Endometriosis and early pregnancy outcomes: a registry-based cohort study in Finland

BACKGROUND: Understanding of the impact of endometriosis on adverse early pregnancy outcomes has evolved in recent decades and has been partly attributed to endocrine dysfunction and chronic inflammation. However, data from population-based cohort studies which include all pregnancy outcomes, with long-term follow-up, are lacking. OBJECTIVE: This longitudinal cohort study aimed to investigate the association between endometriosis and early pregnancy outcomes including miscarriage, ectopic pregnancy, induced abortion, and molar pregnancy. STUDY DESIGN: We conducted a retrospective registry study including all 10,105 people in Finland born after 1972 with a first surgical diagnosis of endometriosis between 1998 and 2012 in the Finnish Care Register for Health Care. We also included 19,526 age- and residence-matched individuals. We collected all registered pregnancies of the study cohorts-both prior to and following surgical diagnosis of endometriosis-from Finnish registries from age 15 until the occurrence of sterilization, bilateral oophorectomy, sequential unilateral oophorectomies, hysterectomy, death, emigration, or the end of the study period, whichever occurred first. We calculated the cumulative incidence and incidence rate of pregnancy. For the specific pregnancy outcomes, we calculated their proportions, risk per 100 pregnancies, and risk ratios adjusting for age, gravidity, and education level. In the endometriosis cohort, there were 256,906 person-years of follow-up, and in the reference cohort, 503,286 person-years. RESULTS: We identified 19,141 pregnancies in 7731 (76.5%) ever-pregnant participants with endometriosis and 43,478 pregnancies in 15,421 (79.0%) ever-pregnant reference participants during a median follow-up time of 26.4 years (interquartile range, 22.5-29.6). The mean number of registry-identified pregnancies per person in the endometriosis cohort was 1.89 (standard deviation, 1.64), compared to 2.23 (1.92) in the reference cohort (P<.001). Of the pregnancies with known outcomes-that is, excluding pregnancies which were ongoing at the end of the study period-72.0% vs75.7% were births (P<.001), 15.6% vs10.4% were miscarriages (P<.001), 3.5% vs1.5% were ectopic pregnancies (P<.001), 8.8% vs12.3% were induced abortions (P<.001), and 0.1% vs0.1% were molar pregnancies (P=.99) in the endometriosis and in the reference cohorts, respectively. Individuals with endometriosis had lower pregnancy rates compared to the reference cohort, with an incidence rate ratio of 0.87 (95% confidence interval, 0.86-0.89) when adjusted for education level and birth year. Conversely, they experienced significantly higher age-adjusted risks of miscarriage (adjusted risk ratio 1.51; 1.43-1.60) and ectopic pregnancy (2.45; 2.21-2.71). The risk of induced abortion was lower in those with endometriosis (0.69; 0.65-0.74), and there was no significant difference in the risk of molar pregnancy (1.03; 0.60-1.56). CONCLUSION: Individuals with surgically verified endometriosis had a lower lifetime incidence of pregnancy compared to those without endometriosis. Among pregnancies in those with endometriosis, there were fewer births and induced abortions, but a higher risk of early pregnancy loss due to increased proportions of miscarriage and ectopic pregnancy.