VAN DEN EEDEN, STEPHEN K. PhD*,†; HABEL, LAUREL A. MPH*,†; SHERMAN, KAREN J. PhD*; MCKNIGHT, BARBARA PhD*,‡; STERGACHIS, ANDY PhD†,§; DALING, JANET R. PhD*,†
BETWEEN 1966 and 1981, physician office visits for condyloma acuminata increased by over 450%.1 This increase has been attributed both to increased awareness by the public and by health practitioners and to an increase in the prevalence of the disease.2 This increase is important, in part, because human papillomavirus (HPV) has been associated with the occurrence of anogenital and upper respiratory cancers.3–9 However, the HPV types implicated in these cancers are typically different from those characteristically found in condyloma.2 It remains unclear if anogenital cancers and condyloma have a set of risk factors in common.
Most epidemiologic studies conducted to data have been cross‐sectional in design and have used highly selected populations such as university students,10 sexually transmitted disease (STD) clinic patients,11,12 pregnant women,13 and family planning clinic patients.14,15 The few controlled studies focused primarily on women16; little is known about the risk factors for condyloma among men.17 This stands in contrast to the epidemiologic characterization of other STDs such as gonorrhea or genital herpes.18–20
To evaluate what factors may predispose a man to the occurrence of condyloma, we conducted a case‐control study of condylomata acuminatum among members of a health maintenance organization. Our focus was on sexual history and those factors that may be shared with HPV‐related anogenital cancers.
The methods for this study were published before.21 A summary of the methods is provided below.
Subjects for this study were selected from enrollees of the Group Health Cooperative of Puget Sound (GHC) who received their primary care at one of four clinics. Group Health Cooperative is a health maintenance organization serving residents of western Washington State. The GHC population is generally representative of the geographic population, although the extremes of the income scale are likely to be underrepresented. The clinics included in this study cover approximately one third of GHC's enrollees. Both men and women were studied; see Habel et al22 for the results for women.
Patients were men, 18 years or older, with a incident or recurrent diagnosis of external genital warts at one of the four clinics between April 1, 1987 and September 30, 1991. Patients were identified by a manual review of all treatment record forms at the four clinics of GHC for the study period for a diagnosis of genital or venereal warts or condyloma. These forms are completed for each outpatient visit and include reason for the visit. Incident or primary patients were those seeking care who had not received prior medical care for condyloma. Recurrent patients were defined as men who had received prior care for condyloma. Potential control subjects were randomly selected from the GHC membership file and frequency matched to patients by age (5‐year groups) and clinic site.
Permission to contact the patient was sought from the primary care physician of patients. Patients, where permission was obtained, and control subjects were sent a letter introducing the study and inviting participation. Eligible subjects could indicate their unwillingness to participate by returning a postcard. If no card was received within 2 weeks, the potential subject was called by a member of the study staff. Interested subjects were contacted by a male interviewer, and an interview was scheduled at a time and location convenient to the subject. The median time between the visit and interview was 8 months.
Of the 307 men identified with condyloma, 179 (58.3%) were successfully interviewed. Patients were not interviewed for the following reasons: 11 (3.6%) physician refusal, 78 (25.7%) subject refusal, and 39 (12.7%) unable to locate. Two hundred ninety‐two control subjects were identified and invited to participate. Of these, 188 (64.4%) were successfully interviewed; for 15 (5.1%) the physician refused, 67 (22.9%) refused interview, and 22 (7.5%) could not be located.
Interview responses used in the analysis of this data refer to those occurring before a reference date. For patients, the reference date was the date of the physician diagnosis of genital warts. For control subjects, the reference date of the patient for whom they were selected was used.
After obtaining informed consent, interviews were conducted at a place of convenience to the subject, most often the subject's home. The interviewer used a structured questionnaire to collect information pertaining to the subject's demographic and lifestyle characteristics and medical, diet, sexual, and contraceptive histories. For condyloma, subjects were asked if they ever had genital warts. If the subject responded yes, information was sought on the number of episodes of genital warts, location(s), and treatment(s). For a small number of patients, we were able to obtain a small sample of tissue for viral typing (no data available).
The odds ratio (OR) was used to approximate the relative risk associated with exposure using multiple logistic regression models fit by maximum likelihood.16 Ninety‐five percent confidence intervals (CIs) were calculated by using the normal approximation and the standard errors. In these analyses, categoric variables were used for age (<20, 20‐29, 30‐39, 40+), race (white, black, other), household income (<$15,000, 15‐30,000, >30,000‐45,000, 45,000+), age at first regular intercourse (≤17, 18‐19, 20+), number of sexual partners (0, 1, 2‐4, 5‐14, 15‐29, 30+), and frequency of condom or female barrier contraceptive methods (rarely [<5% of the time], sometimes [5%‐49%], usually [50%‐89%], nearly always [≥90%]). Dichotomous variables were used for education (≤12 years, >12 years), ever having had an STD, being circumcised, ever having had one of a selected group of infections, and recreational drug use. In cases with few individuals reporting exposure (e.g., exposure syphilis, hepatitis, or giardia), exact methods were used to compute the OR and appropriate 95% confidence intervals.
Only nine control subjects and one recurrent patient reported not having any sexual partners. In addition, three control subjects, four recurrent patients, and five incident patients reported not being exclusively heterosexual. Because few subjects reported no sexual partners or not being exclusively heterosexual, we restricted the analysis to men who reported at least one sexual partner and being exclusively heterosexual.
Patients were categorized as recurrent if they reported more than one episode of condyloma and incident if the index visit to the clinic was the first medical attention sought for condyloma. Approximately one third of the men with condyloma were recurrent patients (n = 67) with the remainder being incident patients (n = 109). Men with recurrent disease were older on average than men with an incident condyloma, whereas control subjects were slightly younger than either of the condyloma groups (Table 1). There were no important differences in the distribution of education and income between the groups. Similarly, race was unrelated to being a patient or control subject.
As noted in Methods, few men reported not having sex or not being exclusively heterosexual. Only nine (5.1%) of the patients and three (1.8%) of the control subjects reported not having a primarily or exclusively heterosexual preference (Table 2). Compared with reporting exclusive heterosexuality, the risk of reporting not being exclusively heterosexual was 3.2 (95% CI, 0.7‐15.0) for recurrent patients and 2.5 (95% CI, 0.6‐11.3) for incident patients. Men reporting not being primarily or exclusively heterosexual and the five control subjects who reported not having had sexual intercourse (all patients reported sexual activity) are not included in the remainder of the results presented.
The relative risks associated with sexual activity or factors are presented in Table 2. The risk of incident condyloma was elevated for men reporting first regular sex after age 20 (OR, 2.6; 95% CI, 1.2‐5.6) compared with initiating regular sex before age 17. However, few subjects reported initiating sex after age 20. Compared with control subjects, men with recurrent or incident condyloma more often reported never having been married or being divorced, separated, widowed, or living with someone. Having more lifetime sexual partners was associated with an elevated risk of both recurrent and incident condyloma. Indeed, none of the recurrent patients reported having only one lifetime partner. The number of female partners in the last 5‐year period was also associated with an increased risk of recurrent and incident condyloma. However, no clear pattern of increased risk with increasing number of sexual partners was evident in these data. Men who reported using condoms sometimes (5%‐49% of the time) had an increased risk of incident disease. Although the use of condoms was associated with a lower risk of recurrent disease, the confidence limits on these estimates are broad. Reporting being circumcised was not associated with recurrent or incident condyloma.
A history of genital infections was strongly related to recurrent condyloma, whereas there was little, if any, increased risk of incident condyloma associated with specific infections (Table 3). A modest increased risk of incident condyloma was associated with a history of any STD. No increase in risk was observed for reporting a partner with condyloma in the last 5 years with either recurrent or incident condyloma.
Table 4 presents the relative risk estimates associated with tobacco, alcohol, and recreational drug use. Men with incident condyloma reported being a former or current smoker about as often as control subjects (Table 2). Recurrent patients reported being former smokers less often than control subjects; however, few of these patients or control subjects were former smokers. There was no clear trend in risk of recurrent or incident condyloma associated with alcohol use reported in the last 5‐year period. Use of either marijuana and cocaine was associated with an increased risk of condyloma for recurrent disease, whereas only slight increases were observed for the incident patients.
The relative risk estimates associated with factors or conditions that may reflect compromised immune status are presented in Table 5. Hepatitis was associated with both incident and recurrent disease, although few of the incident patients reported hepatitis. A history of common warts was associated with recurrent, but not incident, condyloma. No association was found between the other factors and condyloma.
We believe the following issues should be borne in mind when interpreting the results of this study. First, the low response rate introduces uncertainty in our findings in that responders may be different from nonresponders with respect to exposures of interest. Of particular concern is that bias may be introduced if this relative difference in responders and nonresponders is differential by case‐control status. Indeed, the lower rates of control subject participation in the younger age groups suggests that this situation may have occurred. Second, the patients were limited to subjects who brought their lesions to medical attention. Men who bring their lesions to medical attention may differ from those that choose not to do so. Thus, these results apply only to this subset of patients. Third, we were unable to determine accurately the true onset of the disease in our subjects. We are concerned that subjects who have the disease for a long period of time before seeking medical attention may alter their behavior with respect to some of the exposures of interest, for example, condom use. This would likely reduce our risk estimate. Fourth, our data represent clinical expression of disease. We were able to perform viral phenotyping only on a small number of patients and found 89% (46 of 52) to have HPV types 6 and/or 11 or type 16. Occult disease may be present in our control subjects (and may have occurred in patients before the current episode). Furthermore, the patients may be incorrectly classified as incident or recurrent condyloma. This potential misclassification of disease status could bias our results. Fifth, for recurrent patients, we did not have any way to determine if the lesion was a truly new episode of disease or the reactivation of an existing lesion. Thus, we may be misclassifying some patients with too many episodes of disease.
Although some epidemiologic characteristics of condyloma, such as incidence and prevalence by age and sex, have been reported,23 few, if any, controlled studies have examined the factors that predispose a man (or woman) to developing condylomata acuminata. Although transmission through sexual conduct was established some time ago,13 the few studies that have been conducted to describe how sexual and other factors are related to condyloma have been uncontrolled or conducted among highly selected populations. The one population‐based study was a review of all condyloma acuminatum cases in the Rochester, Minnesota area from 1950 to 1978.23 We are unaware of other epidemiologic studies of condyloma in men similar to the present study.
These data provide evidence about the relative occurrence of recurrent and initial episodes of condyloma among men. Before initiating this study, the patient group was expected to be primarily composed of men seeking treatment for their first episodes of condyloma. However, roughly one third of the men being seen for condyloma were for recurrent disease. It is unknown how the proportion of recurrent to initial events in this study compares with other unselected populations because, to our knowledge, this is the first study to describe this feature among men from an unselected general population. Although we could not estimate the risk of condyloma associated with this factor because age was a matching factor, the age distribution of our study population is similar to other published series.13,23
Our finding of increased risk associated with having multiple sex partners is in agreement with the established mode of transmission for this disease. Having many lifetime partners or many partners in the most recent period is a potent predictor of both recurrent and incident condyloma. The other factors associated with elevated risks may well be markers of sexual behavior, for example, behavioral patterns that are reflected in the choice of sexual partners. Other studies reported similar findings. Bergman and Nalick11 reported that men whose sexual partners in the last 6 months were infected were significantly more likely to be infected than men whose partners had no evidence of condyloma. Cross‐sectional studies of military conscripts24 and young males in detention25 found men with multiple partners and infrequent condom use to be associated with risk of STDs or condyloma.
Whether barrier contraceptive methods, including condom and female methods, offer protection against infection with HPV is not clear. We found no clear pattern of risk associated with condom or female barrier contraception. Although cross‐sectional studies found infrequent condom use to be associated with infection,24 these studies did not control for other factors, such as number of sexual partners. Our data do not support the protective advantage of using barrier contraceptive methods, although the complex events or circumstances associated with using barrier methods were not fully assessed.
We found no relationship between smoking and condyloma after adjusting for sexual history factors. This is in contrast to other STDs3,7,26,27 and the risk of anogenital cancers that have been linked to HPV.28 Recurrent, but not incident, condyloma was associated with a history of common warts. This is in contrast to the results with women.22 Other studies found no association between common or plantar warts and the occurrence of condyloma. This is likely due to differences in the behavior and preferred host tissue of the HPV types found in condyloma and plantar warts.
Several reports, using cross‐sectional data from STD clinic patients, suggested that the lack of circumcision was associated with a decreased risk of genital warts.12,13 Low‐grade trauma on the penis has been hypothesized to facilitate HPV infection or manifestation. The presence of the foreskin would impart some protection from trauma to the glans and upper penile shaft, the most frequent anatomical sites of condyloma in the male.13,23 However, our data show that circumcision had little if any effect on the risk of recurrent or incident condyloma. Our results may differ from the other studies in that ours was a less highly selected population and we were able to control for a variety of other confounders or cofactors. However, this is an important area of further research.
There were several differences in the risk associated with various factors and incident and recurrent condyloma. At least some of these differences are probably due to factors that are not directly associated with incident or recurrent disease but represent either historical or behavioral patterns. Drug use and a history of STDs were associated with recurrent, but not incident, condyloma. We suspect that these factors represent behavior patterns and choices toward sexual activity and choice of partners rather than direct relationships. The differences in recreational drug use probably represent risk behavior over and above the sexual history and behavior ascertained in this study. The difference for STDs may be due to other factors. It may be that men who are infected with an STD may be more likely to be exposed to partners with an STD. However, our data do not support this contention‐we found no apparent association with reporting a partner with genital warts in the last 5 years. Alternatively, because the recurrent patients were more likely to have multiple sex partners, the ability to know the genital wart status of their partners may be diminished. It has also been hypothesized that STDs that inflame or affect the columnar epithelium may prime the tissue for HPV infection.2
Sexual risk factors found in this study are not in total concordance with those found in studies of male anogenital cancers where HPV is thought to play an etiologic role.29 One case control study of penile cancer reported few sexual behavior factors, such as age at intercourse, number of sexual partners, and number of marriages, were not associated with penile cancer.26 However, 15 or more sexual partners was associated with penile cancer in a second case‐control study.30 A third case‐control study reported inconsistent associations with penile cancer.31 Case‐control studies found number of sex partners to be consistently associated with increased risk of male anal cancer.3,27
In contrast to the sexual factors, tobacco use, either measured as current smoking or heavy use, was associated with penile and anal cancer in virtually all of these studies.3,7,27,31,32 The present study found no such association, although there is some suggestion of increased risk for recurrent disease and for condyloma among women.22 These dissimilarities likely represents either differences in the natural history of these diseases or in what they represent in the different types of studies. Smoking may be a cofactor with HPV infection for neoplastic development but may not influence the risk of HPV infection itself.
In summary, a large proportion of our patients had recurrent disease. Having a history of multiple sex partners and prior STD were associated with condyloma in men. However, several differences in the risk profile of recurrent and incident condyloma were observed and likely represent behavioral patterns rather than direct relationships. The factors found to be associated with condyloma in this study are similar, but not the same, for anogenital cancer believed to be associated to HPV infection.
1. Centers for Disease Control. Condyloma acuminatum-United States, 1966-1981. MMWR 1983; 23:306-308.
2. Koutsky LA, Galloway DA, Holmes KK. Epidemiology of genital human papillomavirus infection. Epidemiol Rev 1988; 10:122-163.
3. Daling JR, Weiss NS, Hislop TG, et al. Sexual practices, sexually transmitted diseases, and the incidence of anal cancer. N Engl J Med 1987; 317:973-977.
4. Brinton LA, Nasca PC, Mallin K, et al. Case-control study of in situ and invasive carcinoma of the vagina. Gynecol Oncol 1990; 38:49-54.
5. Brinton LA, Nasca PC, Mallin K, Baptiste MS, Wilbanks GD, Richart RM. Case-control study of cancer of the vulva. Obstet Gynecol 1990; 75:859-866.
6. Maden C, Beckmann AM, Thomas DB, et al. Human papillomaviruses, herpes simplex viruses, and the risk of oral cancer in men. Am J Epidemiol 1992; 135:1093-1102.
7. Maden C, Sherman KJ, Beckmann AM, et al. History of circumcision, medical conditions, and sexual activity and risk of penile cancer. J Natl Cancer Inst 1993; 85:19-24.
8. Sherman KJ, Daling JR, Chu J, Weiss NS, Ashley RL, Corey L. Genital warts, other sexually transmitted diseases, and vulvar cancer. Epidemiology 1991; 2:257-262.
9. Kjaer SK, Dahl C, Engholm G, Bock JE, Lynge E, Jensen OM. Casecontrol study of risk factors for cervical neoplasia in Denmark. II. Role of sexual activity, reproductive factors, and venereal infections. Cancer Causes Control 1992; 3:339-348.
10. Wheeler CM, Parmenter CA, Hunt WC, et al. Determinants of genital human papillomavirus infection among cytologically normal women attending the University of New Mexico student health center. Sex Transm Dis 1993; 20:286-289.
11. Bergman A, Nalick R. Prevalence of human papillomavirus infection in men. Comparison of the partners of infected and uninfected women. J Reprod Med 1992; 37:710-712.
12. Cook LS, Koutsky LA, Holmes KK. Circumcision and sexually transmitted diseases. Am J Public Health 1994; 84:197-201.
13. Oriel JD. Natural history of genital warts. Br J Vener Dis 1971; 47:1-13.
14. Bauer HM, Hildesheim A, Schiffman MH, et al. Determinants of genital human papillomavirus infection in low-risk women in Portland, Oregon. Sex Transm Dis 1993; 20:274-278.
15. Oh MK, Feinstein RA, Pass RF. Sexually transmitted diseases and sexual behavior in urban adolescent females attending a family planning clinic. J Adolesc Health Care 1988; 9:67-71.
16. Daling JR, Sherman KJ, Weiss NS. Risk factors for condyloma acuminatum in women. Sex Transm Dis 1986; 13:16-18.
17. Oriel JD. Identification of people at high risk of genital HPV infections. Scand J Infect Dis Suppl 1990; 69:169-172.
18. Barnes RC, Holmes KK. Epidemiology of gonorrhea: current perspectives. Epidemiol Rev 1984; 6:1-30.
19. Larsson PG, Platz-Christensen JJ, Sundstrom E. Is bacterial vaginosis a sexually transmitted disease? Int J STD AIDS 1991; 2:362-364.
20. Pessione F, Dolivo M, Casin I, et al. Sexual behavior and smoking: risk factor for urethritis in men. Sex Transm Dis 1988; 15:119-122.
21. Bairati I, Sherman KJ, McKnight B, et al. Diet and genital warts: a case-control study. Sex Transm Dis 1994; 21:149-154.
22. Habel LA, Van Den Eeden SK, Sherman KJ, McKnight B, Stergachis A, Daling JR. Risk factors for incident and recurrent condyloma acuminata among women: a population-based study. Sex Transm Dis 1998;25:285-292.
23. Chuang TY, Perry HO, Kurland LT, Ilstrup DM. Condyloma acuminatum in Rochester, Minn., 1950-1978. I. Epidemiology and clinical features. Arch Dermatol 1984; 120:469-475.
24. Hippelainen M, Syrjanen S, Koskela H, Pulkkinen J, Saarikoski S, Syrjanen K. Prevalence and risk factors of genital human papillomavirus (HPV) infections in healthy males: a study on Finnish conscripts. Sex Transm Dis 1993; 20:321-328.
25. Shafer MA, Hilton JF, Ekstrand M, et al. Relationship between drug use and sexual behaviors and the occurrence of sexually transmitted diseases among high-risk male youth. Sex Transm Dis 1993; 20:307-313.
26. Brinton LA, Li JY, Rong SD, et al. Risk factors for penile cancer: results from a case-control study in China. Int J Cancer 1991; 47:504-509.
27. Holly EA, Whittemore AS, Aston DA, Ahn DK, Nickoloff BJ, Kristiansen JJ. Anal cancer incidence: genital warts, anal fissure or fistula, hemorrhoids, and smoking. J Natl Cancer Inst 1989; 81:1726-1731.
28. zur Hausen H. The role of papillomaviruses in anogenital cancer. Scand J Infect Dis Suppl 1990; 69:107-111.
29. Daling JR, Sherman KJ. Relationship between human papillomavirus infection and tumours of anogenital sites other than the cervix. In: Munoz N, Bosch FX, Shah KV, Meheus A, eds. The Epidemiology of Cervical Cancer and Human Papillomavirus. Lyon: International Agency for Research on Cancer, 1992:223-241.
30. Daling JR, Sherman KJ, Hislop TG, et al. Cigarette smoking and the risk of anogenital cancer. Am J Epidemiol 1992; 135:180-189.
31. Hellberg D, Valentin J, Eklund T, Nilsson S. Penile cancer: is there an epidemiological role for smoking and sexual behaviour? Br Med J (Clin Res Ed) 1987; 295:1306-1308.
32. Holmes F, Borek D, Owen-Kummer M, et al. Anal cancer in women. Gastroenterology 1988; 95:107-111.