Effects of Intermission and Resumption of Long-term Testosterone Replacement Therapy on Body Weight and Metabolic Parameters in Hypogonadal in Middle-aged and Elderly Men

Aksam Yassin; Yousef Almehmadi; Farid Saad; Gheorghe Doros; Louis Gooren


Clin Endocrinol. 2016;84(1):107-114. 

In This Article


In this observational study of 262 elderly men with hypogonadism (defined as a combination of symptoms of T deficiency with a serum T level ≤12·0 nmol/l), receiving treatment with T undecanoate injections, 147 men had an intermission of their T administration for reasons of reimbursement of costs or prostate cancer for a mean of 16·9 months. This intermission was not designed, but its occurrence provided an opportunity to monitor the effects of withdrawing T treatment on measures of obesity, metabolic factors and well-being. These effects could be set against the course in men who continued their T administration uninterruptedly.

As expected, in Group I serum total and free T levels returned to the levels before T administration, while levels of SHBG increased to pretreatment values. Body weight, BMI and waist circumference increased significantly but did not quite return to pretreatment values. Fasting glucose and serum HbA1c returned to baseline levels or were even higher. Similar patterns were observed for serum cholesterol, LDL cholesterol, triglycerides, HDL cholesterol, total cholesterol/HDL ratio and non-HDL cholesterol, as well as systolic blood pressure, pulse pressure and liver transaminases. The effects on diastolic blood pressure and on the inflammation marker CRP were more benign. Upon the resumption of T administration, serum levels of total and free T reached similar levels as before the intermission but serum SHBG declined further. Over the post-intermission observation period of mean 14·5 months, body weight, BMI and waist circumference declined but did not reach the lower levels achieved over 65·5 months before the intermission, but serum glucose and HbA1c did. This was also the case with serum lipids and systolic blood pressure and pulse pressure, and liver transaminases.

Our results strongly indicate that the beneficial changes induced by normalization of serum T are not sustained when T administration is halted and serum levels return to hypogonadal values. But these beneficial effects do manifest themselves again upon the resumption of T administration restoring serum T to normal.

There are almost no studies monitoring the effects of halting T administration to men with hypogonadism, particularly in the middle-aged and elderly. In a recent study, we compared the effects of T administration to two groups of men: a group of 450 men between the ages of 32 and 65 years (mean 56·10 ± 6·29) and a group of 111 men >65 years (mean 68·45 ± 2·91).[19] The symptoms of T deficiency and the benefits of restoring serum T in men were not different between younger men and men >65 years of age. We interpreted this to indicate that age is not a significant factor in the symptomatology and benefits that men receive from restoring their T levels to normal.

In a Japanese study, beneficial psychological effects were maintained over a period of 55 months following cessation of T administration.[20] But in a study of middle-aged hypogonadal severely obese men, 1-year T treatment improved cardiometabolic and hormonal parameters. Upon withdrawal of T, a return back to hypogonadism occurred within 6 months, with loss of beneficial effects on cardiovascular and body composition improvements that had been attained.[21] The results of this study[21] and ours would seem to argue that the hypogonadism encountered in elderly men is truly a state of hypogonadism where the effects of restoring T levels to normal are lost when serum T falls back to hypogonadal values. The effects of T withdrawal manifested themselves indeed within a short interval of approximately 6 months which is in agreement with the study by Francomano et al.[21]

The results of our study would seem to imply that T replacement in hypogonadal men is long-term. This has been accepted for young men but has raised concerns for men over 50–60 years of age. Of late, particularly, the implications of T treatment for cardiovascular disease have been hotly debated. The data are not sufficient to arrive at a final verdict,[6,22] but critical analysis of literature does not give rise to immediate concerns.[23] The same applies to the risks of prostate cancer. Also with regard to prostate disease, analysis of the data is reassuring.[24]

There are now guidelines[5,25] that help physicians to let elderly men benefit from T treatment while prudently avoiding pitfalls that may occur in the course of T administration, be they related or not to T treatment.

Our study has methodological shortcomings. First of all, the study is observational and not placebo-controlled. The intermission in T administration was not randomized but dictated by insurance policy (no reimbursements of costs) or incidental prostate disease. During the intermission, anthropometric values, blood pressure, glycaemic control and lipids returned to their less favourable pretreatment levels. Whether (drug) interventions or changes in medication dosage took place during this interval was not assessed in this study.

Nevertheless, our observations are unambiguous and point in the expected direction and should be followed up by studies with a better design.