Abstract
BACKGROUND: The relationship between COVID-19 vaccines and menstrual disturbance is unclear, in part because researchers have measured different outcomes (e.g., delays vs. changes to cycle length) with various study designs. Menstrual disruption could be a decisive factor in people's willingness to accept the COVID-19 vaccine. METHODS: We searched Medline, Embase, and Web of Science for studies investigating menstrual cycle length, flow volume, post-menopausal bleeding, and unexpected or intermenstrual bleeding. Data were analyzed using fixed-effects meta-analysis with Shore's adjusted confidence intervals for heterogeneity. FINDINGS: Seventeen studies with >1·9 million participants were analyzed. We found a 19% greater risk of increase in menstrual cycle length as compared to unvaccinated people or pre-vaccination time-periods (summary relative risk (sRR): 1·19; 95% CI: 1·11-1·26; n = 23,718 participants). The increase in risk was the same for Pfizer-BioNTech (sRR: 1·15; 1·05-1·27; n = 16,595) and Moderna vaccines (sRR: 1·15; 1·05-1·25; n = 7,523), similar for AstraZeneca (sRR: 1·27; 1·02-1·59; n = 532), and higher for the Janssen (sRR: 1·69; 1·14-2·52; n = 751) vaccine. In the first cycle after vaccination, length increased by