Presentation Authors: Richard Fantus*, Joshua Halpern, Chicago, IL, Cecilia Chang, Evanston, IL, Mary Keeter, Nelson Bennett, Chicago, IL, Brian Helfand, Evanston, IL, Robert Brannigan, Chicago, IL
Introduction: Millions of Americans each year trial new diets hoping to lose weight, gain strength or increase their energy level. While multiple studies have examined the benefits of low-fat and Mediterranean diets, the effects of these diet regimens on serum testosterone (T) are unknown. We hypothesized that men who restrict their dietary fat intake may be at increased risk for decreased serum T, a steroid hormone with cholesterol precursors.
Methods: The National Health and Nutrition Examination Survey (NHANES) database was queried from 1999-2004 and 2011-2012. Men 18-80 years who completed the two-day dietary history and underwent serum T testing were included. Diets were categorized as low-fat (American Heart Association), Mediterranean (inclusive of low-fat), or non-restrictive. Multivariable modeling was used to determine the relationship between diet and serum T.
Results: Among 7316 men who met inclusion criteria, 1160 (15.9%) met criteria for a low-fat diet, 1924 (26.3%) a Mediterranean diet, and 4920 (67.2%) a non-restrictive diet. Mean serum T was 435.5 Â± 6.7 ng/dL, and 838 men (26.7%) had serum T < 300 ng/dl. Compared to men with non-restrictive diets, mean T was lower among men with low-fat (410.8Â±8.1 vs 443.5Â±7.3, p=0.005) and Mediterranean diets (412.9Â±9.1 vs 443.5Â±7.3, p=0.002). Multivariable analysis controlling for age, body mass index (BMI), activity level, diabetes, comorbidities, and prostate cancer showed men with non-restrictive diets had higher serum T compared to those adhering to low-fat (parameter estimate[PE] -57.2, 95% confidence interval [CI] -105.6 - -8.8, p < 0.05) or Mediterranean diets (PE -26.2, 95% CI -61.9 - 9.6, p=0.15). Men adhering to a low-fat diet were more likely to have a T level < 300 ng/dl compared to their non-restrictive counterparts (odds ratio [OR] 2.33, 95% confidence interval [CI] 0.77-7.034, p=0.13), which approached but did not reach statistical significance.
Conclusions: In a nationally representative sample, men adhering to restrictive diets had lower serum T levels. This remained true even when controlling for common covariates such as comorbidities, age, BMI and activity levels. Future prospective research is required to both corroborate these findings and elucidate the mechanisms by which restrictive dieting may affect serum T.