 10.

USE OF COLPOSCOPE IN CHILDHOOD SEXUAL ABUSE EXAMINATIONS

877

Figure 7. A 4-month-old female infant with a perineal groove (arrow a) extending from the fossa navicularis to the internal anal sphincter. Supine position. Colposcopic photograph (magnification x 10).

with depressions or dimples in approximately one half of the children with this entity. The abnormality of the underlying superficial division of the external anal sphincter muscle is the basis for the name given to this finding that is now called "diastasis ani." Two other perianal anomalies that may be difficult to differentiate from lesions caused by an injury are perianal skin folds and perineal grooves. Perianal folds are tags of tissue that are always found in the midline just beyond the anal verge (Fig. 6). In the study of nonabused prepubertal children, they were found only in girls and invariably anterior to the anus. 17 Unlike sentinel tags, which are formed in response to a fissure, no disruption of the tissues immediately beneath the fold will be found, even with the use of magnification. Perineal grooves may be mistaken for a laceration or a fissure and are easily seen without the aid of a colposcope. 2 These congenital lesions, which run from the fossa navicularis to the internal anal sphincter, form a shallow groove along the length of the median raphe (Fig. 7). They are created by a failure of the skin overlying the perineal body to fuse during fetal development. They are usually asymptomatic and require no therapy because they will gradually become epithelialized with normal skin. 32 SEXUALLY TRANSMITTED DISEASES The documentation of inflammation or discharge caused by a sexually transmitted disease does not require the use of a colposcope. This instrument can be of help, though, in the detection of the early lesions of condyloma acuminata. Through the application of a dilute acetic acid

878

JOHN MCCANN

solution and the use of magnification, even small lesions can be identified. 10 Although rarely necessary for the diagnosis, "strawberry spots" or ecchymotic papillae on the vaginal wall, which indicate the presence of the parasite Trichomonas vaginalis, are more easily seen with the use of the colposcope. 28 Conversely, the presence of a sexually transmitted disease may occasionally be ruled out when an examiner, using a colposcope, is successful in finding a foreign body that proves to be the cause of a vaginal discharge in a young child.

SUMMARY The addition of the colposcope to the armamentarium of the medical investigator of childhood sexual abuse has many advantages. The ability to accurately record anatomic findings has implications for the medical community, the judicial system, the accused, the victim, and the family. Through the use of the photographs produced by this instrument, examiners can inspect physical findings without the time constraints imposed by the short attention span of the young child. If necessary, colleagues can be consulted and interpretations discussed. The availability of these photographs to the courts has reduced the need to re-examine the child for another opinion. The colposcopic photograph has also proved to be an excellent teaching and research tool. As an aid to teaching, the photographs and slides produced by this instrument help the instructor demonstrate anatomic findings while allowing the student time to ask questions. As a research tool, this instrument has opened up a myriad of possibilities for medical examiners. It has facilitated the collection of clinical data, it has allowed the standardization of examination techniques, and with the help of computers it has made possible the sophisticated analysis of the information collected. The colposcope has limitations. Aside from its cost, it is a difficult instrument to use in the examination of the young child. The time required for an examination can increase substantially as the examiner attempts to maneuver the scope into a proper position. During this procedure, the maintenance of the child in a suitable state of relaxation, while avoiding further emotional trauma, can be a challenge. The reality that the photograph is two dimensional and represents only the findings at that moment will always be a limiting factor in its use as a means of assessing a child's anatomy. Even the multimethod approach employed to offset this problem may compound the situation by further increasing the length of the examination. Most of these and other dilemmas encountered in the use of the colposcope can be solved by additional experience with this instrument. Despite the improvements brought about by the introduction of the colposcope, more advanced technology may be needed to help solve some of the problems currently plaguing medical examiners. The use of video tape could provide a solution to the documentation of the changes that occur in the soft tissues as the child moves or becomes more or less relaxed. The potential of the computer appears unlimited, and its application to the

USE OF COLPOSCOPE IN CHILDHOOD SEXUAL ABUSE EXAMINATIONS

879

problem of the interpretation of findings could make a significant contribution to the field. Still other technologies not yet invented may bring advances currently not believed possible. Even as our technical skills grow, the importance of listening to the child and observing his or her behavior cannot be overemphasized. If we are to improve our expertise in the identification of the sexually abused child, then we must continue to develop both our physical diagnostic skills as well as our proficiency in the behavioral sciences. Throughout this process, the welfare of the child must always be considered. The avoidance of further abuse must be the primary concern, whether it is by the hand of the perpetrator or through the efforts of a well-meaning investigator.

REFERENCES 1. Adams JA, Ahmad M, Phillips P: Anogenital findings and hymenal diameter in children referred for sexual abuse examination. Adolesc Pediatr Gynecoll:123-127, 1988 2. Adams JA, Horton M: Is it sexual abuse? Confusion caused by a congenital anomaly of the genitalia. Clin Pediatr 28:146-148, 1989 3. Adams JA, Phillips P, Ahmad M: The usefulness of colposcopic photographs in the evaluation of suspected child sexual abuse. Adolesc Pediatr Gynecol 3:75-82, 1990 4. Altcheck A: Vulvovaginitis, vulvar skin disease, and pelvic inflammatory disease. Pediatr Clin North Am 28:397-432, 1981 5. Berkowitz CD: Sexual abuse of children and adolescents. Adv Pediatr 30:275-312, 1987 6. Cowell CA: The gynecologic examination of infants, children and young adolescents. Pediatr Clin North Am 28:247-266, 1981 7. Davis AJ, Emans Sf Human papilloma virus infection in the pediatric and adolescent patient. Pediatrics 115:1-9, 1989 8. Dube R, Hebert M: Sexual abuse of children under 12 years of age: A review of 511 cases. Child Abuse NegI12:321-330, 1988 9. Emans SJ, Woods ER, Flagg NT, et al: Genital findings in sexually abused, symptomatic and asymptomatic girls. Pediatrics 79:778-785, 1987 10. Goff CW, Burke KR, Rickenback C, et al: Vaginal opening measurements in prepubertal girls. Am J Dis Child 143:1366-1368, 1989 11. Herman-Giddens ME, Frothingham TC: Prepubertal female genitalia: Examination for evidence of sexual abuse. Pediatrics 80:203-208, 1987 12. Hobbs CJ, Wynne JM: Buggery in childhood-A common syndrome of child abuse. Lancet: 792-796, 1986 13. Hobbs CJ, Wynne JM: Child sexual abuse-an increasing rate of diagnosis. Lancet 2:837841, 1987 14. Hobbs CJ, Wynne JM: Management of sexual abuse. Arch Dis Child 62:1182-1187, 1987 15. Marshall WN, PuIs T, Davidson C: New child abuse spectrum in an era of increased awareness. Am J Dis Child 142:664-667, 1988 16. McCann J, Voris oj, Simon M, et al: Genital findings in prepubertal females selected for non-abuse: A descriptive study. Pediatrics, in press 17. McCann J, Voris J, Simon M, et al: Perianal findings in prepubertal children selected for nonabuse: A descriptive study. Child Abuse NegI13:179-193, 1989 18. McCann J, Wells R, Voris J, et al: Comparison of genital examination techniques in prepubertal females. Pediatrics 85:182-187, 1990 19. McCauley J, Gorman RL, Guzinski G: Toluidine blue in the detection of perineal lacerations in pediatric and adolescent sexual abuse victims. Pediatrics 78: 1039-1043, 1986 20. Merlob NM, Reesner SH: Types of hymen in the newborn infant. Eur J Obstet Gynecol Reprod BioI (Israel) 22:225-228, 1986 21. Muram 0: Child sexual abuse-Genital tract findings in prepubertal girls: I. The unaided medical examination. Am J Obstet Gynecol 160:328-332, 1989

880

'JOHN MCCANN

22, Muram D: How you can detect and treat child sex abuse. Contemp Obstet Gynecol 31:34-48, 1988 23. Muram D: Rape, incest, trauma: The molested child. Clin Obstet Gynecol 30:754-761, 1987 24. Muram D, Elias S: Child sexual abuse--genital tract findings in prepubertal girls: II. Comparison of colposcopic and unaided examinations. Am J Obstet Gynecol 160:333335, 1989 25. Norvell MK, Benrubi GI, Thompson RJ: Investigation of microtrauma after sexual , intercourse. J Reprod Med 29:269-271, 1964 26. Paul DM: "What really did happen to Baby Jane?"-The medical aspects of the investigation of alleged sexual abuse of children. Med Sci Law 26:85-102, 1986 27. Reinhart M: Sexually abused boys. Child Abuse Negl11:229-235, 1987 28. Ricci LR: Medical forensic photography of the sexually abused child. Child Abuse Negl 12:305-310, 1988 29. San Filippo JS, Schikler KN: IdentifYing the sexually molested preadolescent girl. Pediatr Ann 15:621-624, 1986 30. Spencer M, Dunklee P: Sexual abuse of boys. Pediatrics 78:133-138, 1986 31. Steiner H, Taylor T: Description and recording of physical signs in suspected child sexual abuse. Br J Hosp Med 40:348-351, 1988 32. Stephens FD, Smith ED: Ano-rectal Malformations in Children. Chicago, Year Book Medical Publishers, 1971, pp 114-116 33. Teixeira WR: Hymenal colposcopic examination in sexual offenses. Am J Forensic Med Pathol 3:209-214, 1981 34. White S, Ingram D: Vaginal introital diameter in the evaluation of sexual abuse. Child Abuse NegI13:217-224, 1989 35. Woodling BA, Heger A: The use of the colposcope in the diagnoses of sexual abuse in the pediatric age group. Child Abuse NegllO:111-114, 1986 36. Woodling BA, Kossoris PD: Sexual misuse: Rape, molestation and incest. Pediatr Clin North Am 28:481-499, 1981

Address reprint requests to John McCann, MD Valley Medical Center Department of Pediatrics 445 South Cedar Avenue Fresno, CA 93702