A prospective randomized study was conducted to evaluate the use of adding oestradiol valerate 6 mg per os daily to intravaginal micronized progesterone (600 mg daily) as luteal supplements. The study comprised 378 infertile women superovulated with a gonadotrophin releasing-hormone agonist (GnRHa) and human menopausal gonadotrophins (HMG) for in-vitro fertilization (IVF) or zygote intra-Fallopian transfer (ZIFT). The clinical pregnancy rate was similar (29%) whether or not oestradiol valerate was added to intravaginal progesterone. Eighteen out of twenty-two endometrial biopsies were in phase, and morphological evaluations of the two luteal supplementation groups were not different. Serum hormone profiles in singleton pregnancies showed a similar day of appearance of human chorionic gonadotrophin (HCG) in both protocols but significantly lower oestradiol concentrations arose in the group without oestradiol valerate. In 32% of the singleton pregnancies, the first appearance of HCG occurred later than day 12 after HCG injection; in those ongoing pregnancies, corpus luteum rescue--as measured by significantly lower serum oestradiol and progesterone concentrations--was compromised. This study provided no evidence of any benefit of routinely supplementing GnRHa/HMG cycles with oestradiol valerate in addition to intravaginal micronized progesterone.