Postpartum return of ovarian activity in nonbreastfeeding women monitored by urinary assays

Ronald H. Gray, Oona M. Campbell, Howard A. Zacur, Miriam H. Labbok, Susan L. MacRae

Research output: Contribution to journalArticlepeer-review


The present authors currently are studying 95 breastfeeding women to find predictors of ovulation. In preparation for these studies, they investigated the return of ovarian activity in 22 postpartum nonbreast-feeding women, using steroid and LH assays on daily early morning urine samples and twice weekly blood samples as measures of endocrine status. In addition, they observed steroid excretion in 16 regularly menstruating nonpostpartum women to establish the normal range of values. The 16 regularly menstruating women ranged in age from 21 to 44 years and had an average of 0 to 2 children. The 22 nonbreast-feeding women had a mean age of 24.9 years and a mean parity of 2.6 children. Urine samples were collected daily, and blood samples were collected twice a week. Pregnanediol 3α-glucuronide (PdG) and estradiol-17β-glucuronide (E2G) were measured in unex-tracted urine. The occurrence of ovulation was inferred from a rise in urinary LH and/or an increase in urinary PdG greater than 2 SD from baseline and a reversal of the E2G:PdG ratio. Two statistical indices were used to define the normal limits of PdG excretion during the luteal phase: 1) the mean peak level of the PdG rise was defined as the highest smoothed daily luteal phase PdG values, and the lower limit of normal was defined as the lowest 10th percentile; and 2) the area under the PdG curve was used to measure overall luteal phase PdG excretion. The 16 regularly menstruating nonpostpartum women had ovulatory cycles and thus provided data to establish normal values for luteal phase PdG excretion. Median peak PdG excretion during the luteal phase was 8.13 μg/ml, and the lower 10th percentile was 4.0 μg/ml. The median area under the luteal phase PdG curve was 36 μg/ml, and the lower 10th percentile was 20 μg/ml. Therefore, the lower limit of normal for luteal phase PdG excretion was set at 4.0 μg/ml for the mean peak level and at 20 μg/ml for the area under the curve. The 22 nonbreast-feeding women contributed information on 49 menstrual episodes, which were classified as ovulatory on the basis of a significant rise in LH and/or PdG and a reversal of the E2G:PdG ratio. The luteal phase was classified as normal if all measures of luteal phase PdG excretion were within the normal ranges, and luteal phase length was greater than 8 days; as abnormal if PdG excretion was less than the lower limit of normal, or luteal phase length was 8 days or less; and as equivocal if there were contradictions between the criteria for PdG excretion. Thirty-two per cent of the postpartum women had no ovulation before their first postpartum menstrual episodes, whereas only 15 per cent of the second and subsequent menstrual episodes were anovulatory. The mean time from delivery to first menses was 45.0 ± 10.1 (SD) days, and the mean time from parturition until first inferred ovulation was 45.2 ± 16.9 days. This reflects the high proportion of anovulatory first menses. No woman ovulated before 25 days after delivery. Eleven of the 15 ovulatory first postpartum cycles had clearly abnormal luteal phases (7.3 per cent), as compared to six of 23 ovulatory second or subsequent cycles (26.1 per cent). The three equivocal luteal phases had mean PdG excretion within the normal range, but a marginal area under the PdG curve. These data indicate an excess of abnormal first cycles postpartum, and if all abnormalities are considered, 81.8 per cent (18 of 22) of first postpartum cycles and 37.0 per cent (10 of 27) of second or subsequent cycles had unequivocal evidence of disturbance in ovulation, luteal phase PdG excretion, and/or luteal phase length. Although two women failed to ovulate throughout the study, there was no propensity for individual women to have repeated abnormal cycles. A high proportion (50.0 per cent) of abnormal luteal phases were deficient in PdG excretion but of normal length, whereas one-quarter of the abnormal phases were deficient in PdG excretion and of short duration. These distributions were similar for first and subsequent ovulatory cycles. Among the normal luteal phase cycles, there was a progressive increase in PdG excretion from the first to the third postpartum cycles. The mean peak smoothed PdG value for the normal postpartum cycles was 5.64 ± 0.14 μg/ml, which is significantly lower than the peak values for the second and third normal postpartum cycles. Such a progressive rise in PdG excretion was not found in the abnormal luteal phase cycles. The correlation coefficients between serum and urinary concentrations are tabulated by lag time between blood and urine collections in Table 1. All correlation coefficients were highly significant (P < 0.001). The coefficients were highest for progesterone to PdG, and least satisfactory for serum and urinary LH. The coefficients for progesterone to PdG were maximum for the 1- and 2-day lag periods between serum and urine collections, but the estradiol to E2G correlation was highest on the same day and the 1-day lag. These differences in lag times, however, are not significant. The lower correlation coefficients for LH result from failure to obtain serum samples at the time of the LH peak due to the twice weekly venipuncture regimen.

Original languageEnglish (US)
Pages (from-to)620-621
Number of pages2
JournalObstetrical and Gynecological Survey
Issue number10
StatePublished - Oct 1987

ASJC Scopus subject areas

  • Obstetrics and Gynecology


Dive into the research topics of 'Postpartum return of ovarian activity in nonbreastfeeding women monitored by urinary assays'. Together they form a unique fingerprint.

Cite this