The success of an IVF/ICSI treatment depends substantially on the quality of transferred embryos. Among numerous factors affecting embryo quality, ovarian stimulation is an eligible and adjustable one. Despite the established clinical impact of different stimulation protocols, analysis of ovarian stimulation on quality of oocytes and developing embryos is not well known yet.
Nevertheless since before 2000 the combination of GnRH antagonists and gonadotrophins has also been available, the GnRH-agonist long protocol remained the first choice in most IVF centres  as it is in our IVF department, too. We choose GnRH antagonist protocols at the first IVF cycle of the patient almost exclusively when the duration of the pretreatment and ovarian stimulation is limited; this explains the small number of patients in our study in spite of the relatively long trial period.
As previous studies reported [19, 20], we also showed that patients need less HMG ampoulles and the length of stimulation is shorter with the use of GnRH antagonists protocol. We aspirated significantly more follicles and we retrieved significantly more oocytes with the use of GnRH agonist; most of the comparative studies of GnRH analogues had similar results [9, 19, 20].
These clinical aspects have been evaluated in several studies, but only a recent study focused on the differences in embryo quality according to the type of gonadotrophin used for ovarian stimulation .
Like in some previous studies, there was no significant difference between the rate of mature metaphase II oocytes in the two groups in our study, however this parameter was examined only by a few workgroups and one study was made on a special group of non-obese PCOS patients [18, 19, 24].
Several studies focused on the role of oocyte quality in predicting treatment outcome. Granularity in the perivitelline space seems to be a physiological phenomenon in oocytes and it could be enhanced by exposure to high dosages of gonadotrophins . In a recent study stimulation protocol prooved to influence significantly the zona pellucida score (agonist protocol resulted in better score compared to the antagonist one) . The presence of intracytoplasmic abnormalities can refer to the quality of the oocyte [12, 13]. Otsuki et al. confirmed pronucleus sized translucent vacuoles in oocytes as tubular-type smooth endoplasmic reticulum clusters (sERCs) . sERC positive oocytes were observed more frequent in GnRH agonist short protocols compared to long ones. Comparing GnRH agonist long proctocol to GnRH antagonist cycles in our study we have also found significantly less oocytes with cytoplasmic abnormalities after administrating GnRH agonists in long protocol.
According to the method of fertilization (IVF or ICSI) there was no significant difference between the two groups in the rate of normally fertilized oocytes; this parameter was similar in both groups independent of the method of fertilization. In previous studies the rate of the normally fertilized oocytes was examined during conventional IVF and ICSI treatments together only: cumulatively there was no significant difference between the two groups . However these parameters were similar in the two groups, accordingly the lower count of retrieved oocytes, the average number of normally fertilized zygotes was significantly lower in the antagonist cycles. Normal nucleolar distribution was also significantly lower in the antagonist group. This parameter has not been examined yet in previous studies in view to the type of GnRH analogue used for stimulation. The zygotes with normal pronuclear morphology are supposed to develop most likely top-quality embryos [27–29].
Dynamics of early embryonic development could reflect the developmental potential of the embryo. The first cleavage can be examined with the breakdown of the pronuclear membrane (the start of the M phase of the first cell cycle) and directly with the presence of cleavage, because the duration of the M phase is relatively constant (3-4 hours) [30, 31]. It is known that early cleavage is a strong indicator of the quality and the viability of the embryos [30, 32, 33], although a recent study showed higher implantational potential for early-cleavage embryos only with the use of GnRH agonists . We observed significantly higher number of blastomers in the antagonist group on day 2, while the higher presence of early cleavage in this group did not reach statistical significance.
There was no significant difference in the rate of the top quality embryos. (Embryo quality and development were examined also on day 3 [64-72 hours postinsemination], but only the results of day 2 are analyzed in this study, because part of the embryos are used to be transferred on day 2 already. Hence the results of day 3 would not be representative in this study). Other studies had similar results in the rate of the top quality embryos, but the early cleavage, the number of multinucleated blastomeres (multinucleated embryos have poor implantation potential ) and the amount of fragmentation have not been yet examined during comparative studies of GnRH analogues.
The clinical pregnancy rate, which shows the effectiveness of the treatments, was examined in all of the comparative studies. Co-administrating GnRH agonists during gonadotrophin ovarian stimulation seems to result in higher pregnancy rates, however this difference proved not to be significant in most of the studies [19, 20, 24]. Despite of the not significant difference in our study in pregnancy rates, the presented odds ratios indicate higher pregnancy rates by GnRH agonists. The higher number of cycles with supernumerous embryos appropriate for cryopreservation in the agonist group favours the patient avoiding a repeated ovarian stimulation and oocyte retrieval procedure in a contingent next treatment.