The aim of our study was to investigate the correlation between serum malondialdehyde levels and serum estradiol concentrations in healthy human female subjects. Nine hundred and fifty-five blood samples, from infertile women undergoing controlled ovarian hyperstimulation treatment with recombinant follicle-stimulating hormone, were collected for estradiol and malondialdehyde measurements. Five groups were formed according to serum estradiol levels: Group I (<50 pg/ml), group II (50-299 pg/ml), group III (300-999pg/ml), group IV (1000-1999pg/ml) and group V (greater than or equal to2000 pg/ml). One-way analysis of variance was used for comparisons. Mean malondialdehyde concentrations were 1.74 +/- 0.24mmol/ml (group I), 1.53 +/- 0.20mmol/ml (group II), 1.69 +/- 0.24mmol/ml (group III), 1.77 +/- 0.21 mmol/ml (group IV) and 1.86 +/- 0.20 mmol/ml (group V), respectively. Mean serum malondialdehyde level at physiological estradiol concentrations (50-199pg/ml, group II) was significantly (p<0.01) lower than the mean malondialdehyde levels in other groups. Mean malondialdehyde concentrations among the remaining groups did not significantly differ. Our findings suggest that in vivo lipid peroxidation might be increased when circulating estradiol concentrations are below (<50 pg/ml) or above (>300 pg/ml) the physiological limits. High blood estradiol levels in human female subjects during ovarian stimulation with exogenous gonadotropins could be associated with increased serum malondialdehyde concentrations.