This retrospective pooled analysis indicates that the incidence of early and late LH rises prior to and during ganirelix treatment is low and comparison of the clinical outcome shows that, in contrast to women with an early LH rise, women with a late LH rise may have a reduced chance of ongoing pregnancy.
The incidence of LH rises on stimulation day 5 was significantly lower than on day 6 and the incidence during ganirelix treatment tended to be lower in subjects who started treatment on day 5 rather than on day 6, indicating that an earlier initiation of ganirelix treatment may further lower the incidence during ganirelix treatment as observed in this study. The limitation of the retrospective approach of this study is acknowledged.
Women with or without LH rises had the same average age; thus, it may be difficult to predict which women will have an increased risk for a premature LH rise. Prediction of ovarian response may be facilitated by pretreatment assessment of the antral follicle count and of anti-Müllerian hormone, both of which have been shown to be reliable predictors of ovarian response [16, 17]. When women start ovarian stimulation with a relatively high dose of rFSH they may subsequently present with more follicles and higher serum estradiol concentrations during the early follicular phase, thereby reaching the threshold for LH surge earlier during stimulation. In a previous study in which women were treated with a fixed starting dose of 225 IU FSH, employment of an algorithm of criteria to prevent premature LH surges resulted in GnRH antagonist initiation earlier than stimulation day 6 in two thirds of patients (138/208 initiated on days 4 or 5). This earlier start resulted in an improved clinical outcome .
In contrast to previous publications that have reported premature LH peaks in up to 22% of subjects during (mild) stimulation while treated with a GnRH antagonist [19–21], the current analysis indicates that LH rise prior to and during ganirelix treatment is a comparatively infrequent event in women aged up to 39 years with regular menstrual cycles undergoing ovarian stimulation prior to in vitro fertilization or intracytoplasmic sperm injection. In addition, the incidence of a concomitant rise in P was even lower, which for the early LH rises may be related to the low number of LH receptors during the first days of stimulation .
While women with an early LH rise prior to ganirelix treatment had higher numbers of oocytes retrieved compared with women without an LH rise, the impact of this early LH rise seems limited as the chance of ongoing pregnancy was not compromised. This may be explained by a possible short-lived pulse of LH rather than a precursor to luteinization and premature ovulation. In contrast, women with a late LH rise during ganirelix treatment had a lower ovarian response and had a lower chance of ongoing pregnancy, though the OR for ongoing pregnancy for women with an LH rise versus women without LH rise was not significant. Some of these LH rises may have been caused by drug noncompliance; however, these cannot be documented because ganirelix concentrations were not measured during these trials. Other LH rises may indeed be related to either premature luteinization or a low ovarian response, and the causes require further investigation.
While a daily dose of 0.25 mg ganirelix has been demonstrated to be an effective dose to prevent premature LH rises during multiple follicular development in in vitro fertilization cycles, there is some evidence that a GnRH antagonist is less effective in blocking the positive feedback effect of estradiol in women with mild or nonstimulated ovaries [23, 24]; this is in agreement with the lower ovarian response of women with an LH rise during ganirelix treatment in the current retrospective analysis. The cause of these LH rises remains to be elucidated but it has been suggested that a reduced production of gonadotropin surge-attenuating factor may be involved, which is hypersecreted in the case of induced multiple follicular development .
Late LH rises during GnRH antagonist treatment lower the chance of pregnancy and may be further reduced by starting antagonist treatment timely. These results support a flexible regimen of starting ganirelix treatment on day 5 or day 6 of stimulation, depending on the ovarian response.