The aim of this study was to establish the pharmacokinetic profile of serum oestriol (E3 ) concentrations over 24 h following application of vaginal E3 in chronic users (>12 weeks of E3 use). The interindividual and intraindividual differences before and after E3 were examined. Ten women participated. Vaginal cream was omitted for ≥36 h prior to the study days. Blood sampling was performed for E3 , oestradiol and oestrone concentrations prior to cream application and at 1, 2, 3, 5, 8, 10, 12 and 24 h afterwards. In five women, all samples were repeated on a separate day. E3 was absorbed rapidly in most women. Peak serum E3 concentration occurred around 2 h (range 1-5 h). The decline in E3 concentrations was also rapid: falling <100 pmol L-1 in six out of ten women within 8 h and returning to ≤ 10 pmol L-1 at 24 h in nine out of the ten patients. Interindividual variability for peak concentrations was considerable (mean 546 pmol L-1 ; 95% CI 349-743). Area under the concentration-time curve (AUC) values over a dosage interval also varied widely: mean 2145 pmol.h L-1 ; 95% CI 1422-3233. However, repeated measurements in the same woman were highly (peaks: ρ = 0.94) or moderately (AUC: P = 0.74) correlated. Postmenopausal E3 concentrations are negligible. Serum E3 concentrations of chronic users of E3 cream varied greatly; however, concentrations declined rapidly within 8 h, generally reaching 'postmenopausal' levels by 24 h. The basis for the variation between subjects needs further elucidation. Additional research is required to establish the safety of topical E3 .