Association of follicular phase length (fpl) and early pregnancy loss (epl) in women with unexplained infertility undergoing ovarian stimulation with intrauterine insemination (os-iui) cycles

Pregnancy loss < 20 weeks gestations is the most common adverse event in early pregnancy, occurring in 10–20% of clinically recognized pregnancies. Given common occurrence, significant associated physical and psycho-social morbidities, it is imperative to identify patient and cycle characteristics that may lead to increased risk for miscarriage. In this study, we aimed to evaluate the association between FPL and EPL among women with unexplained infertility undergoing OS-IUI cycles. We hypothesized that women with short FPL may have increased risk of EPL due to inability to attain full oocyte maturation or inadequate development of the endometrium. Secondary analysis of a prospective, randomized, multicenter clinical trial investigating pregnancy, live-birth, and multiple pregnancy rates following OS-IUI, the Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation (AMIGOS) trial. Among 2546 cycles from 869 AMIGOS participants, there were 320 achieved pregnancies defined by an initial rise in serum beta-hCG levels. FPL was evaluated as a categorical variable defined by quintiles (q1: ≤ 11 days, q2: 12 days, q3: 13 days, q4: 14-15 days, and q5: ≥16 days). EPL was defined as any achieved pregnancy that did not result in a live birth. Pattern of EPL by FPL quintile was evaluated using the Cochran Armitage trend test. Risk ratios (RR) and 95% confidence intervals (CI) were calculated using modified Poisson regression models with robust standard errors. Covariates evaluated in multivariable models included age, race/ethnicity, BMI, parity, duration of infertility, AMH, number of follicles >16 mm) and treatment group (clomiphene/letrozole vs gonadotropins). Age was the only covariate that changed the point estimates by more than 10% and thus adjusted models control only for age. Of the 320 achieved pregnancies, 10 were lost to follow up and excluded from the analysis. After excluding 3 cases who pursued induced abortion/selective reduction, there were overall 102 (33.3%) pregnancy losses. EPL was lowest for the 1st FPL quintile (≤ 11 days, 19.4%) and highest for 5th FPL quintile ( ≥16 days, 45.5%; trend test p=0.02), although rates did not monotonically rise with each increasing FPL quintile. When ≥16 days was used as the referent, FPL ≤ 11 days was associated with a 62% lower risk of EPL [adjusted aRR 0.38 (95% CI 0.21, 0.70)]. When stratified by treatment group, the 1st FPL quintile was similarly associated with reduced risk of EPL in the Clomiphene/Letrozole group [aRR 0.21 (95% CI 0.08, 0.75)] but this association was attenuated in the gonadotropin group [aRR 0.62, 95% CI 0.31, 1.24)]. Risk of EPL may be lowest in OS-IUI treatment cycles characterized by reduced FPL.