Refining the moose serum progesterone threshold to diagnose pregnancy

• Published in: Conservation physiology Vol. 11; no. 1; p. coad003
• Main Authors: Struck, Madeline, Severud, William J, Chenaux-Ibrahim, Yvette M, J. Isaac, Edmund, Brown, Janine L, Moore, Seth A, Wolf, Tiffany M
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 2023
• Subjects: serum progesterone; moose; receiver operator characteristic curve; pregnancy; Alces alces
• ISSN: 2051-1434; 2051-1434
• DOI: 10.1093/conphys/coad003

We refined the threshold of serum progesterone concentrations that indicated pregnancy in moose using samples from collared moose with known pregnancy status in Grand Portage Indian Reservation in northeastern Minnesota, USA. This refinement increased historical population-level pregnancy rates from an average of 81% to 91% of adult female moose. Abstract Pregnancy determination is necessary for sound wildlife management and understanding population dynamics. Pregnancy rates are sensitive to environmental and physiological factors and may indicate the overall trajectory of a population. Pregnancy can be assessed through direct methods (rectal palpation, sonography) or indicated using hormonal assays (serum progesterone or pregnancy-specific protein B, fecal progestogen metabolites). A commonly used threshold of 2 ng/ml of progesterone in serum has been used by moose biologists to indicate pregnancy but has not been rigorously investigated. To refine this threshold, we examined the relationship between progesterone concentrations in serum samples and pregnancy in 87 moose (Alces alces; 64 female, 23 male) captured from 2010 to 2020 in the Grand Portage Indian Reservation in northeastern Minnesota, USA. Pregnancy was confirmed via rectal palpation (n = 25), necropsy (n = 2), calf observation (n = 25) or characteristic pre-calving behavior (n = 6), with a total of 58 females determined pregnant and 6 not pregnant; 23 males were included to increase the non-pregnant sample size. Using receiver operating characteristic analysis, we identified an optimal threshold of 1.115 ng/ml with a specificity of 0.97 (95% confidence interval [CI] = 0.90–1.00) and a sensitivity of 0.98 (95% CI = 0.95–1.00). Progesterone concentrations were significantly higher in cases of pregnant versus non-pregnant cows, but we did not detect a difference between single and twin births. We applied our newly refined threshold to calculate annual pregnancy rates for all female moose (n = 133) captured in Grand Portage from 2010 to 2021. Mean pregnancy rate during this period was 91% and ranged annually from 69.2 to 100%. Developing a reliable method for determining pregnancy status via serum progesterone analyses will allow wildlife managers to assess pregnancy rates of moose without devoting substantial time and resources to palpation and calf monitoring.