Objective: Compare the pharmacokinetics of vaginal progesterone gel (Crinone 8%, 90 mg) with that of oral progesterone (Prometrium, 100 mg).Design: Open-label, randomized, parallel-group protocol.Setting: Outpatient clinic.Patient(s): Twelve healthy postmenopausal women.Intervention(s): Six subjects each were randomized to receive progesterone, which was administered either as 90 mg of progesterone gel (Crinone 8%) given vaginally or 100 mg progesterone in a capsule (Prometrium) given orally.Main Outcome Measurement(s): Serum progesterone levels were measured by both radioimmunoassay (RIA) and liquid chromatography-mass spectrometry (LC-MS).Result(s): Progesterone given vaginally resulted in greater bioavailability with less relative variability in absorption than oral progesterone (mean AUC 0 - 2 4 = 1.48 +/- 0.16 ng . h/mL per milligram vs. 0.035 +/- 0.0052 ng . h/mL per milligram). Mean C m a x for oral progesterone was much lower than that of vaginal progesterone (i.e., 2.20 +/- 3.06 ng/mL vs. 10.51 +/- 0.46 ng/mL). Mean T m a x occurred earlier for oral progesterone than for Crinone (1.00 +/- 0.41 hours vs. 7.67 +/- 3.67 hours). Radioimmunoassay is inappropriate for determining serum progesterone levels after oral administration, because it provided erroneously high values that were approximately eightfold higher than those obtained with LC-MS.Conclusion(s): Crinone (progesterone gel) given vaginally results in greater bioavailability with less relative variability than oral progesterone, thus providing more reliable delivery of progesterone, compared with oral progesterone. Measuring circulating progesterone with use of direct RIA is not appropriate after oral progesterone administration.