First live birth before surgical verification of endometriosis-a nationwide register study of 18 324 women.

study questionDo women with endometriosis have lower first live birth rate before surgical diagnosis than women without verified endometriosis? SUMMARY ANSWER: Compared to reference women, the incidence of first live birth was lower in women prior to surgical verification of endometriosis irrespective of the type of endometriosis. WHAT IS KNOWN ALREADY: Endometriosis is associated with pain and reduced fertility. The mechanism of infertility is partly explained by anatomical, endocrinological, and immunological changes. Over the past decades, the treatment of both endometriosis and infertility has evolved. Knowledge of fertility far before surgical diagnosis of endometriosis in large cohorts and of different types of endometriosis has been lacking. The diagnostic delay of endometriosis is long, 6-7 years. STUDY DESIGN, SIZE, DURATION: Retrospective population-based cohort study focused on the time period before the surgical verification of endometriosis. All women with surgical verification of endometriosis in 1998-2012 were identified from the Finnish Hospital Discharge Register and the reference cohort from the Central Population Register. Data on deliveries, gynecological care, and sociodemographic factors before the surgical diagnosis were gathered from Finnish national registers maintained by the Finnish Institute for Health and Welfare, the Digital and Population Data Services Agency, and Statistics Finland. PARTICIPANTS/MATERIALS, SETTING,

methodsAll women aged 15-49 years at the time of surgical verification of endometriosis (ICD-10: N80.1-N80.9) in Finland during 1998-2012 were identified (n = 21 620). Of them, we excluded women born in 1980-1999 due to the proximity of the surgical diagnosis (n = 3286) and women left without reference (n = 10) for the final endometriosis cohort of 18 324 women. From the final cohort, we selected sub-cohorts of women with isolated diagnosis of ovarian (n = 6384), peritoneal (n = 5789), and deep (n = 1267) endometriosis. Reference women were matched by age and residence and lacked registered clinical or surgical diagnosis of endometriosis (n = 35 793). The follow-up started at the age of 15 years and ended at the first birth, sterilization, bilateral oophorectomy, hysterectomy, or until the surgical diagnosis of endometriosis or corresponding index day-whichever came first. Incidence rate (IR) and the incidence rate ratio (IRR) of first live birth before the surgical verification of endometriosis with corresponding CIs were calculated. In addition, we reported the fertility rate of parous women (the number of all children divided by the number of parous women in the cohort) until the surgical verification of endometriosis. The trends in first births were analysed according to the women's birth cohort, type of endometriosis, and age. MAIN RESULTS AND THE ROLE OF CHANCE: Surgical diagnosis of endometriosis was set at the median age of 35.0 years (IQR 30.0-41.4). Altogether 7363 women (40.2%) with endometriosis and 23 718 (66.3%) women without endometriosis delivered a live born infant before the index day (surgery). The IRs of the first live birth per 100 person-years were 2.64 (95% CI 2.58-2.70) in the endometriosis cohort and 5.21 (95% CI 5.15-5.28) in the reference cohort. Between the endometriosis sub-cohorts, the IRs were similar. The IRR of the first live birth was 0.51 (95% CI 0.49-0.52) between the endometriosis and reference cohorts. Fertility rate per parous woman before the surgical diagnosis was 1.93 (SD 1.00) and 2.16 (SD 1.15) in the endometriosis and reference cohorts (P < 0.01). The median age at the first live birth was 25.5 (IQR 22.3-28.9) and 25.5 (IQR 22.3-28.6) years (P = 0.01), respectively. Between the endometriosis sub-cohorts, women in the ovarian sub-cohort were the oldest at the time of surgical diagnosis with the median age of 37.2 years (IQR 31.4-43.3), (P < 0.001). Altogether 44.1% (2814) of the women with ovarian, 39.4% (2282) with peritoneal, and 40.8% (517) with deep endometriosis delivered a live born infant before the diagnosis. IRRs between the endometriosis sub-cohorts did not differ. Fertility rate per parous woman was lowest, 1.88 (SD 0.95), in the ovarian sub-cohort compared to 1.98 (SD 1.07) in the peritoneal and 2.04 (SD 0.96) in deep endometriosis (P < 0.001). Women with ovarian endometriosis were oldest at first live birth compared to women in other sub-cohorts with a median age of 25.8 years (IQR 22.6-29.1) (P < 0.001). Cumulative distributions of first live birth were presented according to age at first live birth and birth cohorts of the participants. LIMITATIONS, REASONS FOR CAUTION: The increasing age at first live birth, increasing practice of clinical diagnostics, conservative treatment of endometriosis, a possible effect of coexisting adenomyosis, and use of artificial reproductive treatments should be considered when assessing the results. In addition, the study is limited due to possible confounding effects of socioeconomic factors, such as level of education. It should be noted that, in this study, we assessed parity only during the years preceding the surgical verification of endometriosis. WIDER IMPLICATIONS OF THE

findingsThe need for early diagnosis and relevant treatment of endometriosis appears clear given the impairment of fertility prior to its surgical verification. STUDY FUNDING/COMPETING INTEREST(S): The study was funded by the Hospital District of Helsinki and Uusimaa and by Finska Läkaresällskapet. The authors report no conflicts of interest. All authors have completed the ICMJE Disclosure form. TRIAL REGISTRATION NUMBER: N/A.