The luteal phase of cycles stimulated with human menopausal gonadotropins (hMG) may be characterized by aberrant hormone levels, altered endometrial development, and shortened length. Luteal phase support with supplemental progesterone or hCG has been recommended to help correct these problems and thus improve pregnancy rates, but the efficacy of such regimens is controversial. Therefore, a randomized cross-over study was performed to evaluate the effects of luteal phase hCG administration on pregnancy rates during ovulation induction with hMG. Sixty-seven infertile women were randomly assigned to either group A (N = 33) or group B (N = 34). Non-treatment cycles (no luteal phase support) were alternated with treatment cycles, in which patients received 2500 IU hCG on the third, sixth, and ninth days after the ovulatory dose of 10,000 IU hCG. Patients in group A received supplemental hCG in odd-numbered cycles, whereas group B was given luteal support in even-numbered cycles. The mean number of cycles per patient was 2.2 and 2.3 for groups A and B, respectively. Analysis of 151 cycles revealed a cycle fecundity of 0.15 for 72 hCG-supported cycles, versus 0.13 for 79 nonsupported cycles (P = not significant). Midluteal progesterone levels were significantly higher in supported (45.6 ng/mL) versus unsupported cycles (31.9 ng/mL) (P < .001). There were no significant differences in the mean peak estradiol levels in hCG-supported versus-unsupported cycles. We conclude that hCG support of the luteal phase is not routinely warranted in hMG-stimulated cycles.
(C) 1992 The American College of Obstetricians and Gynecologists