Rational approach to categorizing testosterone levels using community-based reference ranges
Reference ranges for testosterone in men generated using liquid chromatography tandem mass spectrometry in a community-based sample of healthy nonobese young men in the Framingham Heart Study and applied to three geographically distinct cohorts. Bhasin S, Pencina M, Jasuja GK, et al. J Clin Endocrinol Metab 2011;96(8):2430−2439.
This study1 generated reference limits for total and free testosterone levels in a community-based sample of nonobese healthy young (19-40 years) men enrolled in the Framingham Heart Study third generation cohort.2 These reference limits were then applied to three geographically distinct cohorts of community dwelling men drawn from the Framingham Heart Study (FHS) generations 2 and 3,2 the European Male Aging Study (EMAS)3,4 and the Osteoporotic Fractures in Men Study (MrOS).5 A ‘gold standard’ assay method, liquid chromatography tandem mass spectrometry (LC-MS/MS)6 was used throughout to determine total testosterone levels; free testosterone levels were calculated using a published law-of-mass-action equation.7
Researchers then investigated whether men deemed to have low total and free testosterone levels by the proposed reference limits had a higher prevalence of physical dysfunction, sexual symptoms, and diabetes mellitus, the three categories of conditions most consistently associated with low testosterone levels. Physical function measures (including low walking speed, difficulty climbing stairs, self-reported mobility limitation and frailty) and diabetes were investigated in all three cohorts; sexual symptoms (including decreased morning erections, erectile dysfunction and decreased frequency of sexual thoughts) were available only in EMAS.
What is known
The accurate diagnosis of androgen deficiency is based on the determination of whether circulating testosterone levels are low or normal.8-10 However, approaches to generate rigorous testosterone reference ranges have been limited by the inherent selection bias when hospital- or clinic-based patient samples have been used, and by the accuracy of the assay methods. Thus, the categorization of total and free testosterone levels into normal or low values has been associated with a substantial risk of misclassification.
What this study adds
This study has several strengths. It used a ‘gold standard’ assay method with high specificity, sensitivity and accuracy to measure total testosterone levels in a general population. The FHS reference cohort is a large community-based sample of healthy men, both young and older, approaching the definition of an optimum sample as described by the International Federation of Clinical Chemistry.11-13 The researchers then showed that, consistently across the three geographically distinct samples, men with low total and free testosterone levels, according to these thresholds, had a higher prevalence of the three categories of conditions consistently associated with low testosterone levels: physical dysfunction, sexual symptoms, and diabetes.
Prospective randomized trials and incident outcome studies will be necessary to determine how well the discriminating thresholds identified in this study can be applied to the clinical diagnosis of testosterone deficiency.