Novel reference range values for serum testosterone: a cross-sectional study of 200,000 males

Purpose To better understand the effects of aging, metabolic syndrome, diurnal variation, and seasonal variation on serum testosterone levels in the context of current guideline statements on testosterone deficiency. Methods This cross-sectional study utilized the United Kingdom Biobank. Physical examination, anthropomorphic measurements, and laboratory evaluation were performed at the time of enrollment from 2006 to 2010. The primary outcomes were the effect of age, the presence of metabolic syndrome, the time of day, and the month of the year on serum testosterone levels. Results Among 197,883 included men, the 5th, 25th, 50th, 75th and 95th percentile testosterone levels in men without metabolic syndrome were significantly higher than those in men with metabolic syndrome at every decade of life (p < 0.001). The average testosterone level within each group (men without metabolic syndrome vs. men with) was clinically similar across decade of life (12.43 in 40's 12.29 in 50's 12.24 in 60's vs. 10.69 in 40's 10.56 in 50's 10.63 in 60's respectively). Average testosterone levels decreased with blood draws later in the day ranging from 10.91 to 12.74 nmol/L (p < 0.01). Similarly, there was seasonal variation in serum testosterone ranging from 11.86 to 12.18 nmol/L (p < 0.01). Conclusions We found significant variation in serum testosterone according to the presence of metabolic syndrome and time of laboratory draw, but not according to age. These data challenge the prior dogma of age-related hypogonadism and favor an individualized approach towards serum testosterone measurement and interpretation. However, further studies are needed to correlate these population-based data with individuals' hypogonadal symptoms.