Relationship Between Testosterone and Mortality in Men

Abstract & Commentary

By Harold L. Karpman, MD, FACC, FACP, Clinical Professor of Medicine, UCLA School of Medicine. Dr. Karpman reports no financial relationship to this field of study.

Synopsis: Absent or low testosterone levels in men appear to have a pathogenic role in the development of cardiovascular disease resulting in increased cardiovascular and all-cause mortality and are not simply "markers" for illness or wellness.

Source: Khaw K, et al. Circulation. 2007;116:2694-2701.

The relationship between endogenous testosterone concentrations and overall health in men is still not well established and, in fact, it is quite controversial in many respects. A recent trial in 87 elderly men confirmed the results of many previous trials15-18 that revealed no significant benefit from exogenous testosterone replacement on body composition, physical performance or quality-of-life.3,4 Despite the fact that, in some of these studies, testosterone administration was associated with extremely severe adverse health effects such as sudden cardiac death and liver disease,15-18 exogenous testosterone is still being widely used because of the growing belief among patients and even among many physicians that hypogonadism is frequently associated with poor health and that the use of relatively low doses of exogenous testosterone has benefits for health and well-being1,2 whether or not low endogenous testosterone levels are present.1,2

The prospective relationship between endogenous testosterone concentrations and mortality due to all causes, cardiovascular disease and cancer occurring in a nested case-control study of 11,606 men was examined and reported by Khaw and his colleagues from Cambridge School of Clinical Medicine and the Royal Marsden Hospital in London.5 The relationship between a single baseline endogenous testosterone concentration and death in 825 men who did not have cancer or cardiovascular disease was studied and compared to the control group of 1489 men who were still alive and who were matched for age and the date of the baseline visit. Endogenous testosterone concentrations at baseline were inversely related to mortality due to all causes (825 deaths), cardiovascular disease (369 deaths), and cancer (304 deaths). After excluding deaths which occurred in the first two years after the beginning of the study (and adjusting for age, date of visit, body mass index, systolic blood pressure, blood cholesterol, cigarette smoking, diabetes mellitus, alcohol intake, physical activity, social class, blood levels of dehydroepiandrosterone sulfate, androstanediol glucuronide, and sex hormone binding globulin), increasing testosterone concentrations appeared to be inversely related to mortality due to all causes, cardiovascular causes and cancer with an approximately 25 to 30% lower risk of total mortality in the highest compared with the lowest quartile of testosterone level.


Patients who exhibit reductions in serum testosterone levels resulting from aging or chronic disease have signs and symptoms similar to those seen in classical male hypogonadism and are associated with increased fat mass, decreased lean body mass, decreased muscle strength, and a diminished quality of life.6 Numerous studies have reported findings similar to Khaw's results.5 For example, it has been reported that men whose total testosterone levels were in the lowest quartile were 40% more likely to die than were men with higher androgen levels independent of age, adiposity, lipid levels, adipokines and lifestyle.7 Low testosterone levels have been predictive of increased risk of death due to cardiovascular and respiratory disease which is not surprising because these levels have been found to be independently associated with many of the individual risk factors of heart disease; for example, low testosterone levels are inversely related to fat mass in men which is an independent predictor of cardiovascular death.8 Also, systolic and diastolic blood pressure levels,9 arterial calcification,10 intima-media thickness,11 type 2 diabetes mellitus,12 abnormal lipid profiles,13 inflammatory cytokines,14 and the metabolic syndrome are all important cardiac risk factors and are inversely related to the endogenous testosterone level. Despite the fact that the Khaw study5 was extremely well conducted, it should be noted that the testosterone values were based on only a single measurement of total testosterone and that measurements were not made of the free or bioavailable testosterone levels which are said to be more accurate than total testosterone levels especially in obese and diabetic subjects.

Multiple studies have demonstrated that absent or low testosterone levels are not simply "markers" for illness or wellness since they appear to play a pathogenic role in the development of cardiovascular disease and are are associated with increased all-cause and especially cardiovascular mortality. But what should clinicians do at this time—administer androgen therapy to elderly patients? There is no credible evidence suggesting that androgen should be administered to men with normal endogenous androgen levels and the final answer as to whether or not androgen therapy should be administered to men with low testosterone levels is not available at this time. One cannot assume that testosterone replacement will reduce or eliminate the increased risks apparently present in the patient with low endogenous testosterone levels and therefore, long-term, double-blind, randomized, placebo-controlled trials of androgen replacement in men of all ages with low testosterone levels and in elderly men with low and with normal testosterone levels are needed to accurately evaluate the effects of such therapy on cardiovascular disease, cardiovascular death and all-cause mortality. Hopefully, these trials will answer numerous questions such as the critical blood level below which treatment should be started, the optimal dose, what is the target testosterone level to be reached and, most important, the long-term safety of such therapy. The need for these large outcome studies has now been recognized and therefore it is expected that the Institute of Medicine which in the past has not recommended funding of an adequate study in this important area will soon alter its previous stance and recommend funding of these studies.

For the time being, since endogenous testosterone concentrations in men appeared to be universally related to mortality due to cardiovascular disease and all causes, clinicians will have to decide without the benefit of adequate outcome studies or approved guidelines as to whether or not androgen therapy is appropriate or still too risky to give to men with low androgen levels in the absence of pituitary or testicular disease and whether the apparent benefits of such therapy are worth the recognized risks.


1. Liverman CT, et al. Testosterone and aging. Washington, DC: Institute of Medicine, National Academies Press; 2004.

2. Kaufman JM, Vermeulen A. Endocr Review. 2005;26:833-876. Epub 2005 May 18.

3. Nair KS et al. N Engl J Med. 2006; 355:1647-1659,

4. Page ST, et al. N Engl J Med. 2007;356:635-636.

5. Khaw K, et al. Circulation. 2007;116:2694-2701.

6. Basaria S, Dobs AS. Am J Med. 2001; 110:563-572.

7. Laughlin G, et al. J Clin Endocrin Metab. 2008;93:68-75. Epub 2007 Oct. 2.

8. Van den Beld AW, et al. J Clin Endocrinol Metab. 2000;85:3276-3282.

9. Khaw KT, et al. J Hypertens. 1988;6:329-332.

10. Hak AE, et al. J Clin Endocrinol Metab. 2002;87:3632-3639.

11. Muller M, et al. Circulation. 2004; 109: 2074-2079.

12. Stellato RK, et al. Diabetes Care. 2000;23:490-494.

13, Haffner SM, et al. Relationship of sex hormones to lipids and lipoproteins in nondiabetic men. J Clin Endocrinol Metab. 1993;77:1610-1615.

14. Maggio M, et al. J Clin Endocrinol Metab. 2006; 91:345-347.

15. Bagatell CJ, et al. N Engl J Med. 334:707-714.

16. Gruenewald DA, et al. J Am Geriatr Soc. 2003;51:101-115.

17. Rhoden EL, et al. N Engl J Med. 2004;350:482-492.

18. Westaby D, et al. Lancet. 1977;2:262-263.