The impact of total gonadotropin dose on spontaneous abortion rates and early pregnancy outcomes in infertile couples undergoing gonadotropin induction/intrauterine insemination cycles

Abstract

Studies suggest that ovarian hyperstimulation with high dose gonadotropin regimens in women undergoing IVF may affect endometrial receptivity, as well as oocyte and embryo quality [1, 2]. Furthermore, high doses of gonadotropins increase aneuploidy rates in animal models and are suspected to increase aneuploidy in human embryos [3]. The potential impact of high dose gonadotropins on spontaneous abortion rates (SABR) and early pregnancy outcomes in gonadotropin induced/intrauterine insemination (Gn/IUI) cycles is unknown. To evaluate the impact, if any, of total gonadotropin dose on SABR (surrogate outcome for aneuploidy), and early pregnancy outcomes in Gn/IUI cycles. Design: Retrospective cohort. Setting: Academic fertility center Interventions: A total of 2099 Gn/IUI cycles (12/2003-1/2016; 923 patients) were analyzed. Exclusion criteria included: i) age older than (≥) 38 years, ii) body mass index (BMI) above (≥) 35 kg/m2, iii) diagnosis of diminished ovarian reserve or polycystic ovary syndrome. Cycles were stratified by total gonadotropin dose i) into two groups: ≤ and > 75th percentile (Gr1 and Gr2, respectively), and ii) into quartiles (Q1-Q4). Outcome Measures: Primary outcome measure was the rate of SAB per cycle. Secondary outcome measures included: clinical, multiple, and non-clinical (ectopic/biochemical) pregnancy rates (CPR, MPR, and non-CPR, respectively). Statistics: t-test, Mann Whitney U- and χ2- tests were used as appropriate for comparisons between groups 1 and 2. Using logistic regression models, odds ratios (ORs) and 95% confidence intervals (CI) were calculated for SAB, clinical, and non-clinical pregnancy across quartiles adjusting for potential confounders (age, BMI, day-3 FSH, infertility diagnosis and endometrial thickness). P<0.05 was considered statistically significant. Table 1 summarizes the demographics and cycle characteristics of the study population stratified by the 75th percentile cut-off. Patients receiving total gonadotropin dose >75th percentile, when compared to those receiving doses ≤75th percentile, experienced similar SABR (18.6 vs. 12.9%, p = 0.19; for Gr2 vs. Gr1, respectively). However, CPR was significantly higher in the former (>75th) compared to the latter (≤75th) group (19.1 vs. 13.2% for Gr2 vs. Gr 1, respectively, p <0.05), while MPR and non-CPR were comparable between the two groups (13.5 vs. 12.9%, p = 0.88; and 7.7 vs. 6.6%, p = 0.72 for Gr2 vs. Gr 1, respectively). Similarly, when calculating adjusted OR (95%CI) across quartiles of total Gn dose, using Q1 as the reference group, we noted no differences in the incidence of SAB and non-clinical pregnancy (Fig. 1A-B), while the OR for clinical pregnancy was significantly higher in both Q3 and Q4 and nearly significant in Q2 compared to Q1 (Fig. 1C). Despite evidence of potential adverse effects of higher Gn doses on oocyte and embryo quality and possibly on endometrial receptivity in IVF cycles, the overall lower doses of stimulation medication administered during Gn/IUI cycles, do not seem to have a similar negative impact on SABR and early pregnancy outcomes.