当前位置: 首页 > 期刊 > 《小儿科》 > 2005年第10期 > 正文
编号:11343145
Anogenital and Respiratory Tract Human Papillomavirus Infections Among Children: Age, Gender, and Potential Transmission Through Sexual Abus
http://www.100md.com 《小儿科》
     Departments of Pediatrics Otolaryngology, Wake Forest University School of Medicine, Winston-Salem, North Carolina

    ABSTRACT

    Objectives. To evaluate human papillomavirus (HPV) presentation among children <13 years of age and its association with suspected child sexual abuse (CSA), and to assess sexual abuse consideration among different clinical services treating these children.

    Methods. Records of children <13 years of age from 1985 to 2003 were selected for review if the children had a HPV-related International Classification of Diseases, Ninth Revision, code or had been examined in the CSA clinic. Abstracted data included demographic features, clinical findings, clinical services involved, age at diagnosis, age when care was first sought, and age when symptoms were first noted.

    Results. HPV was identified by clinical examination and/or biopsy for 124 children, 40 with laryngeal lesions, 67 with anogenital lesions, 10 with oral lesions, and 7 with both anogenital and oral lesions. The mean age at HPV diagnosis was 4.0 ± 2.9 years, compared with 6.4 ± 3.0 years for 1565 HPV-negative children. Among 108 HPV cases with data for age when symptoms were first noted, the mean age was 3.3 ± 2.9 years (median: 2.2 years) for children with anogenital and oral HPV and 2.4 ± 2.3 years (median: 1.9 years) for children with laryngeal HPV. Among HPV-positive patients, 56% were female, compared with 82% of HPV-negative children. Fifty-five (73%) of 75 children with anogenital HPV infections were referred to the CSA clinic for evaluation, compared with none of 49 children with laryngeal or oral HPV infections treated by the otolaryngology service. Laryngeal cases presented earlier than anogenital and oral lesions. Abuse was considered at least possible for 17 of 55 children with any CSA evaluation. The mean age of likely abused, HPV-positive children was 6.5 ± 3.8 years (median: 5.3 years), compared with 3.6 ± 2.3 years (median: 2.6 years) for likely not abused, HPV-positive children. The likelihood of possible abuse as a source of HPV infection increased with age. The positive predictive value of HPV for possible sexual abuse was 36% (95% confidence interval: 13–65%) for children 4 to 8 years of age and 70% (95% confidence interval: 35–93%) for children >8 years of age.

    Conclusions. The data from this epidemiologic study of HPV suggest that many anogenital and laryngeal HPV infections among preadolescent children are a result of nonsexual horizontal transmission, acquired either perinatally or postnatally. It seems that many children >2 years of age acquire HPV infection from nonsexual contact. Different subspecialties vary greatly in their suspicion and evaluation of CSA. At this time, there remains no clear age below which sexual abuse is never a concern for children with anogenital HPV infections. Every case needs a medical evaluation to determine whether enough concern for abuse exists to pursue additional investigations.

    Key Words: child sexual abuse human papillomavirus anogenital warts laryngeal papillomatosis positive predictive value

    Abbreviations: HPV, human papillomavirus CSA, child sexual abuse STD, sexually transmitted disease CPS, child protective services CI, confidence interval PPV, positive predictive value ICD-9, International Classification of Diseases, Ninth Revision PCR, polymerase chain reaction

    Human papillomavirus (HPV) is prevalent worldwide and has no race, age, or gender barriers. HPV can infect all skin and mucosal sites. The majority of papillomatous lesions (warts) regress spontaneously within 2 years when immunocompetent hosts mount an effective cell-mediated immune response.1,2 More than 130 types of HPV have been identified to date. Only one third of these tend to cause anogenital, oral, or laryngeal infections, with types 6, 11, and 16 predominating among these.2,3

    Determining the means of acquisition of anogenital HPV infections among children, for whom potential transmission through child sexual abuse (CSA) is often a consideration, continues to be a major dilemma in pediatrics. Children may acquire anogenital, laryngeal, or oral HPV infections in a variety of ways. Perinatal infection may occur transplacentally (hematogenously) via amniotic fluid during gestation and delivery and through direct exposure to cervical and genital lesions during birth.2,4–7 In addition to sexual abuse, postnatal infections can be acquired through heteroinoculation or autoinoculation from nongenital mucocutaneous HPV sources and fomite transmission.2,4,6–18 The history and physical examination often provide little definitive guidance regarding the mode of acquisition for a given child.19

    Because sexual abuse is a well-documented cause of anogenital HPV infections among children, an American Academy of Pediatrics policy statement issued in 1999 stated that condyloma acuminata suggests abuse if the infection was not acquired perinatally, although the time frame for perinatal acquisition was not delineated.6,20–31 The commonly accepted upper age limits for perinatal transmission are 12 to 24 months.6,8,11,13–15,32 Anogenital warts discovered among children more than 24 to 36 months of age are often assumed to have been acquired through sexual abuse.2,14,21,27,29,33,34

    The incubation period of HPV infection is thought to range from several weeks to several years.35 Determinations of this period are complicated by the occurrence of latent infections, which can be lifelong and can reactivate to produce lesions.3,6,32 Therefore, consideration of the incubation period, especially in comparison with time elapsed between birth and recognition of HPV infection for a child, is seldom of help in distinguishing between perinatal and postnatal acquisition of HPV infection.

    It has been recognized that subclinical genital infections (flat warts that can be very difficult to distinguish visually from surrounding uninfected tissues) and latent infections are more common manifestations of HPV infections than genital warts among adults.36–38 Subclinical HPV lesions also have been documented for both abused2,27,31 and nonabused3,39,40 children. Diagnosis of anogenital HPV infection among older children thus may reflect late recognition of subclinical lesions or recent reactivation of latent infection, rather than recent acquisition of infection (which could be sexually or nonsexually acquired); therefore, the implications of anogenital HPV infections among young children remain a subject of debate.31

    The focus on potential abuse among children with anogenital HPV infections stands in stark contrast to the current guidelines for the treatment of children with laryngeal papillomatosis provided in the otolaryngology and pediatrics literature. These guidelines, including those in the 2003 Red Book,35 state almost uniformly that juvenile-onset laryngeal papillomatosis is generally the result of perinatally acquired HPV infection,2,35,41–48 implying that consideration of sexual abuse is unnecessary in these cases. Other than a single case series of 5 children (age: 3–10 years) with oral or laryngeal papillomas for whom there was suspicion but no final determination of abuse, there are no suggestions that sexual abuse should be considered when the diagnosis of laryngeal papillomatosis is made for children.32

    Observations provided in support of this view of laryngeal papillomatosis include the following. (1) Genital HPV infections are common among women of childbearing age, with cumulative 2- to 3-year prevalences of HPV DNA detection among sexually active young women in the range of 30% to 80% in recent years.1,49–51 (2) The triad of being a vaginally delivered firstborn of a mother <20 years of age is common among children with laryngeal papillomatosis.52 (3) The illness is uncommon among infants delivered through cesarean section.48 (4) Of mothers whose children have laryngeal papillomatosis, 30% to 60% have a history of genital warts. These observations support a role for perinatal transmission in anogenital HPV infection in childhood.

    Studies of the epidemiologic features of anogenital HPV infections among preadolescent children have generally suffered from small sample size. Only 74 cases of anogenital warts among children were reported before 1990.53 Only one subsequent study described a series of >50 cases of anogenital infection.10 The epidemiologic features of laryngeal papillomatosis have been explored in greater depth and a national pediatric registry has been established in the United States.46 To our knowledge, no studies to date have compared HPV infections of the respiratory and anogenital tracts among children from the same institution during the same time period.

    We undertook this study to enhance our understanding of the epidemiologic features of laryngeal and anogenital HPV infections in childhood and the implications these may have for potential sexual abuse. The information on HPV infection in this study is limited to the spectrum of visible disease (warts or papillomas), because these are the manifestations in childhood that generally cause clinical symptoms or lead to medical attention.

    METHODS

    HPV cases were identified through review of the medical records of children <13 years of age who were treated at Brenner Children’s Hospital (Winston-Salem, NC) between 1985 and 2003. Records were selected for review if they had 1 of the following International Classification of Diseases, Ninth Revision (ICD-9), codes: 078.10, 078.11, or 078.19 (which can be used for diagnoses of condyloma or genital warts); 210.0, 210.1, 210.2, 210.4, 210.6, 210.7, 210.8, or 210.9 (benign neoplasms of the lip, oral cavity, pharynx, or laryngeal regions); 212.1, 212.2, or 212.3 (benign neoplasms of the larynx, trachea, or carina); 221.1 or 221.2 (benign neoplasms of the vagina or vulva); or 222.1 (benign neoplasms of the penis). These ICD-9 codes were chosen on the basis of local coding practices and review of available codes that may be used for HPV. In addition, visit records were reviewed for all children <13 years of age who were examined in the CSA clinic between July 1993 and December 2002, to assess age and gender distributions for comparison with children with HPV diagnoses, as well as to permit estimation of the sensitivity and specificity of HPV for CSA. Records were excluded only if there was documentation of consensual sexual activity before diagnosis of HPV.

    Physician clinical diagnosis of HPV cases was accepted for this study.54 In the CSA clinic, the general practice during the study period was to spare children biopsy procedures unless the lesions suggested but were not classic in appearance for HPV. Classic lesions were considered to be anogenital warts that were multiple and verrucous in appearance, consistent with condyloma acuminata. Small flat lesions, or lesions that were few in number but suggested warts, were biopsied for confirmation of HPV. Laboratory evaluation for syphilis was performed for most children, which ensured that lesions classified as HPV were not in fact condyloma lata. The general practice of the otolaryngology service was to excise or debulk papillomatous lesions of the oral cavity or larynx, with pathologic examinations being performed with this tissue. The age of children relative to their diagnosis of HPV was assessed in 3 ways, ie, (1) age at which the diagnosis of HPV infection was made, (2) age at which the child was first evaluated by a physician for the lesion (or referable symptoms) diagnosed subsequently as HPV infection, and (3) age at which the child’s caretaker(s) first noted the lesion (or referable symptoms) diagnosed subsequently as HPV infection. For laryngeal papillomas, referable symptoms were usually progressive hoarseness or stridor.

    Records were reviewed for evidence that CSA was considered by the evaluating clinical service. This was considered to have occurred if (1) there was mention of consideration of sexual abuse in the medical records, (2) cultures or tests for sexually transmitted diseases (STDs) were performed, or (3) child protective services (CPS) or law enforcement was involved. The types of investigations performed for possible sexual abuse were abstracted. Nine children with HPV infections underwent only partial evaluations in the CSA clinic (eg, forensic interview or discussion of the case with the CSA clinic physician by the referring provider). Levels of suspicion of abuse were very low for these children; they were combined with those who underwent full CSA clinic evaluations into a group having "any CSA clinic evaluation." A full CSA clinic evaluation included interview of the caretaker/parent, forensic interview of the child, behavioral inventory, medical history, general physical examination, anogenital examination with a colposcope, and laboratory cultures for STDs, as deemed appropriate on the basis of history and physical findings.

    For children examined in the CSA clinic, history factors used to make sexual abuse determinations included (1) results of interviews with a parent/caregiver, (2) results of a forensic interview of the child, (3) information available from CPS and/or law enforcement investigations of possible sexual abuse (eg, confessions of perpetrators), and (4) age of the child when lesions/referable symptoms were first noted and age at the time of the first evaluation for these lesions or symptoms. Generally, the latter information was recorded in the CSA clinic medical records. A forensic interview with a child was considered positive for sexual abuse if there was any age-appropriate disclosure suggesting abuse. Parental/caregiver history was positive if there was any expressed concern regarding CSA and negative if there was no concern or if the parent was unsure.

    Oral, genital, anal, and cutaneous physical examination findings were abstracted. A child with normal anogenital findings and negative results for any STD tests performed was considered negative for abuse. Physical findings were considered definitive for abuse if there was (1) evidence of acute anogenital injury (eg, bruising or laceration), (2) total absence of the hymen posteriorly or inferiorly, (3) deep notching (more than one half the width) of the hymen for a prepubertal child, or (4) evidence of another STD. Findings considered suggestive for sexual abuse included (1) <1-mm narrowing in the posterior/inferior hymen, (2) scarring in the genital or anal area outside the midline, (3) a significant difference in the width of the hymen on one side of the posterior rim, compared with the other side, or (4) anal dilation of >2 cm in the knee-chest position. For the purposes of this study, children were defined as (1) unlikely sexually abused if there was no positive or suggestive evidence of abuse from the history or physical examination investigations, as outlined above, or (2) likely sexually abused if any of the history findings were positive, as defined above, or physical findings were suggestive or definitive for sexual abuse.

    Analysis of bivariate associations of categorical variables were performed with 2 procedures. Differences between groups with continuous outcomes variables were analyzed with Mann-Whitney U tests for 2 groups and Kruskal-Wallis tests for >2 groups; t tests were used to determine 95% confidence intervals (CIs) for continuous variables, and equal variances were not assumed. Odds ratios and 95% CIs were determined with logistic regression analyses. SPSS 12.0 software (SPSS, Inc, Chicago, IL) was used these analyses. The exact method was used to determine 95% CIs for single binomial proportions.

    RESULTS

    Identified Cases

    Of 129 potential HPV cases identified on the basis of ICD-9 codes, 4 did not have evidence of HPV infection on review of the medical records and one 12-year-old female subject known to be consensually sexually active was excluded. The 124 HPV cases identified on the basis of clinical diagnosis and/or biopsy had 4 types of clinical presentation; 40 (32%) had laryngeal papillomas, 10 (8%) had oral papillomas, 67 (54%) had anogenital warts, and 7 (6%) had both anogenital warts and oral papillomas. There were 1565 children without evidence of HPV infection who had visits to the CSA clinic during the study period for evaluation of possible sexual abuse.

    Age

    The children with HPV infection were younger than those without HPV infection who were examined in the CSA clinic (4.0 ± 2.9 years versus 6.4 ± 3.0 years, P < .001) (Table 1). The age distribution of the 124 HPV cases is shown in Fig 1. Among the 4 HPV infection presentation types, children with laryngeal papillomatosis were younger at the time of diagnosis than were the other groups (laryngeal papillomas versus anogenital warts and oral papillomas, P .018; laryngeal papillomas less than anogenital warts with oral papillomas, P = .078) (Table 1). The ages of the 3 anogenital wart and/or oral papilloma groups were not statistically different, and these subjects were grouped together for subsequent analyses.

    Age was evaluated in 3 ways for both anogenital and laryngeal HPV infections, ie, (1) age at which the diagnosis of HPV infection was made, (2) age at which the child was first evaluated by a physician for the lesion (or referable symptoms) diagnosed subsequently as HPV, and (3) age at which the child’s caretaker(s) first noted the lesion (or referable symptoms) diagnosed subsequently as HPV. Children with laryngeal papillomas were younger than those with anogenital warts and/or oral papillomas in all 3 measures (Fig 2), although the difference in the ages when symptoms were first noted did not reach statistical significance. The median age for each measure was 2.2 years, except for the age when symptoms were first noted for laryngeal papillomas, for which the median was 1.9 years.

    Gender

    The proportion of male patients was greater among those with HPV infection than among the HPV-negative children examined in the CSA clinic (44% vs 18%, P < .001). For the 2 subgroups of HPV infection, male patients predominated slightly (52%) among the 40 cases of laryngeal papillomas but female patients represented 60% of those with anogenital warts and/or oral papillomas. This difference was not statistically significant (P = .29) (Table 1). The gender distributions among the 84 children with anogenital and/or oral papillomas (40% male and 60% female) and HPV-negative children examined in the CSA clinic (18% male and 82% female) were different (P < .001).

    Race/Ethnicity

    There were no differences in race/ethnicity (1) between the children with HPV infection and those without HPV infection or (2) among the 4 HPV infection subgroups as listed in Table 1.

    Biopsy and Serotyping

    Biopsy confirmation was not required for this study, but 66 cases (53%) had biopsy specimens sent for histopathologic examinations, which showed findings consistent with HPV infection. The biopsy frequency varied according to clinical service. Forty-seven of 49 children with laryngeal or oral papillomas who were treated by the otolaryngology service underwent biopsy (Table 2). Children with anogenital warts, with or without associated oral papillomas, underwent biopsy in 19 (26%) of 74 cases. The 1 oral papilloma case not treated by the otolaryngology service was not biopsied. HPV serotype was available for only 4 cases, with types 6, 11, and/or 16 being present in each case.

    Skin Warts and Parental HPV Infection

    Five of 124 children had concurrent skin (nongenital and nonoral) warts. Three of these patients had a parent with a history of HPV infection. Three had anogenital warts; 1 had an oral papilloma; and 1 had a laryngeal papilloma. The age at the time of diagnosis ranged from 5.3 to 9.5 years. Among 54 HPV cases with information regarding the presence or absence of maternal HPV infection, 10 mothers had a history of genital warts, 11 had a history of cervical neoplasia, and 5 had a history of HPV infection that was otherwise undefined in the records. Among 35 cases with information regarding the presence or absence of paternal HPV infection, 1 father had penile warts and 14 had a history of HPV infection that was otherwise undefined in the records.

    Considerations of CSA Among Clinical Services

    The clinical service that first evaluated the child, either making the diagnosis of HPV infection or referring the child for CSA clinic evaluation, and whether sexual abuse was considered were ascertained from the medical records. Otolaryngology (49 cases, 40%), dermatology (23 cases, 18%), and pediatric surgery or urology (20 cases, 16%) were the most common clinical services evaluating children in this series (Table 2). In 17 cases, either the CSA clinic was the site of the first medical evaluation (direct referral from CPS or law enforcement) or the source of referral to the CSA clinic was unclear. Overall, the possibility of CSA was considered in 71 (57%) of the 124 cases, and 55 children (44%) were referred to the CSA clinic for evaluation (Table 2).

    The abuse consideration and referral practices varied among the clinical services to which these children presented initially, with none of the 49 children examined by otolaryngology for laryngeal or oral papillomas being referred or considered for abuse. Among the 75 children evaluated by other clinical services, the percentages of children for whom abuse was considered and who were referred to the CSA clinic for additional evaluation were 95% and 73%, respectively (Table 2). Sexual abuse was at least considered for 98% of these girls (44 of 45 patients) and 90% of the boys (27 of 30 patients; P = .34); 69% of girls (31 of 45 patients) underwent full CSA evaluations, compared with 50% of boys (15 of 30 patients; P = .16). There was no association between race/ethnicity and (1) the clinical service that first evaluated or referred the child to the CSA clinic or (2) whether the child underwent any or full CSA clinic evaluation.

    Role of Sexual Abuse in HPV Transmission

    Of the 55 children with HPV infections who underwent at least partial evaluation in the CSA clinic, 17 (31%) were considered likely sexually abused. None of these children had evidence of other STDs. Among the 1565 HPV-negative children evaluated in the CSA clinic, 59% were likely sexually abused. Girls were more likely than boys to have positive forensic interviews and physical findings of abuse in examinations among the HPV-negative children but not among the HPV-positive children (Table 3). However, the sample size of HPV-positive children did not have sufficient power to identify differences on the order of 10% as statistically significant. The types of evidence supporting the determinations of sexual abuse are listed in Table 3.

    Among the 55 HPV-positive children, there was no association of gender with sexual abuse in this sample. The age characteristics of the 17 children likely and the 38 children unlikely to have been sexually abused are presented in Fig 3 and Table 4. Those likely to have been abused were older than those unlikely to have been abused at the age of diagnosis (difference in means: 3.0 years; 95% CI: 1.1–4.9 years; P = .001), the age when care was first sought (difference in means: 2.9 years; 95% CI: 0.8–5.0 years; P = .005), and the age when lesions or symptoms were first noticed (difference in means: 3.6 years; 95% CI: 1.3–5.9 years; P = .001). Risk of abuse increased with increasing age, with an odds ratio of 1.4 (95% CI: 1.1–1.8) for each increasing 1 year of age. HPV-positive children 4 to 8 years of age were 2.9-fold more likely (95% CI: 0.7–12.4) and those >8 years of age were 12.1–fold more likely (95% CI: 2.3–63.6) to have been sexually abused than were HPV-positive children <4 years of age.

    This sample of HPV-positive and -negative children with an assessment of sexual abuse status (likely positive or likely negative) allows estimation of the sensitivity, specificity, likelihood ratios, and predictive values of visible anogenital warts for sexual abuse. These are presented in Table 5.

    Among these 55 cases, information on the presence or absence of maternal HPV infection was available for 38 and paternal HPV infection for 31. A history of maternal or paternal HPV infection was not associated with likely sexual abuse for these subsets of children with HPV infections (Table 4).

    DISCUSSION

    Modern studies using polymerase chain reaction (PCR)-based methods have shown the prevalence of genital HPV infection to be as high as 80% among sexually active young women.49 Anogenital and respiratory tract HPV infections among children have increased in recent decades, paralleling the increasing trend of genital HPV infections among adults that began in the 1960s.8,32,55,56 Laryngeal papillomatosis occurs at a rate of 4 cases per 100000 in the pediatric population. Approximately 7 of 1000 children born to mothers with vaginal condyloma develop laryngeal papillomatosis during childhood.41 Concern regarding sexual abuse as the explanation for recognized anogenital HPV infections among children continues to be high among health care providers who evaluate these children, but there seems to be little concern about sexual abuse when the site of infection is the upper respiratory tract. Should there be more balance between the perspectives of these 2 sets of providers

    In this study of anogenital and respiratory tract HPV infections among children <13 years of age, male patients represented a greater proportion (44% vs 18%) of the 124 children with papillomas than of the 1565 children without recognized HPV infection who were referred to the CSA clinic for evaluation of possible sexual abuse. Male patients represented 52% of 40 children with laryngeal papillomas and 40% of 84 children with anogenital and/or oral warts (not statistically different between the 2 HPV infection groups). Those with HPV infections also were significantly younger (median age: 3.1 vs 6.0 years) than children without HPV infections who were examined in the CSA clinic. However, there was considerable overlap in age between the 2 groups.

    We also observed that children with laryngeal papillomatosis were younger on average at the time of diagnosis than were those with anogenital warts and/or oral papillomas (median age: 2.3 vs 3.3 years). Those with oral papillomas seemed more similar in age to the children with anogenital warts than to those with laryngeal papillomas. The differences between the anogenital and laryngeal groups lessened and were no longer statistically significant when based on the age when lesions or symptoms referable to the HPV infection were first noted by parents (1.9 vs 2.2 years).

    Anogenital Lesions, Age, Gender, and Sexual Abuse

    Girls are 2 to 4 times more likely than boys to be sexually abused.57,58 The ratio of girls to boys among the 1565 children without recognized HPV lesions who were examined in the CSA clinic was similar (4:1). Taken together, the younger age distribution and greater proportion of male patients among the children with anogenital or laryngeal HPV infections, compared with HPV-negative children referred to the CSA clinic because of concerns about possible sexual abuse, suggest that perinatal HPV transmission, which does not seem to have a gender bias, plays a substantial but not universal role in HPV infection among these children.

    Among the 55 children with anogenital warts who were evaluated in the CSA clinic, 17 (31%) had findings at least suggesting sexual abuse. The positive predictive value (PPV) of anogenital warts for sexual abuse among children 2 to 12 years of age was 37% (95% CI: 23–52%). For children 4 to 8 years of age and for those >8 years of age, the PPVs were 50% and 70%, respectively. HPV-positive children 4 to 8 years of age were 2.9-fold more likely (95% CI: 0.7–12.4) and those >8 years of age were 12.1-fold more likely (95% CI: 2.3–63.6) to have been sexually abused than were HPV-positive children <4 years of age. These observations support an increasing role for sexual abuse as a cause of HPV lesions with increasing age in childhood. With the definition of abuse we used (inclusion of possible as well as definite cases), the odds ratios and PPVs for anogenital warts indicating sexual abuse may represent overestimates, but not to a degree that would allow the potential for abuse to be dismissed among these children.

    Our overall PPV was higher than most of those reported in 8 studies that evaluated the frequency of sexual abuse among preadolescent children with HPV infections (visible lesions). Among the 5 series with 31 to 75 cases, the frequency of sexual abuse was 3% to 10%.10,13,15,59,60 Three series with smaller numbers reported frequencies of 24% (n = 25),61 35% (n = 17),62 and 91% (n = 11).26 These frequencies are equivalent to PPVs. Six of these series consisted of boys and girls, whereas 2 included only girls.26 Some of the variation among these studies, including ours, may be attributable to differences in the evaluations performed to investigate possible sexual abuse, the population studied, or referral pattern bias.55

    Given the substantial role of perinatal transmission for anogenital and respiratory tract HPV infection among children, along with the increasingly likelihood of abuse with increasing age, an important question that remains unanswered is the following: "Is there a reasonable age above which abuse should be considered as a highly likely cause of HPV infection among children" Among the studies and reviews that suggest 24 months of age as the upper limit for anogenital warts to be attributable to perinatal transmission,6,8,11,13,14,32,33,35,60 this dividing line seems based primarily on consideration of the incubation period of HPV; 2 years is considered "outside the plausible incubation period of perinatal transmission of HPV."2 However, longer incubation periods (5 years) are allowed for laryngeal papillomas.2

    The mean and median ages of the children with anogenital HPV in our study were 4.5 and 3.3 years, respectively. This is similar to the mean ages of 4.2 and 4.4 years in 2 case series of children born to mothers with known HPV infection.15,60 Children with anogenital warts whose mothers had visible genital lesions at the time of delivery had a mean age of presentation of 2.8 years.15 In a series of children referred for evaluation of possible sexual abuse, the mean age of those with anogenital warts was 3.4 years.62 Collectively, these studies suggest that the majority of preadolescent children with anogenital warts present well beyond 2 years of age. If a minority of anogenital HPV cases among preadolescents result from sexual abuse and the majority of cases occur at >2 years of age, then the use of age alone (eg, a threshold of 24 months) to declare sexual abuse as the cause of anogenital HPV infection could subject many innocent families to false accusations and potentially damaging intrusions.

    Also, discordance of infant and maternal HPV serotypes is seen in approximately one third of cases in prospective studies.7 There is a growing body of evidence that demonstrates that anogenital HPV lesions among children are sometimes caused by serotype 2, a HPV type commonly responsible for nongenital cutaneous lesions.8,10,11,13,14,32,60 This suggests that some cases of anogenital HPV among children are the result of postnatal acquisition through casual contact, although abuse through "fondling" may be an explanation in some cases. Fomite transmission of anogenital HPV through bidet use, shared undergarments, and bathing towels was demonstrated in a prospective study.16

    Laryngeal Papillomatosis and Consideration of Sexual Abuse

    On the basis of current literature guidance, the observation that sexual abuse was considered for 71 of 75 children with anogenital HPV lesions but none of 49 children evaluated for laryngeal or oral lesions by our otolaryngology clinic is not surprising. This serves to illustrate the current divide in approach that exists with respect to infection with the same virus in different anatomic sites.

    The ages of children with laryngeal papillomatosis in our series (mean and median of 3.1 and 2.3 years, respectively, at diagnosis) (Fig 2) are similar to those in other recent studies. A series of 73 laryngeal cases reported in 2004 demonstrated a mean age of 3.7 years (range: 4 months to 12 years).48 In a national registry of cases seen in 22 centers from 1996 through March 2002, among 480 children for whom age at diagnosis was known, 383 (80%) were <7 years and 62% were <4 years of age.46 The age distribution of laryngeal papillomas is clearly bimodal, with juvenile onset peaking between 2 and 5 years of age and adult onset peaking between 20 and 30 years of age.2,47 Seropositivity for HPV IgM follows a similar bimodal distribution, with peaks between 2 and 5 years and 13 and 16 years of age.63

    Although these and other data support a substantial role for perinatal transmission of HPV among children with laryngeal papillomatosis, these do not necessarily support a conclusion that perinatal transmission is universal in juvenile laryngeal papillomatosis. Adult respiratory tract HPV infections are readily attributed to oral-genital contact.2,43,44,47 It also remains unclear whether different sites of lesions among children (oropharyngeal versus laryngeal or tracheal) have different implications for the presence or absence of sexual abuse.

    What Do We Do Now

    Although CSA is not the likely cause of HPV in the majority of cases involving <4-year-old children with anogenital warts, it still should be considered. We continue to recommend that children with anogenital HPV lesions undergo a complete medical evaluation for CSA, including an interview of parents or caretakers, an age-appropriate interview of the child by an interviewer experienced in the area of CSA evaluation, a thorough physical examination, and evaluation for other STDs, on the basis of history risk factors and physical findings. After such evaluation, select referral to CPS may be made if there is a reasonable basis for suspicion for abuse other than the presence of HPV lesions. We encourage consideration of potential abuse for children with laryngeal papillomas, especially older children, and possibly for those with oropharyngeal lesions.

    The primary limitation of our study is its retrospective nature. We had less comprehensive information regarding birth history, household composition, and HPV infection status among adults and other children in the home than we would have liked. Possibly we misclassified the abuse status of some of the children in our series. Identification of sexual abuse is not straightforward at any age and can be especially difficult among young children, given the importance of victim disclosure in the determination of whether abuse has occurred. If anything, we think we erred toward overclassifying children as having been abused, which would mean we overestimated the PPV of HPV for CSA. A large number of classification errors, with significant overall underclassification of children as having been sexually abused, would be required to alter the primary implications of this study. Our data also may not be readily generalizable to populations in which subclinical or latent HPV infections are detected. Of note, our age and gender characteristics are similar to those reported in other studies, and this is the first study to address the issues of age at diagnosis versus age at initial presentation, age at the time the infection first might have been evident, and differences between clinical specialties.

    This study focused on the presence of visible, exophytic, HPV lesions. Clinical diagnosis identifies only HPV infections that manifest as condyloma acuminata. With the advent of molecular methods, it has become clear that HPV is much more widely prevalent than visible lesions alone suggest. Cytology studies show that 3% to 11% of women have had HPV. Prevalence increases to 33% with visual colposcopy and to 26% to 80% with PCR techniques.64,65 Intermittent shedding of the virus from nonvisible lesions also can lead to false-negative test results.2,66 The importance of nonvisible HPV infection among children, with respect to potential CSA and long-term oncogenic potential, remains unclear.

    CONCLUSIONS

    The majority of anogenital and laryngeal HPV infections among preadolescent children are likely the result of nonsexual horizontal transmission. However, sexual abuse is a possible cause, perhaps more so for anogenital lesions than for laryngeal lesions (or perhaps not, especially among older children). Postnatal nonsexual infections also may occur. There is no definite age cutoff that distinguishes between modes of transmission of HPV, but the likelihood of sexual abuse increases with age. Many children >2 years of age with anogenital HPV infections are unlikely to have suffered sexual abuse. All children with anogenital warts should undergo evaluation by someone with expertise in CSA. Referral to CPS for young children with anogenital warts should be guided by overall history and clinical findings. CPS referral likely should still be routine for children who present with anogenital warts at >4 years of age. CPS evaluation of the household may be revealing sometimes when thorough evaluation of the child and questioning of the primary caretakers have not been. When referring patients to CPS, physicians must acknowledge that the HPV lesions might have been acquired through nonsexual means.

    Additional research is needed to better ascertain potential indicators of sexual abuse versus other mechanisms of HPV transmission, including issues of latency and subclinical lesions. Determining the age at which symptoms of infection were first evident may be useful in some cases and should be reported in future clinical studies. The utility of investigation for the presence of HPV infection among other family members,8 serotyping of HPV strains, use of PCR-based diagnostic tests, and use of standardized forensic interviews and examinations of children with evidence of HPV infection should be evaluated. The dichotomy of practice, with respect to possible sexual abuse, for anogenital versus laryngeal HPV infection merits additional discussion and more formal prospective study. Meanwhile, we must tread carefully the fine line between protecting children from additional harm and plunging innocent families into disruption and duress.

    FOOTNOTES

    Accepted May 26, 2005.

    No conflict of interest declared.

    Dr Sinclair’s current address is: Department of Pediatric Emergency Medicine, Emergency Medical Services, The Children’s Mercy Hospital, Kansas City, MO 64108.

    REFERENCES

    Richardson H, Kelsall G, Tellier P, et al. The natural history of type-specific human papillomavirus infections in female university students. Cancer Epidemiol Biomarkers Prev. 2003;12 :485 –490

    Vanchiere J, Demmer G. Human polyomaviruses and papillomaviruses. In: Feigin R, Cherry J, Demmler G, Kaplan S, eds. Textbook of Pediatric Infectious Diseases. 5th ed. Philadelphia, PA: Saunders; 2004:1809–1831

    Gibbs NF. Anogenital papillomavirus infections in children. Curr Opin Pediatr. 1998;10 :393 –397

    Syrjanen S, Puranen M. Human papillomavirus infections in children: the potential role of maternal transmission. Crit Rev Oral Biol Med. 2000;11 :259 –274

    Armbruster-Moraes E, Ioshimoto LM, Leao E, et al. Presence of human papillomavirus DNA in amniotic fluids of pregnant women with cervical lesions. Gynecol Oncol. 1994;54 :152 –158

    Gutman LT, Herman-Giddens ME, Phelps WC. Transmission of human genital papillomavirus disease: comparison of data from adults and children. Pediatrics. 1993;91 :31 –38

    Rintala MA, Grenman SE, Puranen MH, et al. Transmission of high-risk human papillomavirus (HPV) between parents and infant: a prospective study of HPV in families in Finland. J Clin Microbiol. 2005;43 :376 –381

    Armstrong DK, Handley JM. Anogenital warts in prepubertal children: pathogenesis, HPV typing and management. Int J STD AIDS. 1997;8 :78 –81

    Bergeron C, Ferenczy A, Richart R. Underwear: contamination by human papillomaviruses. Am J Obstet Gynecol. 1990;162 :25 –29

    Cohen BA, Honig P, Androphy E. Anogenital warts in children: clinical and virologic evaluation for sexual abuse. Arch Dermatol. 1990;126 :1575 –1580

    Fairley CK, Gay NJ, Forbes A, et al. Hand-genital transmission of genital warts An analysis of prevalence data. Epidemiol Infect. 1995;115 :169 –176

    Gorzen J, Robertson I, Inoue M, et al. Detection of HPV DNA in genital condylomata acuminata in female prepubescent children. Adolesc Pediatr Gynecol. 1989;2 :224 –229

    Handley J, Hanks E, Armstrong K, et al. Common association of HPV 2 with anogenital warts in prepubertal children. Pediatr Dermatol. 1997;14 :339 –343

    Lowy G. Sexually transmitted diseases in children. Pediatr Dermatol. 1992;9 :329 –334

    Obalek S, Jablonska S, Favre M, et al. Condylomata acuminata in children: frequent association with human papillomaviruses responsible for cutaneous warts. J Am Acad Dermatol. 1990;23 :205 –213

    Pacheco B, DiPaola G, Ribas J, et al. Vulvar infections caused by human papilloma virus in children and adolescents without sexual contact. Adolesc Pediatr Gynecol. 1991;4 :136 –142

    Pao CC, Tsai PL, Chang YL, et al. Possible non-sexual transmission of genital human papillomavirus infections in young women. Eur J Clin Microbiol Infect Dis. 1993;12 :221 –222

    Rock B, Naghashfar Z, Barnett N, et al. Genital tract papillomavirus infection in children. Arch Dermatol. 1986;122 :1129 –1132

    Kellogg ND, Parra JM, Menard S. Children with anogenital symptoms and signs referred for sexual abuse evaluations. Arch Pediatr Adolesc Med. 1998;152 :634 –641

    Gibson PE, Gardner SD, Best SJ. Human papillomavirus types in anogenital warts of children. J Med Virol. 1990;30 :142 –145

    Gutman LT. Sexual abuse and human papillomavirus infection. J Pediatr. 1990;116 :495 –496

    Neinstein LS, Goldenring J, Carpenter S. Nonsexual transmission of sexually transmitted diseases: an infrequent occurrence. Pediatrics. 1984;74 :67 –76

    Vallejos H, Del Mistro A, Kleinhaus S, et al. Characterization of human papilloma virus types in condylomata acuminata in children by in situ hybridization. Lab Invest. 1987;56 :611 –615

    American Academy of Pediatrics, Committee on Child Abuse and Neglect. Guidelines for the evaluation of sexual abuse of children: subject review. Pediatrics. 1999;103 :186 –191

    Gutman LT, St Claire K, Herman-Giddens ME, et al. Evaluation of sexually abused and nonabused young girls for intravaginal human papillomavirus infection. Am J Dis Child. 1992;146 :694 –699

    Herman-Giddens ME, Gutman LT, Berson NL. Association of coexisting vaginal infections and multiple abusers in female children with genital warts. Sex Transm Dis. 1988;15 :63 –67

    Siegfried E, Rasnick-Conley J, Cook S, et al. Human papillomavirus screening in pediatric victims of sexual abuse. Pediatrics. 1998;101 :43 –47

    Davis AJ, Emans SJ. Human papilloma virus infection in the pediatric and adolescent patient. J Pediatr. 1989;115 :1 –9

    Hanson RM, Glasson M, McCrossin I, et al. Anogenital warts in childhood. Child Abuse Negl. 1989;13 :225 –233

    Muram D. Anal and perianal abnormalities in prepubertal victims of sexual abuse. Am J Obstet Gynecol. 1989;161 :278 –281

    Stevens-Simon C, Nelligan D, Breese P, et al. The prevalence of genital human papillomavirus infections in abused and nonabused preadolescent girls. Pediatrics. 2000;106 :645 –649

    Yoshpe NS. Oral and laryngeal papilloma: a pediatric manifestation of sexually transmitted disease Int J Pediatr Otorhinolaryngol. 1995;31 :77 –83

    Bingham EA. Significance of anogenital warts in children. Br J Hosp Med. 1994;52 :469 –472

    Goldenring JM. Condylomata acuminata in the evaluation of child sexual abuse. Arch Dermatol. 1987;123 :1265 –1266

    Committee on Infectious Diseases. Human papillomavirus. In: Pickering L, ed. Red Book: 2003 Report of the Committee on Infectious Diseases. 26th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2003:448–451

    Handsfield HH. Clinical presentation and natural course of anogenital warts. Am J Med. 1997;102(5A) :16 –20

    Koutsky L. Epidemiology of genital human papillomavirus infection. Am J Med. 1997;102(5A) :3 –8

    Oriel JD. Genitoanal papillomavirus infection: a diagnostic and therapeutic dilemma. Semin Dermatol. 1990;9 :141 –147

    Smith EM, Johnson SR, Cripe T, et al. Perinatal transmission and maternal risks of human papillomavirus infection. Cancer Detect Prev. 1995;19 :196 –205

    Smith EM, Johnson SR, Cripe TP, et al. Perinatal vertical transmission of human papillomavirus and subsequent development of respiratory tract papillomatosis. Ann Otol Rhinol Laryngol. 1991;100 :479 –483

    Wiatrak BJ. Overview of recurrent respiratory papillomatosis. Curr Opin Otolaryngol Head Neck Surg. 2003;11 :433 –441

    Mounts P, Shah KV. Respiratory papillomatosis: etiological relation to genital tract papillomaviruses. Prog Med Virol. 1984;29 :90 –114

    Tseng CJ, Liang CC, Soong YK, et al. Perinatal transmission of human papillomavirus in infants: relationship between infection rate and mode of delivery. Obstet Gynecol. 1998;91 :92 –96

    Derkay CS. Recurrent respiratory papillomatosis. Laryngoscope. 2001;111 :57 –69

    Puranen MH, Yliskoski MH, Saarikoski SV, et al. Exposure of an infant to cervical human papillomavirus infection of the mother is common. Am J Obstet Gynecol. 1997;176 :1039 –1045

    Reeves WC, Ruparelia SS, Swanson KI, et al. National registry for juvenile-onset recurrent respiratory papillomatosis. Arch Otolaryngol Head Neck Surg. 2003;129 :976 –982

    Shykhon M, Kuo M, Pearman K. Recurrent respiratory papillomatosis. Clin Otolaryngol. 2002;27 :237 –243

    Wiatrak BJ, Wiatrak DW, Broker TR, et al. Recurrent respiratory papillomatosis: a longitudinal study comparing severity associated with human papilloma viral types 6 and 11 and other risk factors in a large pediatric population. Laryngoscope. 2004;114(suppl 104) :1 –23

    Brown DR, Shew ML, Qadadri B, et al. A longitudinal study of genital human papillomavirus infection in a cohort of closely followed adolescent women. J Infect Dis. 2005;191 :182 –192

    Winer RL, Lee SK, Hughes JP, et al. Genital human papillomavirus infection: incidence and risk factors in a cohort of female university students. Am J Epidemiol. 2003;157 :218 –226

    Ho GY, Bierman R, Beardsley L, et al. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med. 1998;338 :423 –428

    Shah KV, Stern WF, Shah FK, et al. Risk factors for juvenile onset recurrent respiratory papillomatosis. Pediatr Infect Dis J. 1998;17 :372 –376

    Boyd AS. Condylomata acuminata in the pediatric population. Am J Dis Child. 1990;144 :817 –824

    Bonnez W, Reichman R. Papillomaviruses. 5th ed. Philadelphia, PA: Churchill Livingstone; 2000

    Frasier LD. Human papillomavirus infections in children. Pediatr Ann. 1994;23 :354 –360

    Stumpf PG. Increasing occurrence of condylomata acuminata in premenarchal children. Obstet Gynecol. 1980;56 :262 –264

    Finkelhor D, Hotaling G, Lewis IA, et al. Sexual abuse in a national survey of adult men and women: prevalence, characteristics, and risk factors. Child Abuse Negl. 1990;14 :19 –28

    DeAngelis CD. Nurse practitioner redux. JAMA. 1994;271 :868 –871

    Handley J, Dinsmore W, Maw R, et al. Anogenital warts in prepubertal children: sexual abuse or not Int J STD AIDS. 1993;4 :271 –279

    Obalek S, Jablonska S, Orth G. Anogenital warts in children. Clin Dermatol. 1997;15 :369 –376

    Obalek S, Misiewicz J, Jablonska S, et al. Childhood condyloma acuminatum: association with genital and cutaneous human papillomaviruses. Pediatr Dermatol. 1993;10 :101 –106

    Smith-McCune K, Horbach N, Dattel B. Incidence and clinical correlates of human papillomavirus disease in a pediatric population referred for evaluation of sexual abuse. Adolesc Pediatr Gynecol. 1993;6 :20 –24

    Cason J, Kaye JN, Jewers RJ, et al. Perinatal infection and persistence of human papillomavirus types 16 and 18 in infants. J Med Virol. 1995;47 :209 –218

    Cason J, Mant CA. High-risk mucosal human papillomavirus infections during infancy and childhood. J Clin Virol. 2005;32(suppl 1) :S52 —S58

    Moscicki AB. Genital HPV infections in children and adolescents. Obstet Gynecol Clin North Am. 1996;23 :675 –697

    Wilbur DC, Stoler MH. Testing for human papillomavirus: basic pathobiology of infection, methodologies, and implications for clinical use. Yale J Biol Med. 1991;64 :113 –125(Kelly A. Sinclair, MD, Ch)