Is early age at menarche a risk factor for endometriosis? A systematic review and meta-analysis of case-control studies.

OBJECTIVE: To review published studies evaluating early menarche and the risk of endometriosis. DESIGN: Systematic review and meta-analysis of case-control studies. SETTING: None. PATIENT(S): Eighteen case-control studies of age at menarche and risk of endometriosis including 3,805 women with endometriosis and 9,526 controls. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Medline and Embase databases were searched from 1980 to 2011 to locate relevant studies. Results of primary studies were expressed as effect sizes of the difference in mean age at menarche of women with and without endometriosis. Effect sizes were used in random effects meta-analysis. RESULT(S): Eighteen of 45 articles retrieved met the inclusion criteria. The pooled effect size in meta-analysis was 0.10 (95% confidence interval -0.01-0.21), and not significantly different from zero (no effect). Results were influenced by substantial heterogeneity between studies (I(2) = 72.5%), which was eliminated by restricting meta-analysis to studies with more rigorous control of confounders; this increased the pooled effect size to 0.15 (95% confidence interval 0.08-0.22), which was significantly different from zero. This represents a probability of 55% that a woman with endometriosis had earlier menarche than one without endometriosis if both were randomly chosen from a population. CONCLUSION(S): There is a small increased risk of endometriosis with early menarche. The potential for disease misclassification in primary studies suggests that this risk could be higher