Assessment of luteal function with progesterone enzyme immunoassays in the horse mare

Progesterone (P(4)) concentrations in the horse mare have conventionally been measured by radioimmunoassay (RIA). A commercial quantitative, competitive enzymelinked immunoassay (CELISA) has been recently introduced to the horse breeding industry along with a qualitative CELISA. The objectives of this study were to evaluate these two assays for their ability to detect transitional, cyclic, and pregnancy stages as well as early embryonic death in horse mares. The quantitative CELISA demonstrated a significant difference in P(4) levels during estrus and at Day 24 of pregnancy. There was a significant difference in P(4) concentrations in mares at ovulation and at Days 17 to 18 of pregnancy. However, there was no significant difference between pregnant and nonpregnant P(4) values from Day 1 until Day 16 of pregnancy. There was a significant difference in P(4) concentrations at ovulation and at the peak of the luteal phase in nonpregnant mares. There was also a significant difference between mares in the transitional period and Day 16 pregnant mares. The intraassay coefficient of variation was 3.1%, while the interassay coefficient of variation was 6.2%. The qualitative CELISA provided true positive diagnoses of 76.5% of the time, true negatives 67.3% of the time, false positives 4.1% of the time, and false negatives 3.1% of the time. The test sensitivity was 89.3% and its specificity was 95.7%. The CELISA provided 93% valid positive diagnoses of pregnancy on Day 21, 99% on Day 26, and 100% on Day 30. The use of either of these assay systems will provide a useful adjunct to any breeding program that utilizes one or more of the currently accepted diagnostic techniques, including teasing, palpation, and ultrasound. These assays introduce a new, inexpensive method of confirming the reproductive status in the mare.