Our 6-year of fertility treatment data on PCOS patients showed the incidence rate of UPOR in prolonged protocol is 24.8% (444/1789) which was first exactly reported. It is difficult to accurately predict the ovarian response(23) and the management of UPOR is still an open field(24). Therefore, it is necessary to formulate strategies for such special and general population and our research would make contributions to it.
Our present study unprecedentedly compared the outcomes between IVF/M and IVF for patients in prolonged protocol era. The results showed that in the IVF/M Group compared with the IVF Group, a smaller number of oocytes retrieved, a smaller proportion achieved maturity and the embryonic development were worse, which was consistent with previous study. However, the comparison of fertilization rate in two treatments was not consistent. Gremeau AS et al. found lower fertilization rate in IVM which was contrast to our studies while Walls ML et al. found the fertilization rate in IVF/M Group were higher than IVF without statistic difference (25, 26). This may result from that ICSI was advocated as the preferred method for metaphase II stage oocytes fertilization derived from IVF/M for the hardening of the zona pellucida during maturation culture in our study.
In fresh ET, we found a lower hCG positive rate, clinical pregnancy rate and live birth rate. However, we found that IVF/M wasn’t a significant independent factor for fresh ET non-live birth. The dominant factors that account for non-live birth hasn’t been found in fresh ET through logistic regression analysis. Further research should be well designed to explore the factors that associated to the live birth rate in fresh ET.
In our present paper, IVF/M showed significantly lower clinical pregnancy rate, lower cumulative live birth rate, and higher miscarriage rate. IVM was the only independent factor that associated to non-live birth in complete egg retrieval. 36 patients in IVM Group got no available blastocyst which account for 23.1% non-live birth and only 5.6% in IVF Group. We analyzed the laboratorial factors that may have effect on cumulative live birth rate, and we performed blastocysts transfer only. Therefore, the number of available blastocysts and high-quality blastocysts were the main significant factors. We suspected that IVF/M got less available blastocysts and high-quality blastocysts which eventually affect the cumulative birth rate. The future profile for embryologists was to devote to the cultivation of blastocysts in IVF/M which was coincides with the conception of blastocyst-transferred in ART(17, 22, 27).
IVM had been often compared with IVF in luteal phase long protocol and GnRH-anta protocol (25, 26, 28, 29). The development and advantages of IVM such as elimination or nominal use of ovarian stimulation, minimizing the risk of OHSS, patient-friendly, cost-effective were out of question. So was the lower live birth rate than IVF. However, IVM strategy in previous study were normally performed with or without FSH administration. Our study unprecedentedly analyzed switching IVF to IVM after a full dose of GnRH-a. Prolonged pituitary downregulation has gained great results in IVF for the effect on endometrium and LH level(19, 30). Though the lower live birth rate than IVF limited the utilization of IVF/M and the popularity among patients. We realized that live birth rate of 38.9% for the fresh ET and the cumulative live birth rate of 35.3% in IVM were comparable to that reported before or even higher (16, 25, 26). Therefore, we have reasons to confirm the feasibility of IVF/M in prolonged protocol for PCOS patients with UPOR. GnRH-a could regulate endometrium through vascular endothelial growth factor, endometrial interleukin 1 system, αvβ3 vitronectin and endometrial receptivity markers(31–33). As is known, LH would be inhibited after downregulation. But the relationship between LH level and ART outcomes was still on debate(34). No study has been demonstrated on whether LH level would affect IVM outcomes. Therefore, IVM between different protocols and how low LH level would make a difference on IVM results still need further study. We have faith that technology will bring us a better IVM outcomes and IVM will be simple and safe strategy.
In IVF Group, we found a high incidence rate of OHSS, and moderate to severe OHSS was 8.0% which was very similar as reported before(35). It was a common complication of PCOS in ovarian stimulation and was associated with the number of oocytes retrieved and E2 level on hCG trigger day(36). Dominique de Ziegler thought IVM with non-stimulated follicles in PCOS was not indicated in the new GnRH antagonist for OHSS can be prevented by other methods(29). We wish to find ways to reduce the moderate to severe OHSS rate in GnRH-a protocol which would threat female’s health. Researchers have explored letrozole, bromocriptine, prednisone to decrease the total incidence of moderate to severe OHSS. However, the effect was still controversial(37–40). However, the absence of OHSS only has been reported in IVM strategy(41). Therefore, for patients on a high risk of OHSS, IVF/M must be informed to the patient that it can completely prevent OHSS.
IVM treatment has gained reputation of economic for eliminating or minimizing the use of gonadotrophins. However, in our IVF/M Group, patients were administrated at least 8–10 days ovarian stimulation of adequate gonadotropin. Some patients were even injected more FSH to resist the continuous UPOR. Unluckily the follicles made no further development and had to switching IVF to IVM. In those patients, we achieved less live birth with high cost. Thus, for preventing OHSS and saving cost, we suggested that patients should be performed IVM as earlier as the patients were undergoing moderate to severe OHSS tendency. Even patients obtained no live birth, the puncture of small follicles can be equal to the treatment of transvaginal ovarian drilling which might benefit the next ovarian stimulation cycle(42).
The pregnancy complications and neonatal complications in IVF/M were similar as those in IVF. No evidence showed there was any safety issue in IVF/M strategy.
There are two limitations of this study. Firstly, this is a retrospective single-center study which requires a large multicenter, prospective randomized controlled study to confirm the feasibility of IVF/M in prolonged protocol urgently. Secondly, a study of the comparation between IVF/M in prolonged protocol urgently and IVM should be performed to provide solid evidence to the feasibility of IVF/M.