Herpes Simplex Virus Type 2 Status at Age 26 Is Not Related to Early Circumcision in a Birth Cohort

Objective: The objective of this study was to determine if circumcision in early childhood affects the risk of acquiring herpes simplex virus type 2 (HSV-2) infection.

Study: Study members were born in 1972-1973 in Dunedin, New Zealand. Circumcision status was sought at age 3, when the cohort was established. Information about sexual behavior was obtained at ages 21 and 26. Serum was tested for HSV-2 antibodies at age 26 for 435 men (82.9% of the surviving cohort).

Results: Of eligible men, 40.2% had been circumcised. The prevalence of HSV-2 antibodies was 7.3% in uncircumcised men and 7.4% in circumcised men. Social and sexual factors were very similar between the 2 groups and adjustment had no effect on the association (odds ratio, 1.1; 95% confidence interval, 0.46-2.5). Seroconversion rates according to years since first sexual intercourse were 0.85 and 0.86 per 100 person-years for uncircumcised and circumcised men.

Conclusion: The results support a lack of association between circumcision status and HSV-2 acquisition, although a small effect cannot be ruled out.

EVIDENCE IS MOUNTING THAT UNCIRCUMCISED men are at greater risk of acquiring HIV infection heterosexually.¹ Similarly, new research suggests that the types of human papilloma virus causative for cervical cancer may be associated with uncircumcised male partners.² Such a link between circumcision and cervical cancer was widely discussed in the 1950s but later dismissed.³ The evidence that circumcision reduces sexual transmission of other viral infections is much less clear. It is important to know what affects transmission of herpes simplex virus type 2 (HSV-2) because, as well as being the main cause of symptomatic genital herpes, HSV-2 may facilitate transmission of HIV.⁴

Clinic-based studies have investigated this issue with mixed results,^5-9 and 2 population-based surveys of self-reported genital herpes found no association with circumcision.^10,11 Only 1 previous study has determined HSV-2 incidence serologically, although this was in a clinic population in India.¹ This also found no association with circumcision. We have investigated the relationship between HSV-2 incidence and prevalence and circumcision in a birth cohort in New Zealand using detailed sexual and social histories and serologic measurement at age 26.

Method

The sample was enrolled in the Dunedin Multidisciplinary Health and Development Study, a longitudinal study of a cohort born in Dunedin, New Zealand, between April 1, 1972, and March 31, 1973. The children were first followed up at 3 years of age when 1037 were seen, forming the base sample for the longitudinal study; the mothers were asked then whether their sons had been circumcised. Subsequently, study members were seen every 2 years until 15 years, then at 18, 21, and 26 years.

Questions on sexual behavior, based on those used in the 1990 British National Survey of Sexual Attitudes and Lifestyles, were presented by computer at ages 21 and 26.¹² Age of first intercourse was asked at age 21. Socioeconomic status of the study member's family was based on parental occupation over the first 15 years of life using the Elley-Irving scale. Socioeconomic status of the study members themselves was based on their own current or most recent occupation at age 26 classified using the New Zealand Socio-Economic Index, grouped into 3 categories.

All participants were invited to provide serum samples at the age 21 and 26 assessments, which were then tested for serologic evidence of HSV-2 infection. Sera were tested using an indirect IgG enzyme-linked immunoassay specific to the HSV-2 glycoprotein G (gG-2) and those positive on EIA confirmed by Western blot.

The analysis was restricted to those with serum and circumcision data who reported heterosexual intercourse. Potential confounders were identified from previous analyses of predictors of HSV-2 seroconversion by age 21 and 26 in this cohort.¹³ Chi-squared tests for independence were calculated to determine the relations between circumcision status and measures of socioeconomic status and sexual behavior. A univariate logistic regression model, with 95% confidence intervals (CIs), was used to examine the relationship between circumcision and prevalence of HSV-2. Backward stepwise likelihood ratio logistic regression, which forced circumcision to be entered at each step, and entered those variables examined in the chi-squared analysis, was performed. Seroconversion rates by age 26 were calculated, adjusting for years since first intercourse. When age at first intercourse was not available, exposure was assumed to be 3 years (if first sexually active between age 21 and 26) or 9 (for those active by age 21). The P value was determined using the comparison of incidence rate (large-sample test).¹⁴

Results

Of the original 535 male cohort members seen at age 3, 525 were believed to be still alive at age 26. Of these, 435 (82.9% of the surviving cohort) had serum tested for antibodies to HSV-2 at age 26, had information provided at age 3 on their circumcision status, and reported heterosexual intercourse. The 2 men who reported only sex with other men were excluded.

Of the 435 men, 175 (40.2%) had been circumcised by age 3. There were no significant differences between the circumcised and uncircumcised according to socioeconomic characteristics and sexual behavior (Table 1).

Table 1

Overall, 32 (7.4%) of the men had antibodies to HSV-2 at age 26. The uncircumcised men had a prevalence of 7.3% and circumcised men of 7.4% (odds ratio [CI], 0.98; 95% CI, 0.47-2.0). Adjusting for the characteristics in Table 1 did not affect the relationship (OR, 1.1; 95% CI, 0.46-2.5). The seroconversion rates were 0.85 and 0.86 per 100 person-years since first sexual intercourse for the uncircumcised and circumcised men, respectively (P = 0.89).

Discussion

Early circumcision does not have a strong protective effect on the acquisition of HSV-2 infection by age 26. Remarkably, the social and sexual characteristics of the circumcised and uncircumcised men were almost identical. Hence, unsurprisingly, adjustment for these factors had no influence on the measured association.

Strengths of this study are that it was prospective and population-based with a very high retention rate, HSV-2 infection was determined serologically, and detailed sexual behavior was reported. The use of well-validated questions on sexual behavior and the use of a computer (with safeguards to protect confidentiality) should have enhanced disclosure. Parental reports of early circumcision at age 3 (which we expect to be reliable) were used to assess circumcision status. Information on later circumcision was not sought; however, this is uncommon in comparison to infant or early childhood circumcision. Because the confidence interval was relatively wide, we cannot rule out a small effect of circumcision on HSV-2 acquisition.

The findings are consistent with the more recent population-based studies in showing no protective effect of circumcision on acquisition of HSV-2. This suggests that cross-sectional clinic-based studies may be subject to bias and hence are unreliable. The study does not address whether HSV-2 is acquired more easily by women from an infected uncircumcised man. Finally, there is likely to be a biologic reason why uncircumcised men acquire HIV, but not HSV-2 infection, more easily. This may be because of differences in mode of entry into the genital tract.¹