The more consistently women's male sex partners use condoms, the less likely women are to acquire genital human papillomavirus (HPV) infection, according to a longitudinal study among newly sexually active young women.¹ Compared with their counterparts whose male partners used condoms less than 5% of the time for vaginal intercourse, women whose partners used them at least half the time had a 50% lower risk of infection and women whose partners used them every time had a 70% lower risk. The pattern was similar in analyses restricted to types of HPV associated with a low risk of cervical cancer as well as those associated with a high risk.

Researchers invited female university students aged 18–22 years who were sexually inexperienced or newly sexually active to participate in the study. Every four months, the women underwent gynecologic examinations during which cervical and vulvovaginal samples were collected for HPV testing (by a polymerase chain reaction assay that detects 37 types of the virus) and for Pap testing. The women also recorded information about their daily sexual behavior in a Web-based diary every two weeks. The time that women were considered to be at risk for infection began on the date of first intercourse, and factors potentially affecting the risk of HPV infection—the number of instances of vaginal intercourse, the number of new partners, the frequency of use of condoms by male partners, each partner's circumcision status and each partner's number of previous partners—were summarized for the eight-month period before each HPV test. Factors showing a statistical association at p<.10 in univariate analyses were included in multivariate analyses.

Study results were based on 82 women who reported their first sexual intercourse with a male partner during the study or the two weeks before enrollment. On average, the women were about 19 years old and were followed for 34 months. The median number of instances of sexual intercourse reported was 48 per year, and the median number of new partners reported was one per year.

A total of 40 women experienced 126type-specific HPV infections after first sexual intercourse, corresponding to a 37% cumulative incidence of a first HPV infection over a 12-month period. For every 100 woman-years at risk, there were roughly 38 infections when condoms were used by male partners for 100% of instances of vaginal intercourse in the preceding eight months, 62 infections when condoms were used 50–99% of the time, 160 when condoms were used 5–49% of the time and 89 when condoms were used less than 5% of the time.

In a multivariate analysis, women's likelihood of acquiring an HPV infection decreased significantly as the frequency of condom use increased. Relative to women whose partners used condoms for 5% or fewer instances of intercourse, women whose partners used them 50–99% of the time had a 50% lower risk of infection (hazard ratio, 0.5) and women whose partners used them 100% of the time had a 70% lower risk (0.3). The results were similar when analyses were restricted to infections with high-risk types of HPV, low-risk types of HPV or the four types covered by the HPV vaccine. Moreover, among women whose partners used condoms all of the time, the decrease in the risk of HPV infection did not vary by whether or not the women also had unprotected, nonpenetrative genital contact with their partners.

In addition, women whose partners had not had any previous partners had a markedly lower risk of becoming infected with HPV relative to their counterparts whose partners had had at least one or an unknown number of previous partners (hazard ratio, 0.0). Women who had one new sexual partner and those who had more than one had a sharp increase in risk relative to their counterparts who did not have any new partners (4.8 and 6.9, respectively). Neither the number of instances of vaginal intercourse nor a partner's circumcision status significantly affected the likelihood of acquiring an HPV infection.

A total of 15 women developed precancerous lesions of their cervix after first intercourse, corresponding to a 15% cumulative incidence of these lesions over a period of 24 months. For every 100 woman-years at risk, there were no lesions when condoms were used by male partners for 100% of instances of vaginal intercourse in the preceding eight months, 17 when condoms were used 50–99% of the time, 16 when condoms were used 5–49% of the time and 11 when condoms were used less than 5% of the time.

In a multivariate analysis, the frequency of condom use did not significantly influence women's risk of developing cervical lesions. However, compared with women who did not have any new sex partners, women who had one or more than one new partner had a sharply elevated risk (hazard ratios, 6.5 and 23.3, respectively).

Use of male condoms appears to reduce the risk of HPV transmission from men to women, the researchers conclude; however, they note that the study's findings may not apply to older women or to women of lower socioeconomic status (proxied by lack of college education). Some HPV infections are to be expected despite consistent condom use because the virus can be spread by nonpenetrative genital contact and condoms are not always used correctly, they point out; nonetheless, the benefit observed is "encouraging" because the women studied were new to both intercourse and condom use.

Given the reductions in risk achieved across broad categories of the virus, the researchers assert that even though the HPV vaccine is known to be effective against the four types of the virus that put women at highest risk for cervical cancer, "consistent condom use by their partners may protect women against infection with other high-risk types of HPV…."—S. London

REFERENCE

1. Winer RL et al., Condom use and the risk of genital human papillomavirus infection in young women, New England Journal of Medicine, 2006, 354(25):2645–2654.