What is known
Previously, four retrospective studies have investigated the effect of testosterone therapy on various cardiovascular disease outcomes and mortality; two reported negative effects3,4 while the other two reported beneficial effects.5,6 Two separate retrospective studies of men in the Veterans Affairs Health System using two different databases reported opposite effects of testosterone therapy on all-cause mortality.4,6
Multiple observational studies show that low testosterone levels are associated with an increased number of cardiovascular events.1 Clinical trials examining testosterone therapy have been relatively small and of short duration, and these trials were underpowered to provide conclusive evidence related to cardiovascular events.7 Thus, more studies are needed on the effects of testosterone treatment on cardiovascular disease outcomes.
About this study
The European Heart Journal study retrospectively examined 83,010 male veterans without prior MI or stroke, but with documented low total testosterone levels. The subjects were categorized into 3 groups:
Group 1: Testosterone therapy achieving normalization of total testosterone levels.
Group 2: Testosterone therapy not achieving normalization of total testosterone levels.
Group 3: Did not receive testosterone therapy.
All of the study patients had their total testosterone levels checked at least on two separate occasions. Use of testosterone therapy (injection, gel or patch) was ascertained from the medication prescription of patient medical records.
Definition of low and normal testosterone levels
Testosterone levels can vary significantly between different laboratories, even when they use the same commercial kits.8-10 Therefore, low total testosterone was considered to be present when the total testosterone level was less than the lower limit of the normal laboratory reference range reported for that particular test result. This method of classifying each test result as low or normal based on its respective laboratory reference range permitted inclusion of results from a large number of laboratories in the entire Veterans Affairs Health System during a period of over 14 years, with different test assays and different reference ranges. As there are no universal testosterone thresholds, this also eliminated any bias that may have been introduced by the use of an arbitrary cut-off value.
Testosterone replacement therapy achieved normalization of total testosterone levels in 63% patients while the rest of this group continued to have low testosterone (indicating a sub-optimal testosterone treatment). Mean duration of treatment was 3 years, with a total follow-up of up to 14 years.
What this study adds
The association of testosterone therapy with all-cause mortality, MI and stroke was compared between the three groups.
The all-cause mortality (hazard ratio (HR): 0.44), risk of MI (HR: 0.76), and stroke (HR: 0.64) were significantly lower in Group 1 (n = 43 931, median age = 66 years, mean follow-up = 6.2 years) vs. Group 3 (n = 13 378, median age = 66 years, mean follow-up = 4.7 years).
Similarly, the all-cause mortality (HR: 0.53), risk of MI (HR: 0.82), and stroke (HR: 0.70) were significantly lower in Group 1 vs. Group 2 (n = 25 701, median age = 66 years, mean follow-up = 4.6 years).
There was no difference in MI or stroke risk between Group 2 and Group 3.
It was concluded that normalization of total testosterone levels with testosterone therapy is associated with a significant reduction in all-cause mortality, MI, and stroke during an extended follow-up period.
This study fills an important knowledge gap in the current medical literature on testosterone treatment and cardiovascular disease / mortality outcomes. Notably, this is the first study to demonstrate that significant benefit is observed only if the testosterone treatment dose is adequate to normalize total testosterone levels. Patients who failed to achieve an adequate elevation in total testosterone levels did not experience a reduction in MI or stroke, and had significantly less benefit on mortality. The importance of achievement of therapeutic testosterone levels was previous demonstrated in the BLAST study.11 The study reported here is also the first study in which all included subjects had at least two testosterone measurements to document normalization of testosterone levels after testosterone therapy and adherence to testosterone treatment.
A significant limitation of all retrospective studies on testosterone therapy has been the inability to fully ascertain whether patients in the treatment group actually took the testosterone medications in an adequate dose and adhered to testosterone treatment. One previous retrospective study – which associated testosterone prescriptions with an increased MI risk - did not even report testosterone levels at all!3 The current study overcomes these limitations by assessing follow-up testosterone levels - which is a reliable indicator for adequacy of dosing and treatment compliance – in all included study groups.
This study also differs from previous retrospective studies on testosterone therapy by having a large subject population with extensive follow-up. It clearly shows that effective testosterone therapy is associated with lower rates of MI and stroke in men without pre-existing heart disease, in whom low testosterone levels are documented and effective testosterone therapy (defined as normalization of testosterone levels) is provided. This study also highlights that testosterone therapy should aim for testosterone treatment doses that result in normalization of total testosterone level, as this is a prerequisite to achieve a reduction in MI and stroke.