Robert S Van Howe1. 1. Department of Pediatrics and Human Development, Michigan State University College of Human Medicine, East Lansing, MI, United States.
Abstract
INTRODUCTION: Attempts to find an association between male circumcision and prostate cancer risk have produced inconsistent results. METHODS: Age-standardized prostate cancer incidence, life-expectancy, geographical region, and circumcision prevalence from 188 countries were compared using linear regression analysis. Following a systematic literature review, a meta-analysis was performed on studies meeting inclusion criteria with evaluations of between-study heterogeneity and publication bias. A cost analysis (discounted at 3% and 5% per annum) was performed using the meta-analysis's summary effect and upper confidence interval. RESULTS: Univariate analysis revealed a trend for a positive association between country-level age-standardized prostate cancer incidence (per 100 000 person-years) and circumcision prevalence (β=0.0887; 95% confidence interval [CI)]-0.0560, +0.233), while multivariate analysis found a significant positive association (β=0.215; 95% CI 0.114, 0.316). Twelve studies were included in meta-analysis. The random-effects summary odds ratio of the risk of being genitally intact was 1.10 (95% CI 0.96, 1.26, between-study heterogeneity χ2 15=27.43; p=0.03; I2=82.8%). There was no evidence of publication bias. Cost analysis found infant circumcision was prohibitively costly, returning only between 1.6¢ and 13.8¢ for each dollar expended. CONCLUSIONS: Circumcision may be a positive risk factor on geographical analysis, but not in case-case-controlled studies. Circumcision is not economically feasible for preventing prostate cancer.
INTRODUCTION: Attempts to find an association between male circumcision and prostate cancer risk have produced inconsistent results. METHODS: Age-standardized prostate cancer incidence, life-expectancy, geographical region, and circumcision prevalence from 188 countries were compared using linear regression analysis. Following a systematic literature review, a meta-analysis was performed on studies meeting inclusion criteria with evaluations of between-study heterogeneity and publication bias. A cost analysis (discounted at 3% and 5% per annum) was performed using the meta-analysis's summary effect and upper confidence interval. RESULTS: Univariate analysis revealed a trend for a positive association between country-level age-standardized prostate cancer incidence (per 100 000 person-years) and circumcision prevalence (β=0.0887; 95% confidence interval [CI)]-0.0560, +0.233), while multivariate analysis found a significant positive association (β=0.215; 95% CI 0.114, 0.316). Twelve studies were included in meta-analysis. The random-effects summary odds ratio of the risk of being genitally intact was 1.10 (95% CI 0.96, 1.26, between-study heterogeneity χ2 15=27.43; p=0.03; I2=82.8%). There was no evidence of publication bias. Cost analysis found infant circumcision was prohibitively costly, returning only between 1.6¢ and 13.8¢ for each dollar expended. CONCLUSIONS: Circumcision may be a positive risk factor on geographical analysis, but not in case-case-controlled studies. Circumcision is not economically feasible for preventing prostate cancer.