Secondary Logo

Institutional members access full text with Ovid®

Share this article on:

Measurement of Cortisol and Testosterone in Athletes: Accuracy of Liquid Chromatography–Tandem Mass Spectrometry Assays for Cortisol and Testosterone Measurement in Whole-Blood Microspecimens

Fragala, Maren S.1; Goldman, Scott M.1; Goldman, Mildred M.1; Bi, Caixia1; Colletti, Julia D.; Arent, Shawn M.2; Walker, Alan J.2; Clarke, Nigel J.1

The Journal of Strength & Conditioning Research: September 2018 - Volume 32 - Issue 9 - p 2425–2434
doi: 10.1519/JSC.0000000000002726
Original Research

Fragala, MS, Goldman, SM, Goldman, MM, Bi, C, Colletti, JD, Arent, SM, Walker, AJ, and Clarke, NJ. Measurement of cortisol and testosterone in athletes: Accuracy of LC-MS/MS assays for cortisol and testosterone measurement in whole-blood microspecimens. J Strength Cond Res 32(9): 2425–2434, 2018—Biomarker monitoring provides insight into athletes' training tolerance but is limited by the need for office-based specimen collection. To facilitate self-collection during training, we developed liquid chromatography–tandem mass spectrometry-based tests that measure circulating total cortisol and testosterone using a finger stick volumetric absorptive microsampler. Here, we describe the analytical validation of these tests. Forty-six Division I athletes (18–22 years, 30 women, 16 men) provided a 20-µL finger stick microspecimen and a 5-ml venous blood specimen from the forearm; the venous blood sample was analyzed using both normal volume serum analysis and analysis of dried whole blood (from the microsampler). Liquid chromatography–tandem mass spectrometry on standard serum specimens obtained by venipuncture yielded total cortisol levels of 26.2 ± 11.6 µg·dl−1 (women and men), and total testosterone levels of 37 ± 17 ng·dl−1 in women and 564 ± 171 ng·dl−1 in men. Analytical measurement ranges of the microspecimen assay were 0.3–440 µg·dl−1 (CV <9%) for cortisol and 15 to 20,000 ng·dl−1 (CV <9%) for testosterone. Deming regression and Pearson correlation indicated good test accuracy for the microspecimen tests compared with venipuncture tests for cortisol (y = 0.98x + 1.34, 95% CI of slope = 0.83–1.14; r = 0.92, p < 0.0001) and testosterone (y = 1.06x − 0.01, 95% CI of slope = 0.99–1.14; r = 0.99, p < 0.0001). Similarly, high agreement was observed between finger stick and venous microspecimens for cortisol (y = 1.00x + 0.65, 95% CI of slope = 0.9–1.11; r = 0.96, p < 0.001) and testosterone (y = 0.97x + 2.75, 95% CI of slope = 0.9–1.03; r = 0.99, p < 0.001). These findings suggest the viability of finger stick collection whole-blood microspecimens for assessment of total cortisol and testosterone in athletes.

1Quest Diagnostics, Secaucus, New Jersey; and

2IFNH Center for Health & Human Performance, Rutgers, The State University of New Jersey, New Brunswick, New Jersey

Address correspondence to Dr. Maren S. Fragala, Maren.S.Fragala@QuestDiagnostics.

Copyright © 2018 by the National Strength & Conditioning Association.