Objective: To determine short-term pituitary and ovarian hormonal responses to GnRH agonist (GnRH-a) administered during various phases of the menstrual cycle, in the absence of controlled ovarian hyperstimulation (COH), to determine its independent effect on hormonal parameters previously demonstrated to influence assisted reproductive technology cycle outcome.

Design: Prospective, randomized, controlled crossover study of five regularly cycling women. The GnRH-a, leuprolide acetate (LA), was administered 1 mg SC daily for 5 days beginning on cycle day 3 (early follicular); 8 days post-LH surge (midluteal); or 13 days post-LH surge (late-luteal).

Setting: Clinical research unit at a tertiary care medical center.

Main outcome measures: Serum gonadotropins (LH and FSH) and gonadal steroids (E2, estrone [E1], P, A, and T) measured daily during GnRH-a administration begun in the early follicular, midluteal, or late luteal phase of the menstrual cycle. Gonadotropin pulse amplitude and frequency were determined after frequent serum sampling on the 2nd day of GnRH-a administration in each treatment cycle.

Results: Serum LH elevations, 4- to 10-fold greater than observed for FSH, did not differ by cycle day of GnRH-a initiation. Initial increases in FSH did not differ by cycle day, however, early follicular initiation resulted in a more pronounced suppression of FSH. Mean LH pulse amplitude and frequency increased to a similar extent in all three groups, however, FSH pulse amplitude and frequency varied significantly by cycle day of GnRH-a initiation. Gonadotropin-releasing hormone agonist initiated in the early follicular phase resulted in significant increases in E2, E1, and P levels compared with both midluteal or late luteal. Increases in serum androgens were significantly greater after early follicular and late luteal initiation as compared with midluteal GnRH-a initiation.

Conclusions: Relative FSH suppression and marked androgen elevations in both late luteal and early follicular groups, which may have potential detrimental effects on oocytes of the developing cohort, suggest little advantage of late luteal or early follicular over midluteal initiation of GnRH-a for COH.