GnRHa (alone or combined with hCG) versus hCG alone for ovulation triggering during controlled ovarian stimulation for IVF/ICSI: a systematic review and meta-analysis

Abstract Objective To evaluate whether GnRHa triggering improves oocyte maturation, pregnancy outcomes and safety compared to hCG triggering during controlled ovarian stimulation. Evidence review A systematic review was performed using the following keywords: ‘GnRH agonist’, ‘hCG’, ‘triggering’. Searches were conducted on MEDLINE, EMBASE, the Cochrane Library, ClinicalTrials.gov , and EudraCT for randomised controlled clinical trials from 01 January 1990 to 15 April 2020. The primary outcomes were the total number of retrieved oocytes and the number of mature oocytes. The main secondary outcomes were the number of embryos obtained, the clinical pregnancy rate, the early pregnancy loss rate, the live birth rate, and the incidence of ovarian hyperstimulation syndrome (OHSS). Two independent reviewers carried out the study selection, the bias assessment using the RoB2 tool, and the data extraction according to Cochrane methods. Random-effects meta-analysis was performed followed by prespecified sensitivity and subgroup analyses. Result(s) Our search yielded 1,369 published studies and 216 unpublished studies. After screening the titles and the abstracts, 65 published studies and 25 unpublished abstracts were assessed for eligibility. Of these, we excluded 61 studies. A total of 29 RCT were included. The 26 studies with number of oocytes retrieved enrolled a total of 2,755 women, of whom 1,419 had GnRHa triggering and 1,336 had hCG alone for triggering. A total of 12 studies reported the number of mature oocytes with a total of 1,619 women (806 had GnRHa triggering and 813 had hCG alone for triggering). The mean number of retrieved oocytes [difference in means (95%CI)0.99(0.21, 1.78); p=0.01; n=26] and of mature oocytes [0.68(0.04,1.33); p=0.04; n=12] were statistically significantly higher after GnRHa than after hCG triggering. A similar difference was observed for the number of embryos [0.94(0.19,1.68); p=0.01; n=10]. No differences in the clinical pregnancy rate [risk ratio 1.01(0.90, 1.14); p=0.83; n=23], early pregnancy loss [1.27(0.94, 1.71); p = 0.13; n = 16], and live birth rate [1.00(0.77,1.29); p=0.97; n = 6] were noted. GnRHa was associated with a lower incidence of ovarian hyperstimulation syndrome [odds ratio 0.25(0.08,0.74); p=0.012; n=20]. Moreover, after dual triggering (GnRHa associated to hCG) compared to hCG alone, the meta-analysis showed a statistically significant higher number of oocytes retrieved (p=0.006), of mature oocytes (p=0.04) and clinical pregnancy rate (p=0.02). Conclusion(s) The final triggering using GnRHa allows a higher number of retrieved and mature oocytes to be obtained with comparable clinical outcomes and, after GnRHa alone, a lower OHSS risk, compared to hCG triggering.