The investigation of ovarian function by measurement of urinary oestrogen and pregnanediol excretion.

Urinary oestrogen and pregnanediol excretion was measued daily ("daily monitoring") for a complete cycle in 20 normally menstruating women, in one patient with an anovulatory cycle and for 28 days in a patient with secondary amenorrhoea. The measurements were also performed on urine specimens collected at weekly intervals for 4 to 6 weeks ("weekly tracking") from 506 patients with evidence of abnormal ovarian function. These included 9 patients with primary amenorrhoea, 132 patients with secondary amenorrheoa, 138 patients with oligomenorrhoea and 227 patients with evidence of ovarian dysfunction and cycle lengths of 25 to 42 days. The results were subjected to statistical analysis. In the normal cycles, ovulation could be identified on the criteria of a rising pregnanediol value reaching or exceeding 2-0 mg. per 24 hours for a period of 7 days or more. Valid conclusions on the overall mean oestrogen and pregnanediol values for a complete cycle could be made from the results of weekly tracking, irrespective of which day the tracking commenced. Correlations were obtained by comparing the mean and maximum urinary oestrogen values and the variability of the values with the evidence of ovarian function indicated by the clinical classifications of the patients, the duration of the disorders and the subsequent occurrence of uterine bleeding. Mean oestrogen values of 10 mug. per 24 hours or less were associated with lack of ovarian function. For values higher than this a discriminant function based on both the mean oestrogen value and the variability of the oestrogen values was useful in predicting onset of spontaneous menstruation. A single urine specimen collected 4 to 8 days before onset of menstruation showing a raised pregnanediol value of 2-0 mg. per 24 hours or more provided a valid test for ovulation in women with regular cycles, and a single urine specimen giving an oestrogen value of 10 mug per 24 hours or less gave a valid indication of absent ovarian function in women with amenorrhoea for two years or more. In all other circumstances serial sampling at weekly intervals provided a valid assessment of ovarian activity. Application of these principles allows the greatest amount of information on ovarian function to be obtained with the greatest economy of effort.