Reproducibility of plasma and urinary sex hormone levels in premenopausal women over a one-year period.

Although endogenous sex steroid hormones in premenopausal women may be associated with the risk of breast cancer and other illnesses, direct evidence to support this hypothesis is limited in large part by methodological issues in the conduct of relevant studies. One major unresolved issue is whether a single blood sample (such as is available in most epidemiological studies), collected in a specific phase of the menstrual cycle, reflects long-term levels in that phase. To address this issue, two sets of blood and urine samples were obtained from 87 premenopausal women over a 1-year period in both the follicular and luteal phases. Plasma estradiol, estrone, and estrone sulfate were measured in the blood samples obtained in both phases, whereas progesterone and urinary 2- and 16a-hydroxyestrone were measured in luteal-phase samples only. For all of the women combined, intraclass correlation coefficients (ICCs) ranged, with one exception, from 0.52 to 0.71 for the plasma estrogens and the urinary estrogen metabolites. The sole exception was for estradiol in the luteal phase (ICC = 0.19); inclusion of only women who were ovulatory in both cycles and who collected each sample 4-10 days before their next period resulted in a substantially higher ICC for estradiol in the luteal phase (ICC = 0.62; 95% confidence interval, 0.43-0.78). These data indicate that, for several plasma and urinary sex hormones, a single follicular- or luteal-phase measurement in premenopausal women is reasonably representative of hormone levels in that phase for at least a 1-year period.