The aim of controlled ovarian hyperstimulation (COH) is to induce multiple morphologically and functionally adequate follicles with the aim of harvesting multiple fertilizable oocytes. We compared several treatment regimens with different FSH/LH ratios: group I was the basic 2 human menopausal gonadotropins (2hMG) protocol in which 2 ampules of hMG were administered starting on day 3 of the cycle; group II was the basic 2hMG in which 2 ampules of "pure" follicle stimulating hormone (p-FSH) were added on day 3 and 4; group III was the 2 FSH protocol in which 2 ampules of p-FSH were administered beginning on day 3; finally, group IV was the 4FSH: 4 ampules of p-FSH on day 3 and 4 followed by 2 ampules of p-FSH daily. The reference regimen was the 2hMG. Except for the higher rate of immature oocytes in 2FSH and 2hMG protocols, the number of preovulatory oocytes and fertilization rate were similar in all protocols. No differences occurred in the pregnancy outcome. The low dose Lupron (LDL) stimulation was an experimental protocol applied to two groups of women who had previously failed COH. Eight women who had a premature luteinization and 8 women who showed a low response agreed to participate in the protocol. Ten micrograms of the GnRH agonist leuprolide acetate (Lupron) were injected subcutaneously every 6 hours starting on cycle day 2 and menotropin stimulation was begun on day 3-5, according to individual patient response. The LDL protocol was successful in determining a favourable estradiol pattern and fertilizable oocytes.