In the present paper we examined, whether the combined GnRH-agonist/hMG therapy implies an increased risk of the ovarian hyperstimulation syndrome (OHS). In a retrospective analysis, 525 GnRH-a/hMG cycles were compared with 643 cycles of hMG stimulation, which were simultaneously performed at the Department of Gynecology and Obstetrics of the University of Hamburg. Two different GnRH-agonists were used: Buserelin (Hoechst) given intranasally (410 cycles) and Triptorelin (Ferring) intramuscularly (115 cycles). The clinical results of hMG "only"-therapy revealed an OHS incidence of 7% for grade II and 0.2% for grade III. In contrast, significantly higher incidences were observed after GnRH-a/hMG treatment. In Buserelin/hMG cycles in 23% OHS grade II and in 1.0% OHS grade III occurred, in Triptorelin/hMG cycles in 40% OHS II and in 5.2% OHS III, respectively. The increased incidence of OHS correlated with higher ovarian estrogen production as well as a higher number of follicles following the GnRH-a/hMG stimulation. Furthermore, in GnRH-a/hMG cycles a prolonged duration of follicular maturation occurred due to an increase of the active phase; in addition the amount of hMG-ampoules needed for ovarian stimulation was higher. After GnRH-a/hMG treatment, an endogenous LH-surge was not detected, whereas in 34% of hMG stimulated cycles irregular LH-fluctuations were observed. There was a higher pregnancy rate in GnRH-a/hMG cycles (15%/525 cycles), as compared to hMG stimulation (8%/643 cycles), but the abortion rate was similar (23%, GnRH-a/hMG, versus 13%, hMG). The demonstration of an increased ovarian response leading to better pregnancy rates but also higher risks of OHS is well known from earlier data of hMG stimulation in patients with hypogonadotropic amenorrhoea (WHO group I). This implies that GnRH-agonist pre-treatment shows similar endocrine conditions in normogonadotropic patients.