Idiopathic early ovarian ageing: risk of miscarriage and chance of delivery following ART in a nationwide cohort study

STUDY QUESTION: Is idiopathic reduced ovarian reserve in young women, quantified as low response to ovarian stimulation in ART, associated with a concomitant loss of oocyte quality as determined by risk of pregnancy loss and chance of clinical pregnancy and live birth? SUMMARY ANSWER: Young women with idiopathic accelerated loss of follicles exhibit a similar risk of pregnancy loss as young women with normal ovarian reserve. WHAT IS KNOWN ALREADY: Normal ovarian ageing is described as a concomitant decline in oocyte quantity and quality with increasing age. Conflicting results exist with regard to whether a similar decline in oocyte quality also follows an accelerated loss of follicles in young women. STUDY DESIGN, SIZE, DURATION: This national register-based, historical cohort study included treatment cycles from young women (≤37 years) after ART treatment in Danish public or private fertility clinics during the period 1995–2014. The women were divided into two groups dependent on their ovarian reserve status: early ovarian ageing (EOA) group and normal ovarian ageing (NOA) group. There were 2734 eligible cycles in the EOA group and 22 573 in the NOA group. Of those, 1874 (n ¼ 1213 women) and 19 526 (n ¼ 8814 women) cycles with embryo transfer were included for analyses in the EOA and NOA group, respectively. PARTICIPANTS/MATERIALS, SETTING, METHODS: EOA was defined as ≤5 oocytes harvested in both the first and second cycle stimulated with FSH. The NOA group should have had at least two FSH-stimulated cycles with ≥8 oocytes harvested in either the first or the second cycle. Cases with known causes influencing the ovarian reserve (endometriosis, ovarian surgery, polycystic ovary syndrome, chemotherapy, etc.) were excluded. The oocyte quality was evaluated by the primary outcome defined as the overall risk of pregnancy loss (gestational age (GA) ≤22 weeks) following a positive hCG and further stratified into: non-visualized pregnancy loss, early miscarriage (GA ≤ 12 weeks) and late miscarriage (GA > 12 weeks). Secondary outcomes were chance of clinical pregnancy and live birth per embryo transfer. Cox regression models were used to assess the risk of pregnancy loss. Time-to-event was measured from the day of embryo transfer from the second cycle and subsequent cycles. Logistic regression models were used to assess the chance of clinical pregnancy and live birth. MAIN RESULTS AND THE ROLE OF CHANCE: The overall risk of pregnancy loss for the EOA group was comparable with the NOA group (adjusted hazard ratio: 1.04, 95% CI: 0.86; 1.26). Stratifying by pregnancy loss types showed comparable risks in the EOA and NOA group. The odds of achieving a clinical pregnancy or live birth per embryo transfer was lower in the EOA group compared to the NOA group (adjusted odds ratio: 0.77 (0.67; 0.88) and 0.78 (0.67; 0.90), respectively). LIMITATIONS, REASONS FOR CAUTION: Only women with at least two ART cycles were included. We had no information on the total doses of gonadotropin administered in each cycle. WIDER IMPLICATIONS OF THE FINDINGS: The present findings may indicate that mechanism(s) other than aneuploidy may explain the asynchrony between the normal-for-age risk of miscarriage and the reduced chance of implantation found in our patients with EOA. The results of this study could be valuable when counselling young patients with low ovarian reserve. STUDY FUNDING/COMPETING INTERESTS(S): The study was funded by the Health Research Fund of Central Denmark Region. The authors have no conflict of interest to declare. TRIAL REGISTRATION NUMBER: N/A.