Family size for women with primary ovarian insufficiency and their relatives

Abstract

Abstract STUDY QUESTION How does the number of children in women with primary ovarian insufficiency (POI) compare to the number for control women across their reproductive lifespans? SUMMARY ANSWER Approximately 14% fewer women with POI will have children, but for those able to have children the median number is 1 less than for age-matched controls. WHAT IS KNOWN ALREADY Women with POI are often identified when presenting for fertility treatment, but some women with POI already have children and there remains a low chance for pregnancy after the diagnosis. Further, POI is heritable, but it is not known whether relatives of women with POI have a smaller family size than relatives of controls. STUDY DESIGN, SIZE, DURATION The study was a retrospective case–control study of women with POI diagnosed from 1995 to 2021 (n = 393) and age-matched controls (n = 393). PARTICIPANTS/MATERIALS, SETTING, METHODS Women with POI were identified using ICD9 and 10 codes in electronic medical records (1995–2021) from two major healthcare systems in Utah and reviewed for accuracy. Cases were linked to genealogy information in the Utah Population Database. All POI cases (n = 393) were required to have genealogy information available for at least three generations of ancestors. Two sets of female controls were identified: one matched for birthplace (Utah or elsewhere) and 5-year birth cohort, and a second also matched for fertility status (children present). The number of children born and maternal age at each birth were ascertained by birth certificates (available from 1915 to 2020) for probands, controls, and their relatives. The Mann–Whitney U test was used for comparisons. A subset analysis was performed on women with POI and controls who delivered at least one child and on women who reached 45 years to capture reproductive lifespan. MAIN RESULTS AND THE ROLE OF CHANCE Of the 393 women with POI and controls, 211 women with POI (53.7%), and 266 controls (67.7%) had at least one child. There were fewer children born to women with POI versus controls (median (interquartile range) 1 (0–2) versus 2 (0–3); P = 3.33 × 10−6). There were no children born to women with POI and primary amenorrhea or those <25 years old before their diagnosis. When analyzing women with at least one child, women with POI had fewer children compared to controls overall (2 (1–3) versus 2 (2–4); P = 0.017) and when analyzing women who reached 45 years old (2 (1–3) versus 3 (2–4); P = 0.0073). Excluding known donor oocyte pregnancies, 7.1% of women with POI had children born after their diagnosis. There were no differences in the number of children born to relatives of women with POI, including those with familial POI. LIMITATIONS, REASONS FOR CAUTION The data are limited based on inability to determine whether women were trying for pregnancy throughout their reproductive lifespan or were using contraception. Unassisted births after the diagnosis of POI may be slightly over-estimated based on incomplete data regarding use of donor oocytes. The results may not be generalizable to countries or states with late first births or lower birth rates. WIDER IMPLICATIONS OF THE FINDINGS Approximately half of women with POI will bear children before diagnosis. Although women with POI had fewer children than age matched controls, the difference in number of children is one child per woman. The data suggest that fertility may not be compromised leading up to the diagnosis of POI for women diagnosed at 25 years or later and with secondary amenorrhea. However, the rate of pregnancy after the diagnosis is low and we confirm a birth rate of <10%. The smaller number of children did not extend to relatives when examined as a group, suggesting that it may be difficult to predict POI based on family history. STUDY FUNDING/COMPETING INTEREST(S) The work in this publication was supported by R56HD090159 and R01HD099487 (C.K.W.). We also acknowledge partial support for the Utah Population Database through grant P30 CA2014 from the National Cancer Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors have no conflicts of interest. TRIAL REGISTRATION NUMBER N/A.