Is Men's Health Just a Single Letter?

Abstract

For several years, a marketing campaign to peddle T prescriptions has cleverly centered on the question: “Is it low T? Have an honest conversation with your doctor about all of the symptoms you are experiencing.” “Low T” has become a part of the popular vernacular to describe all sorts of symptoms that might (or might not) be associated with a low serum T level in men. T sales have surged in many countries (1). With the advent of the increased sales of T formulations to treat aging men, questions about the potential adverse effects of T therapy including cardiovascular disease and prostate cancer have emerged. There have been concerns that T might increase adverse cardiovascular events since at least the 1960s. The Coronary Drug Project was designed in 1965 to test the hypothesis that estrogen therapy might decrease recurrent cardiovascular ischemic events in men by raising highdensity lipoprotein cholesterol and lowering serum T levels (2). A high dosage of conjugated estrogen (enough to suppress the male gonadal axis) was associated with increased cardiovascular events (3). Later, a number of epidemiological studies demonstrating an association between cardiovascular disease and low serum T levels in men generated the opposing hypotheses that low T levels might increase the risk of cardiovascular events in men or that normal T might even be protective against cardiovascular events (4). Three recent epidemiological studies have fueled the debate on the effects of T on all-cause mortality and cardiovascular disease. Shores et al (5) reported in a retrospective database study of veterans with low serum T levels that T therapy was not associated with increased mortality or cardiovascular disease. Shores et al (5) reported lower mortality in the subset of men treated with T therapy than men who were not treated; this effect remained after adjustment for age, diabetes mellitus, and prevalent cardiovascular disease. In another retrospective database study of veterans, Vigen et al (6) reported an association between T therapy and an increase in the combined incidence of myocardial infarction, stroke, and overall mortality in men who underwent coronary angiography for evaluation of chest pain and were subsequently found to have low serum total T levels. In the Vigen et al (6) study of veterans, the unadjusted absolute rate of the combined endpoint (myocardial infarction, stroke, and death) was 21% for the untreated men and 10% for the T-treated men. After analysis with a sophisticated adjustment model, this pattern reversed with a lower rate of the combined endpoint (19.9%) in the untreated group compared to the T-treated group (25.7%). A third important epidemiological study on the association between T, cardiovascular disease, and overall mortality appears in this issue of the JCEM. Yeap et al (7) suggest that there is a U-shaped association between serum total or calculated free serum T levels and all-cause mortality, with a similar trend between T levels and cardiovascular mortality. In the Yeap et al study (7), over 4000 Western Australian older men ( 70 years of age) who were a subset of a larger cohort study had total T, dihydrotestosterone, and estradiol levels measured on an early morning blood sample with a highly accurate liquid chromatography-tandem mass spectrometry assay. Free T levels were calculated using a validated empiric formula. The men were followed for an average of 7 years. The primary outcomes were death of any cause and death due to ischemic heart disease as reported to the Western Australia Database Linkage System (WADLS), an intricate, publicly funded database that captures health outcomes including deaths of citizens of Western Australia. Because all residents of Western