Quality assessment
An overview concerning risk of bias ascertained for each article is listed in the appendix [S3 ]. Overall, biases were mostly caused by study design (not including a healthy control group) and patient selection as most studies included women in tertiary referral centres. Most case series lacked describing a hypothesis or did not discuss the biological plausibility of the hypothesis in the light of reports’ observations [26]. Cohort studies and case control studies, as assessed by the NOS scale [25], most frequently used hospital controls instead of community controls, and missing data was poorly reported.