Human papilloma virus infection prior to coitarche

Research

www. AJOG.org

GENERAL GYNECOLOGY

Human papilloma virus infection prior to coitarche Daniela Doerfler, MD*; Astrid Bernhaus, MD, MSc*; Andrea Kottmel, MD; Christine Sam, MD; Dieter Koelle, MD, MSc; Elmar A. Joura, MD OBJECTIVE: The aim of our study was to determine the prevalence and

the natural course of anogenital human papilloma virus (HPV) infections in girls prior to coitarche attending an outpatient gynecological unit.

high-risk HPV DNA in 15 children (13.6%). Two girls testing positive for HPV DNA had clinical apparent warts. After 1 year, 2 children had persistent high-risk HPV DNA, and in 1 case we found a switch from high-risk to low-risk HPV DNA.

STUDY DESIGN: Specimens were taken from the anogenital region of

CONCLUSION: Subclinical genital low- and high-risk HPV infections

114 unselected 4-15 year old girls who were referred consecutively for various gynecological problems.

are common in girls without any history of sexual abuse or sexual activity. We found persistence of genital HPV infection in children, which could be a reservoir for HPV-associated diseases later in life.

RESULTS: Four girls were excluded because of sexual abuse. Low-risk

HPV-deoxyribonucleic acid (DNA) was detected in 4 girls (3.6%) and

Key words: children, girls, human papilloma virus, persistence

Cite this article as: Doerfler D, Bernhaus A, Kottmel A, et al. Human papilloma virus infection prior to coitarche. Am J Obstet Gynecol 2009;200:487.e1-487.e5.

urrently, ⬎ 200 genotypes of human papilloma viruses (HPV) have been identified and around 30 types of HPV cause specifically anogenital infections.1,2 HPV affecting the anogenital region are divided into high-risk and low-risk virus types, depending on their ability to cause malignant disease in the infected epithelium. High-risk types like 16 and 18 are responsible for about 80% of the cases of cervical cancer and are also involved in the pathogenesis of other cancer types like penile cancer3,4 or otorhinolaryngologic cancers.5-7 Whereas HPV types 6 and 11 are responsible for more than 90% of cases of

C

From the Department of Gynecology and Obstetrics (Drs Doerfler, Bernhaus, Kottmel, Sam, and Joura), Medical University of Vienna, Vienna, and the Department of Gynecology and Obstetrics (Dr Koelle), District Hospital Schwaz, Tyrol, Austria. Received July 8, 2008; revised Sept. 25, 2008; accepted Dec. 22, 2008. Reprints: Daniela Doerfler, MD, Department of Gynecology and Obstetrics, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria. daniela.doerfler@meduniwien. ac.at. *Both authors contributed equally to the study and article. 0002-9378/$36.00 © 2009 Mosby, Inc. All rights reserved. doi: 10.1016/j.ajog.2008.12.028

genital warts, they are rarely associated with malignancies. Studies have shown that subclinical infections are the most common manifestation of HPV infection. Different authors reported between 15% and 36% of subclinical infections in sexually active adults.8,9 Nevertheless, the vast majority of HPV infections regress spontaneously within 2 years in immunocompetent hosts.10 Whereas sexual activity is by far the most important way of transmission in adults, there is little known about the mode of transmission and the natural history of HPV infection in children. Although sexual abuse is considered to play an important role of HPV acquisition in children, there is strong evidence for vertical transmission during pregnancy and delivery and horizontal transmission like autoinoculation or heteroinoculation through direct contact or fomites.11-13 The purpose of our study was to determine the prevalence of subclinical anogenital HPV infections in girls prior to sexual activity and the persistence and natural history of subclinical HPV infections. The natural history and epidemiology of HPV infections in children are of particular interest because little is known about the consequences of HPV infections early in life in regard to later development of malignant diseases. Furthermore, the

knowledge of the epidemiology of HPV infection is necessary for designing effective vaccination programs. In view of possible nonsexual transmission ways of HPV infection in early childhood and infancy, current recommendations of HPV vaccination in children just prior to coitarche need to be reappraised. Thus, it could be more beneficial to vaccinate children directly after delivery or during infancy as discussed by Cason and Mant.14

M ATERIALS AND M ETHODS Participants The study sample consisted of 114 unselected consecutive 4-15 year old girls who were referred to our outpatient clinic for children and adolescent gynecology of the University Hospital of Vienna (Austria) for various gynecological problems between June 2000 and June 2001. The main diagnoses for referral were vulvovaginitis (55%), tempoanomalies (8%), ovarian cysts (6%), and suspected sexual abuse (6%). Twentyfive percent of the patients came without referral diagnosis for various reasons. The follow-up period for positive tested participants was extended up to 4 years. Those children (n ⫽ 4) with a history of sexual abuse or suspected sexual abuse were excluded from the study.

MAY 2009 American Journal of Obstetrics & Gynecology

487.e1

Research

General Gynecology

FIGURE 1

Genital warts and intact hymen in a young girl

Vulvar genital warts and intact hymen in an 8-year-old girl without any history of sexual abuse or consensual sexual activity. Doerfler. Human papilloma virus infection prior to coitarche. Am J Obstet Gynecol 2009.

Study procedures Testing for HPV DNA was part of the standard routine examination program for all girls attending the specialized unit. Therefore, no additional approval of the ethical committee was sought because the study was an audit of our routine diagnostic program. Before collecting samples for HPV-deoxyribonucleic acid (DNA) testing, participants were thoroughly examined especially for genital warts (Figure 1) and other suspicious extragenital lesions. Specimens were collected from the vagina and cervix if vaginoscopy or examination with a virgin speculum was possible. If this was not possible, the collecting sites were the vaginal introitus, the inner surface of the labia, the perineum, or the perianal region. Only 1 swab was taken on a regular basis. Hybrid Capture II HPV test (Digene, Gaithersburg, MD) was used to detect HPV DNA from the collected samples. The Hybrid Capture system uses 2 gene probes: gene probe A detects DNA of low-risk HPV types (HPV types 6, 11, 42, 487.e2

43, and 44) and gene probe B detects intermediate and high-risk HPV types (HPV types 16, 18, 31, 35, 39, 45, 51, 52, 56, 58, 59, and 68).15 In cases of positive testing for HPV, participants were followed up after 1 year at the earliest and up to 4 years after the first visit by telephone recall. In the event of repeated positive testing for HPV, participants were recalled a third time. Questionnaires regarding the mode of delivery of participants, clinically apparent warts (anogenital and nongenital), and suspicious Papanicolaou smears and subsequent HPV typing in caretakers and family members were handed out to caretakers of positive testing participants during the follow-up examination. We conducted a telephone interview with the caretaker when the participants did not show up for further examinations.

Statistical analysis Univariate analyses were used to describe the study participants and the frequency of HPV infection in this study population and contingency table analysis to compare frequencies. SPSS 14.0 (SPSS Inc, Chicago, IL) was used for statistical calculations and P ⬍ .05 was regarded statistically significant.

R ESULTS Between June 2000 and June 2001, 114 unselected consecutive 4-15 year old girls prior to coitarche were evaluated in our clinic (Figure 2); 4 girls had to be excluded from the study because of suspected or verified sexual abuse. Two of these excluded girls tested negative for HPV and the other 2 girls tested positive for low-risk HPV. The mean age of the remaining participants (n ⫽ 110) was 9.3 ⫾ 3.4 years. HPV DNA was found in 20 of 110 participants (18.2%); 4 of the children (3.6%) tested positive for low-risk HPV, and 15 children (13.6%) showed positive results for high-risk HPV (Table). One of those girls had clinically apparent genital warts, and the specimen of this girl (0.9%) contained low- and high-risk HPV DNA. Another girl with visible genital warts tested positive only for highrisk HPV- DNA. There was no statistical

American Journal of Obstetrics & Gynecology MAY 2009

www.AJOG.org significant difference in the mean age of positive (n ⫽ 20) and negative (n ⫽ 90) testing subjects (9.2 ⫾ 3.5 years vs 9.7 ⫾ 3.1 years, P ⫽ .883). A meaningful statistical correlation of positive HPV swab testing to the referral diagnosis could not be achieved because of the small numbers in some diagnosis subgroups and the great number of patients without referral diagnosis (many parents were presenting their children without referral by a specialist). For the subsequent follow-up, we excluded the 2 girls with genital warts because we wanted to evaluate the persistence of HPV DNA in children without visible lesions. Furthermore, we excluded 2 participants from follow-up examinations because they started sexual activity between first and second visit. Despite numerous efforts to reschedule appointments for examinations, we lost 8 participants during the follow-up period. We also encountered problems concerning the questionnaires because only 5 caretakers of the participants answered them. One of the caretakers willingly answered the questionnaire during the telephone interview but did not show up for the scheduled follow-up examination. One completed questionnaire was returned by a participant who had to be excluded because of the onset of sexual activity in the meantime. Five of the 8 participants following the invitation for further HPV evaluations refused to answer the questionnaire. According to the 5 returned questionnaires, all of these participants were born vaginally and neither the caretakers nor 1 of the family members had genital or extragenital warts. But there was no verification for this information by a health care provider, although we requested such an examination of the caretakers. In the follow-up, we found no HPV DNA in the sample of 1 girl who tested positive for low-risk HPV during the first visit. We also detected no HPV DNA in the samples of 4 girls who tested previously positive for high-risk HPV. Persistence of low-risk HPV DNA was detected in 2 cases during the follow-up. In 1 subject high-risk HPV DNA was found

General Gynecology

www.AJOG.org during the first visit and low- risk HPV DNA during the follow-up examination. Of the 3 girls remaining positive for HPV DNA testing, 1 did not show up for a third appointment and 1 started sexual activity between the first follow-up examination and the second recall. Nevertheless, we tested the latter girl and found no HPV DNA anymore. The third participant tested negative for HPV DNA 4 years after the first follow-up.

FIGURE 2

Study flow chart First visit

Second visit

Our results show that subclinical genital low- and high-risk HPV infection is common in girls without any history of sexual abuse or consensual sexual activity. Among our examined children, we found HPV DNA in 20 of 110 (18%) anogenital samples. Our findings of the prevalence of anogenital HPV DNA in children are consistent with reported HPV detection rates by polymerase chain reaction (PCR) in the literature ranging between 3% and 55%.16-20 Comparison of published studies about HPV prevalence in children is difficult because different techniques of sample collection and HPV detection are reported. With the Hybrid Capture test, we used a detection method that does not amplify DNA and has a low analytic sensitivity in contrast to HPV detection by PCR.14 Therefore, we believe that we probably underestimated the rate of HPV infection in our subjects. On the other hand, testing of HPV DNA by the Hybrid Capture system is more likely to reveal HPV infections rather than detecting contaminations. We did not use a second method for confirming HPV DNA because there was only 1 test commercially available for us in the routine clinical setting. In reviewing the literature, we found different data concerning the HPV infection rate in probable and confirmed sexually abused children. Siegfried et al13 reported positive testing for HPV 16-DNA by PCR in 2 of 40 patients (5%) referred to their clinic for probable or confirmed sexual abuse, whereas Stevens-Simon et al21 found low- and high-risk HPV DNA by PCR in 5 of 31 girls (16%) with con-

Third visit

90 neg. 2 excluded due to condylomas

110 girls included 20 pos.

2 excluded due to sexual activity 8 lost to follow up

C OMMENT

Research

4 excluded due to sexual abuse

5 neg.

8 tested in 1st follow up

1 lost to follow up 1 excluded due to sexual activity

3 pos. 114 girls age 2 - 15 years

1 tested neg. in 2nd follow up

Study flow chart depicts number of visits, depending on the result of HPV testing. neg., HPV negative; pos., HPV positive. Doerfler. Human papilloma virus infection prior to coitarche. Am J Obstet Gynecol 2009.

firmed or suspected sexual abuse. Furthermore, these authors reported that none of the 9 girls in whom sexual abuse could be ruled out tested positive for HPV DNA. Gutman et al22 also found no HPV DNA in their control group and detected HPV DNA by Southern blotting in 5 of 15 severely sexually abused girls (33%). These findings of HPV DNA detection rates in sexually abused children do not match with our results in girls without any evidence of sexual abuse because a lower rate of HPV DNA detection should be expected in nonabused children when sexual transmission is also considered to be the most important route of HPV infection in children. However, the relatively high rate of HPV DNA in our subjects could reflect an underestimation of sexual abuse. Sinclair et al10 found that nearly one third of 7.3% HPV DNA–positive testing girls (124/ 1689 cases) were likely to be sexually abused. In contrast to the studies of Sinclair et al, 10 Stevens-Simon et al,21 and Gutman et al,22 our enrolled children were not referred to our clinic for suspected sexual abuse, and therefore, the perspective of

our service differed significantly from those of the above mentioned authors. This difference in perspectives seems to be important as already mentioned by Sinclair et al themselves.10 We think that detected HPV DNA or even clinically apparent anogenital warts do not necessarily indicate sexual contact or abuse. Obviously our study population consisted of referred patients, which could be source of bias if we want to extrapolate the results of our study to the general population. But this limitation is also true for all other published data on this topic to date, as far as we know. A further limitation of our study is the fact that only approximately half of the HPV-positive patients were available for control examinations so that hypotheses regarding the natural course of symptom-free HPV infection are on a weak basis. Hence, we take other modes of transmission into account in girls without any evidence of sexual abuse. Putative nonsexual routes of transmission like vertical transmission during delivery, horizontal transmission at bathing, and diapering and other indirect and direct modes of HPV acquisition are described in detail by nu-

MAY 2009 American Journal of Obstetrics & Gynecology

487.e3

Research

General Gynecology

www.AJOG.org

TABLE

Type of HPV detected during examinations HPV positive during first visit 4 low-risk HPV

First follow-up (1 y after first visit)

Second follow-up (2 y after first visit)

1 excluded due to coitarche 1 negative 2 lost to follow-up

..............................................................................................................................................................................................................................................

15 high-risk HPV

2 persistent positive for high-risk HPV 1 switch to low-risk HPV 1 excluded due to condylomas 1 excluded due to coitarche 4 negative 6 lost to follow-up

1 excluded due to coitarche 1 lost to follow-up 1 negative

..............................................................................................................................................................................................................................................

1 high-risk and low-risk HPV

1 excluded due to condylomas

..............................................................................................................................................................................................................................................

Total

.....................................................................................................................................................................................................................................

20 positive

11/20 evaluated

2/3 evaluated

..............................................................................................................................................................................................................................................

HPV, human papilloma virus. Doerfler. Human papilloma virus infection prior to coitarche. Am J Obstet Gynecol 2009.

merous authors.11,17,23-28 Because of the sensitivity of the test kit used, we assume that the detected HPV DNA indicates rather an infection and not a contamination, which motivates us to believe that nonsexual infection may play a larger role in young girls than suspected so far. In regard to the malignant potential of persistent HPV infection in adults, it seems to be worthwhile to evaluate HPV persistence in children. We therefore recalled those girls who tested primarily positive for HPV DNA after at least 1 year and detected in 2 of 8 girls the same types of HPV DNA as in the first visit. However, those 2 girls tested negative for HPV DNA during the second follow-up. Because of a disappointing follow-up rate, we were facing difficulties in the study of the natural history of HPV infections in children because studies with minor subjects depend on the willingness of the caretakers to return for further examinations. As a result of our study with a high percentage of spontaneous change from HPV-positive to HPV-negative swabs and because of the costs of HPV testing, we do not perform routine HPV testing anymore in otherwise HPV symptom-free girls. The current recommendation for prophylactic vaccination against HPV is targeted on young girls and boys who 487.e4

are about to start their sexual activity. We, however, found a high rate of HPV infections in young girls prior to sexual activity and hence unknown but apparently nonsexual ways of acquisition of these infections. Therefore, the recommendation to immunize children just before starting consensual sexual activity can be questioned because children with acquisition of HPV infection during childhood without sexual activity could have less benefit from the vaccination, if HPV infection leads to seroconversion or to chronic cervical shedding of HPV DNA. Prevalence data of HPV infection in the general population of young children are needed to decide whether prophylactic vaccination at birth or in early childhood could be more beneficial as discussed by Cason and Mant.14 Our study demonstrates for the first time that subclinical HPV infection is also common in girls prior to coitarche without any history of sexual abuse or consensual sexual activity. Therefore, we recommend being very careful when suspecting sexual abuse only on the basis of positive HPV testing. On the other hand, HPV testing in population-based representative sample of young girls could be crucial to decide whether vaccination against HPV at a younger age is more

American Journal of Obstetrics & Gynecology MAY 2009

beneficial than the current practice. Furthermore, it is also unclear whether there is a different mode of transmission responsible for clinically visible HPV lef sions and subclinical infection. REFERENCES 1. Garland S. Human papillomavirus update with particular focus on cervical disease. Pathology 2002;34:213-24. 2. De Villiers E-M. Laboratory techniques in the investigation of human papillomavirus infections. Genitourin Med 1992;68:50-4. 3. Pascual A, Pariente M, Godinez JM, et al. High prevalence of human papillomavirus 16 in penile carcinoma. Histol Histopathol 2007;22:177-83. 4. Lont AP, Kroon BK, Horenblas S, et al. Presence of high-risk human papillomavirus DNA in penile carcinoma predicts favorable outcome in survival. Int J Cancer 2006;119:1078-81. 5. Shindoh M, Chiba I, Yasuda M, et al. Detection of human papillomavirus DNA sequences in oral squamous cell carcinomas and their relation to p53 and proliferating cell nuclear antigen expression. Cancer 1995;76:1513-21. 6. Syrjänen SM, Syränen KJ, Happonen RP. Human papillomavirus (HPV) DNA sequences on oral precancerous lesions and squamous cell carcinoma demonstrated by in situ hybridization. J Oral Pathol 1988;17:273-8. 7. Stephen KY, Min KW. In situ hybridization analysis of oral papillomas, leukoplakias, and carcinomas for human papillomavirus. Oral Surg Oral Med Oral Pathol 1991;71:726-9. 8. Koutsky L. Epidemiology of genital human papillomavirus infection. Am J Med 1997;102: 3-8. 9. Worda C, Huber A, Hudelist G, et al. Prevalence of cervical and intrauterine human papillomavirus infection in the third trimester in asymptomatic women. J Soc Gynecol Investig 2005;12:440-4. 10. Sinclair KA, Woods CR, Kirse DJ, Sinal SH. Anogenital and respiratory tract human papillomavirus infections among children: age, gender, and potential transmission through sexual abuse. Paediatrics 2005;116:815-25. 11. 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;43:376-81. 12. Frasier LD. Human papillomavirus infection in children. Pediatr Ann 1994;23:354-60. 13. Siegfried E, Rasnick-Conley J, Cook S, Leonardi C, Monteleone J. Human papillomavirus screening in pediatric victims of sexual abuse. Pediatrics 1998;101:43-7. 14. Cason J, Mant CA. High-risk mucosal human papillomavirus infections during infancy and childhood. J Clin Virol 2005;32(Suppl):52-8. 15. Digene Corporation. General information. Available at: http://www.digene.com. Accessed Jan. 24, 2007.

www.AJOG.org 16. Cason J, Rice P, Best JM. Transmission of cervical cancer-associated human papilloma viruses from mother to child. Intervirology 1998;41:213-8. 17. Pakarian FB, Kaye JN, Cason J, et al. Cancer-associated human papillomaviruses: perinatal transmission and persistence. Br J Obstet Gynaecol 1994;101:514-7. 18. Myhre AK, Dalen A, Berntzen K, Bratlid D. Anogenital human papillomavirus in nonabused preschool children. Acta Paediatr 2003;92:1445-52. 19. Powell J, Strauss S, Gray J, Wojnarowska F. Genital carriage of human papilloma virus (HPV) DNA in prepubertal girl with and without vulval disease. Pediatr Dermatol 2003;20:191-4. 20. Tseng CJ, Lin CY, Wang RL, et al. Possible transplacental transmission of human papillomavirus. Am J Obstet Gynecol 1992;166(1 Pt 1):35-40.

General Gynecology 21. Stevens-Simon C, Nelligan D, Bereese P, Jenny C, Douglas JM Jr. The prevalence of genital human papillomavirus infections in abused and nonabused preadolescent girls. Pediatr 2000; 106: 645-9. 22. Gutman LT, St Claire K, Herman-Giddens ME, Johnston WW, Phelps WC. Evaluation of sexually abused and nonabused young girls for intravaginal human papillomavirus infection. Am J Dis Child 1992;142:694-9. 23. Armbruster-Moraes E, Ioshimoto LM, Leao E, Zugaib M. Presence of human papillomavirus DNA in amniotic fluids of pregnant women with cervical lesions. Gynecol Oncol 1994;52:152-8. 24. Sonnex C, Strauss S, Gray JJ. Detection of human papillomavirus DNA on the fingers of patients with genital warts. Sex Transm Infect 1999;75:317-9. 25. Puranen M, Yliskoski M, Saarikoski S, Syrjanen K, Syrjanen S. Vertical transmission of hu-

Research

man papillomavirus from infected mothers to their newborn babies and persistence of the virus in childhood. Am J Obstet Gynecol 1996;174:694-9. 26. Fredericks BD, Balkin A, Daniel HW, Schonrock J, Ward B, Fraxer IH. Transmission of human papillomavirus from mother to child. Aust N Z J Obstet Gynaecol 1993;33: 30-2. 27. Smith EM, Ritchie JM, Yankowitz J, et al. Human papillomavirus prevalence and types in newborns and parents. Concordance and modes of transmission. Sex Trans Diseases 2004;31:57-62. 28. Syrjänen S, Puranen M. Human papillomavirus infections in children: the potential role of maternal transmission. Crit Rev Oral Biol Med 2000;11:259-74.

MAY 2009 American Journal of Obstetrics & Gynecology

487.e5