Purpose: Does controlled ovarian stimulation (COS) and progesterone (P) luteal supplementation modify the vaginal and endometrial microbiota of women undergoing in vitro fertilization? Methods: Fifteen women underwent microbiota analysis at two time points: during a mock transfer performed in the luteal phase of the cycle preceding COS, and at the time of fresh embryo transfer (ET). A vaginal swab and the distal extremity of the ET catheter tip were analyzed using next-generation 16SrRNA gene sequencing. Heterogeneity of the bacterial microbiota was assessed according to both the Bray-Curtis similarity index and the Shannon diversity index. Results: Lactobacillus was the most prevalent genus in the vaginal samples, although its relative proportion was reduced by COS plus P supplementation (71.5 ± 40.6% vs. 61.1 ± 44.2%). In the vagina, an increase in pathogenic species was observed, involving Prevotella (3.5 ± 8.9% vs. 12.0 ± 19.4%), and Escherichia coli-Shigella spp. (1.4 ± 5.6% vs. 2.0 ± 7.8%). In the endometrium, the proportion of Lactobacilli slightly decreased (27.4 ± 34.5% vs. 25.0 ± 29.9%); differently, both Prevotella and Atopobium increased (3.4 ± 9.5% vs. 4.7 ± 7.4% and 0.7 ± 1.5% vs. 5.8 ± 12.0%). In both sites, biodiversity was greater after COS (p < 0.05), particularly in the endometrial microbiota, as confirmed by Bray-Curtis analysis of the phylogenetic distance among bacteria genera. Bray-Curtis analysis confirmed significant differences also for the paired endometrium-vagina samples at each time point. Conclusions: Our findings suggest that COS and P supplementation significantly change the composition of vaginal and endometrial microbiota. The greater instability could affect both endometrial receptivity and placentation. If our findings are confirmed, they may provide a further reason to encourage the freeze-all strategy.

Background and Objectives: Some biomarkers of ovarian responsiveness to gonadotropins and the total number of retrieved oocytes are known to affect the success rate after controlled ovarian stimulation (COS) and in vitro fertilization (IVF). The aim of this study was to study another putative marker, the Ovarian Sensitivity Index (OSI: (number of retrieved oocytes/total gonadotropin dose) × 1000), assessing whether (a) it correlates with ovarian responsiveness biomarkers, (b) it is an independent predictor of clinical pregnancy, (c) it predicts clinical pregnancy comparably to the number of retrieved oocytes, and (d) it is consistent in the repeated COS cycles of the same woman. Design: retrospective analysis. Setting: public IVF Unit in University Hospital. Cases and Measurements: 1612 patients submitted to 3353 IVF cycles were included, their OSI was calculated and it was correlated with the ovarian responsiveness biomarkers (age, BMI, anti-Mullerian hormone, antral follicle count). The OSI and the total number of oocytes were compared for their value in predicting clinical pregnancy. The inter-cycle consistency of the OSI was estimated in 209 patients who underwent two consecutive cycles in which the ovarian stimulation regimen was changed from the Gonadotropin-releasing Hormone (GnRH)-agonist long protocol to the GnRH-antagonist protocol or vice-versa. Results: The OSI turned out to be significantly related to age and BMI (inversely), the anti-Mullerian hormone (AMH) and the antral follicle count (AFC) (directly), to be an independent predictor of clinical pregnancy, and to correlate with clinical pregnancy better than the total number of oocytes (p < 0.0001 vs. <0.002). In patients who underwent two consecutive COS cycles changing stimulation regimen, the OSI showed 82% consistency. Conclusion(s): The OSI significantly correlates to the currently used biomarkers of ovarian responsiveness; it is an independent predictor of clinical pregnancy; it is more predictive of clinical pregnancy than the total number of oocytes, and is highly consistent in repeated IVF cycles even when the COS protocol changes. These characteristics make the OSI quite suitable to be incorporated into more complex prediction models of IVF outcome.