Indices of ovulation: comparison of plasma and salivary levels of progesterone with urinary pregnanediol.

An increased daily excretion of pregnanediol, relative to that early in the menstrual cycle, is often taken to be evidence that a woman has ovulated. This paper assesses the value of alternative procedures for this purpose. Urine, plasma and saliva samples were collected during a 24-h period from 20 women during the follicular phase and from 20 women during the luteal phase. The 24-h excretion of pregnanediol was compared with (1) the concentration of progesterone in plasma, (2) the concentration of progesterone in saliva, (3) the concentration of pregnanediol in small urine samples, (4) the rate of excretion of pregnanediol and (5) the ratio of pregnanediol to creatinine in small urine samples. Each analyte increased substantially during the luteal phase. The median increases (ratio of luteal to follicular phase values) were 14.8, 3.2, 10.6, 11.9 and 11.1 respectively. By comparison, the median increase in 24-h pregnanediol output was 9.2. When the other analytes were used instead of the 24-h excretion of pregnanediol to assess the possibility of ovulation, the incidence of misclassifications (follicular samples classed as luteal and luteal samples classed as follicular) was 0, 12.8, 5.9, 2.0 and 1.0% respectively. It was concluded that the most satisfactory alternative to the measurement of 24-h pregnanediol output for the biochemical assessment of ovulation based on progesterone production was the measurement of the concentration of progesterone in plasma; the least satisfactory alternative was determination of the concentration of progesterone in saliva. If blood was not available, measurement of the ratio of pregnanediol to creatinine in a small urine sample was the preferred method.