madman
Super Moderator
* Although these epidemiologic data do not prove a causal relationship, they remind us that sex steroid hormones have complex actions and interactions in human health.
Bradley D. Anawalt, MD
University of Washington School of Medicine, Seattle, Washington
A raging debate was settled last year. Or was it?
There has been controversy about whether testosterone therapy to raise serum testosterone concentrations into the normal range increases risk for stroke, myocardial infarction, cardiovascular death, and all-cause mortality among men with low serum testosterone concentrations at baseline (1). The TRAVERSE (Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy ResponSE in Hypogonadal Men) trial was the first randomized, placebo-controlled study that was designed and powered to determine whether testosterone therapy increased risk for major cardiovascular events in men (aged 45 to 80 years in this trial). Published in 2023, this 4-year study of testosterone gel treatment showed no difference in the primary end point, a composite of nonfatal myocardial infarction, nonfatal stroke, or death due to cardiovascular causes (2). After a mean follow-up of 22 months, an adjudicated major cardiovascular event occurred in 7.0% of men in the testosterone group versus 7.3% in the placebo group. Among the testosterone treated men, 3.4% and 5.5% died of cardiovascular causes or any cause, respectively, compared with 4.0% and 5.7% in the placebo group. In TRAVERSE, the testosterone gel dosage was adjusted to maintain serum total testosterone concentrations between 350 and 750 ng/dL (normal range, 264 to 916 ng/dL); median serum testosterone concentrations increased from about 230 to about 375 ng/dL during the treatment phase, and median serum estradiol concentrations increased modestly, from about 20 pg/mL at baseline to about 30 pg/mL during treatment. Serum dihydrotestosterone (DHT) data were not reported. The results suggested that, over the short term, modest-dose testosterone replacement does not increase risk for major cardiovascular events, cardiovascular death, or all-cause mortality.
Although the TRAVERSE findings were reassuring, several questions about the relationship between testosterone and risk for major cardiovascular events and death remain unanswered. These questions include whether high endogenous or exogenous serum testosterone concentrations or a threshold low baseline (untreated) serum testosterone concentration are associated with increased risk for cardiovascular events or mortality. The relationships of serum estradiol and DHT concentrations with cardiovascular risk and death are also unknown. These relationships are interesting because all testosterone formulations increase serum estradiol and DHT concentrations above baseline. Serum DHT increases more markedly with oral and transdermal testosterone formulations because the gut mucosa and skin contain significant amounts of 5a-reductase, the enzyme that converts testosterone to DHT.
In their article, Yeap and colleagues present a meta analysis of prospective cohort studies that suggests that a very low baseline (endogenous) serum testosterone concentration is associated with increased risk for cardiovascular death and all-cause mortality (3). To understand the value of this meta-analysis, it is useful to briefly review its rigorous methodology. In a previously published systematic review, the authors identified 11 prospective cohort studies that measured serum total testosterone (and other sex steroids) by mass spectrometry at baseline and had at least 5 years of follow-up for major cardiovascular events, cardiovascular death, and all-cause mortality (4). Mass spectrometry is considered the most accurate method of testosterone measurement and can also be used to measure DHT and estradiol accurately, whereas widely available commercial immunoassays are inaccurate for measurement of these sexsteroids in men, who typically have low serum concentrations of these 2 metabolites of testosterone.
The authors were able to obtain individual participant data (IPD) from 9 of these studies (which included>20 000 men) and aggregate data statistics from the other 2 studies (which included >3000 men). Aggregate data meta-analysis, the most common form of meta analysis, involves the statistical synthesis of data from similar studies based on published summary statistics. To perform IPD meta-analysis, the authors obtained the raw data from the 9 studies and then reanalyzed the combined data (5). This method permits more sophisticated analysis of combined data from multiple studies and allows more robust testing for associations between secondary variables (such as estradiol and DHT) and outcomes. The authors performed a 2-step IPD analysis. In step 1, the individual data of each of the 9 studies were analyzed separately to obtain aggregate (summary) data, and step 2 entailed an analysis of the combined aggregated data of the 9 studies followed by an analysis of the combined aggregated data of all 11 studies (6).
The primary finding was that high testosterone concentrations at baseline were not associated with increased cardiovascular death and all-cause mortality, but very low serum total testosterone concentrations at baseline were. The testosterone inflection point of around 210 ng/dL was below the baseline median and mean serum testosterone concentrations inTRAVERSE participants. It is tempting to hypothesize that testosterone therapy might have cardiovascular benefits solely in patients with very low concentrations of serum total testosterone.
There were additional provocative findings from secondary and exploratory analyses. First, men with very high serum luteinizing hormone concentrations at baseline (suggesting primary hypogonadism) had higher all-cause mortality. Second, men with low serum total testosterone and high sex hormone–binding globulin (SHBG) concentrations at baseline (suggesting very low unbound or free testosterone concentrations) had higher all-cause mortality, whereas men with low serum total testosterone and low SHBG concentrations at baseline (suggesting normal concentrations of serum free testosterone) had lower all-cause mortality. Third, men with very low serum estradiol concentrations at baseline had higher all-cause mortality. However, the relationship of DHT and outcomes was U-shaped: Men with low and high DHT concentrations had higher cardiovascular and all-cause mortality.
Overall, these epidemiologic data support the hypothesis that hypogonadism is associated with higher cardiovascular and/or all-cause mortality. The data also support the free testosterone hypothesis that states that unbound testosterone is the active form of the hormone—a hypothesis that is somewhat controversial (7, 8). In addition, the current meta-analysis suggests interesting hypotheses about the major hormonal metabolites of testosterone: estradiol and DHT. The finding that a low serum estradiol concentration is associated with higher all-cause mortality adds another reason (in addition to the adverse effects on body fat and bone health) to avoid aromatase inhibitors that are commonly taken by persons who use anabolic steroids (9, 10). The prospect of a U-shaped curve for the relationship between serum DHT and higher cardiovascular risk warrants further study. Although these epidemiologic data do not prove a causal relationship, they remind us that sex steroid hormones have complex actions and interactions in human health.
Bradley D. Anawalt, MD
University of Washington School of Medicine, Seattle, Washington
A raging debate was settled last year. Or was it?
There has been controversy about whether testosterone therapy to raise serum testosterone concentrations into the normal range increases risk for stroke, myocardial infarction, cardiovascular death, and all-cause mortality among men with low serum testosterone concentrations at baseline (1). The TRAVERSE (Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy ResponSE in Hypogonadal Men) trial was the first randomized, placebo-controlled study that was designed and powered to determine whether testosterone therapy increased risk for major cardiovascular events in men (aged 45 to 80 years in this trial). Published in 2023, this 4-year study of testosterone gel treatment showed no difference in the primary end point, a composite of nonfatal myocardial infarction, nonfatal stroke, or death due to cardiovascular causes (2). After a mean follow-up of 22 months, an adjudicated major cardiovascular event occurred in 7.0% of men in the testosterone group versus 7.3% in the placebo group. Among the testosterone treated men, 3.4% and 5.5% died of cardiovascular causes or any cause, respectively, compared with 4.0% and 5.7% in the placebo group. In TRAVERSE, the testosterone gel dosage was adjusted to maintain serum total testosterone concentrations between 350 and 750 ng/dL (normal range, 264 to 916 ng/dL); median serum testosterone concentrations increased from about 230 to about 375 ng/dL during the treatment phase, and median serum estradiol concentrations increased modestly, from about 20 pg/mL at baseline to about 30 pg/mL during treatment. Serum dihydrotestosterone (DHT) data were not reported. The results suggested that, over the short term, modest-dose testosterone replacement does not increase risk for major cardiovascular events, cardiovascular death, or all-cause mortality.
Although the TRAVERSE findings were reassuring, several questions about the relationship between testosterone and risk for major cardiovascular events and death remain unanswered. These questions include whether high endogenous or exogenous serum testosterone concentrations or a threshold low baseline (untreated) serum testosterone concentration are associated with increased risk for cardiovascular events or mortality. The relationships of serum estradiol and DHT concentrations with cardiovascular risk and death are also unknown. These relationships are interesting because all testosterone formulations increase serum estradiol and DHT concentrations above baseline. Serum DHT increases more markedly with oral and transdermal testosterone formulations because the gut mucosa and skin contain significant amounts of 5a-reductase, the enzyme that converts testosterone to DHT.
In their article, Yeap and colleagues present a meta analysis of prospective cohort studies that suggests that a very low baseline (endogenous) serum testosterone concentration is associated with increased risk for cardiovascular death and all-cause mortality (3). To understand the value of this meta-analysis, it is useful to briefly review its rigorous methodology. In a previously published systematic review, the authors identified 11 prospective cohort studies that measured serum total testosterone (and other sex steroids) by mass spectrometry at baseline and had at least 5 years of follow-up for major cardiovascular events, cardiovascular death, and all-cause mortality (4). Mass spectrometry is considered the most accurate method of testosterone measurement and can also be used to measure DHT and estradiol accurately, whereas widely available commercial immunoassays are inaccurate for measurement of these sexsteroids in men, who typically have low serum concentrations of these 2 metabolites of testosterone.
The authors were able to obtain individual participant data (IPD) from 9 of these studies (which included>20 000 men) and aggregate data statistics from the other 2 studies (which included >3000 men). Aggregate data meta-analysis, the most common form of meta analysis, involves the statistical synthesis of data from similar studies based on published summary statistics. To perform IPD meta-analysis, the authors obtained the raw data from the 9 studies and then reanalyzed the combined data (5). This method permits more sophisticated analysis of combined data from multiple studies and allows more robust testing for associations between secondary variables (such as estradiol and DHT) and outcomes. The authors performed a 2-step IPD analysis. In step 1, the individual data of each of the 9 studies were analyzed separately to obtain aggregate (summary) data, and step 2 entailed an analysis of the combined aggregated data of the 9 studies followed by an analysis of the combined aggregated data of all 11 studies (6).
The primary finding was that high testosterone concentrations at baseline were not associated with increased cardiovascular death and all-cause mortality, but very low serum total testosterone concentrations at baseline were. The testosterone inflection point of around 210 ng/dL was below the baseline median and mean serum testosterone concentrations inTRAVERSE participants. It is tempting to hypothesize that testosterone therapy might have cardiovascular benefits solely in patients with very low concentrations of serum total testosterone.
There were additional provocative findings from secondary and exploratory analyses. First, men with very high serum luteinizing hormone concentrations at baseline (suggesting primary hypogonadism) had higher all-cause mortality. Second, men with low serum total testosterone and high sex hormone–binding globulin (SHBG) concentrations at baseline (suggesting very low unbound or free testosterone concentrations) had higher all-cause mortality, whereas men with low serum total testosterone and low SHBG concentrations at baseline (suggesting normal concentrations of serum free testosterone) had lower all-cause mortality. Third, men with very low serum estradiol concentrations at baseline had higher all-cause mortality. However, the relationship of DHT and outcomes was U-shaped: Men with low and high DHT concentrations had higher cardiovascular and all-cause mortality.
Overall, these epidemiologic data support the hypothesis that hypogonadism is associated with higher cardiovascular and/or all-cause mortality. The data also support the free testosterone hypothesis that states that unbound testosterone is the active form of the hormone—a hypothesis that is somewhat controversial (7, 8). In addition, the current meta-analysis suggests interesting hypotheses about the major hormonal metabolites of testosterone: estradiol and DHT. The finding that a low serum estradiol concentration is associated with higher all-cause mortality adds another reason (in addition to the adverse effects on body fat and bone health) to avoid aromatase inhibitors that are commonly taken by persons who use anabolic steroids (9, 10). The prospect of a U-shaped curve for the relationship between serum DHT and higher cardiovascular risk warrants further study. Although these epidemiologic data do not prove a causal relationship, they remind us that sex steroid hormones have complex actions and interactions in human health.
