GnRH agonist for triggering final oocyte maturation: time for a critical evaluation of data.

Sir, We read with great interest the manuscript by Humaidan et al. (2011). In this review, the authors summarize the published evidence regarding the use of GnRH agonist for triggering final oocyte maturation in patients undergoing IVF. By combining the results of six recently published randomized controlled trials (RCTs), the authors suggest that when a modified luteal phase support (LPS) is employed, following GnRH agonist triggering, pregnancy rates are comparable with those achieved after hCG triggering and standard LPS. Moreover, they suggest that GnRH agonist triggering combined with a modified LPS results in elimination of ovarian hyperstimulation syndrome (OHSS). These findings are very interesting, since they support the notion that GnRH agonist triggering when combined with a modified LPS is both an effective and safe treatment. We would like to elaborate on the data presented by the authors and extend the implications of the available evidence on the efficacy and safety of GnRH agonist triggering. The RCTs that have been performed so far in which a modified LPS is employed, when GnRH agonist triggering is used, include two distinct patient populations: (i) normo-ovulatory patients, at a normal risk for OHSS (Humaidan et al., 2006, 2010; Pirard et al., 2006; Papanikolaou et al., 2011) and (ii) patients with polycystic ovary syndrome (PCOS) at a high risk for OHSS (Babayof et al., 2006; Engmann et al., 2008). In addition, two distinct strategies for a modified LPS are used in these two populations. In normo-ovulatory patients, a modified LPS involves LH activity by administration of (i) low-dose bolus hCG (Humaidan et al., 2006, 2010), (ii) multiple doses of rLH (Papanikolaou et al., 2011) and (iii) multiple doses of GnRH agonist (Pirard et al., 2006). The rationale of this approach is mild stimulation of the existing corpora lutea. On the other hand, in PCOS patients, a modified LPS involves intensified support, devoid of LH activity, with the use of intramuscular progesterone and estradiol in the form of oral tablets or transdermal patches (Babayof et al., 2006; Engmann et al., 2008). Apparently, the rationale for the latter approach is to provide adequate luteal support, without, however, stimulating the numerous existing corpora lutea in the high risk for OHSS patients, by providing LH activity. Based on the data presented by Humaidan et al. (2011), if the six aforementioned studies are combined, a significantly lower occurrence of OHSS by 7% [95% confidence interval (CI): 211 to 24] and a nonsignificantly lower delivery rate by 6% (95% CI: 214 to +2) is observed. An alternative approach to the analysis of these data, however, reveals a different picture that might have significant clinical implications. By pooling the data from the four studies (Humaidan et al., 2006, 2010; Pirard, et al., 2006; Papanikolaou et al., 2011) comparing normo-ovulatory patients, who were either triggered by GnRH agonist and received LH activity supplementation during the LP or triggered by hCG and received standard LPS, the significantly lower incidence of OHSS is no longer present [rate difference (RD): 22%, 95% CI: 24 to +1]. At the same time, a non-significant decrease in delivery rates (27%, 95% CI: 216 to +2) is detected in the patients who received LH activity supplementation after GnRH agonist triggering when compared with those triggered by hCG receiving standard LPS. Thus, in normo-ovulatory patients, considered to be at a normal risk for OHSS, GnRH agonist triggering combined with LH activity supplementation during the luteal phase does not significantly decrease the incidence of OHSS. On the other hand, regarding PCOS patients, when GnRH agonist triggering combined with intensified luteal support, devoid of LH activity, is compared with hCG triggering followed by a standard LPS, a significantly lower OHSS rate is present in the GnRH agonist group (RD: 231%, 95% CI: 246 to 217). At the same time, delivery rates between the two groups compared are not significantly different (RD: +1%, 95% CI: 218 to +19). Thus, it seems that in women at a high risk for OHSS, such as those with PCOS, GnRH agonist combined with a modified LPS devoid of LH activity significantly decreases the incidence of OHSS. Based on the above analysis, it cannot yet be concluded that ‘GnRHa triggering is a valid alternative to hCG triggering’ (Humaidan et al., 2011), although it is apparent that GnRH agonist administration when combined with modified LPS activity does achieve OHSS prevention in high-risk women.