Adequacy of androgen replacement influences bone density response to testosterone in androgen-deficient men

Androgen deficiency (AD) leads to bone loss and contributes to osteoporotic fractures in men. Although low bone mineral density (BMD) in AD men is improved by testosterone replacement, the responses vary between individuals but the determinants of this variability are not well defined.

Retrospective review of dual energy X-ray absorptiometry (DEXA) of the lumbar spine and proximal femur in men with established AD requiring regular androgen replacement therapy (ART). After a DEXA scan all men were treated with testosterone implants (800 mg, approximately 6 month intervals). Patients were classified as having a congenital, childhood, or post-pubertal onset, as well as according to the adequacy of treatment prior to their first DEXA scan as untreated, partially treated or well treated.

Men with AD requiring regular ART (n = 169, aged 46.3+/-1.1 years, range 22-84 years) underwent a DEXA scan prior to being treated with testosterone implants (800 mg, approximately 6 month intervals). In cross-sectional analysis at the time of the first DEXA scan untreated men (n = 24) had significantly reduced age-adjusted BMD at all four sites (L1-L4, femoral neck, Ward's triangle and trochanter). Well-treated men (n = 77) had significantly better age-adjusted BMD at all four sites compared with those who were partially treated (n = 66) or untreated (n = 24) with their age-adjusted BMD being normalized. In a longitudinal assessment of men (n = 60) who had two or more serial DEXA scans, at the second DEXA scan after a median of 3 years, men who were previously partially treated (n = 19) or untreated (n = 11) had proportionately greater improvements in BMD, significantly for Ward's triangle (P = 0.025) and the trochanter (P = 0.044) compared with men (n = 30) previously well treated.

The present study demonstrates a positive relationship between adequacy of testosterone replacement and BMD in men with overt organic AD. Additionally, the BMD of well-treated AD men approximates that of age-matched non-AD controls. The greatest BMD gains are made by those who have been either untreated or partially treated, and optimal treatment over time (median 3 years) normalizes BMD to the level expected for healthy men of the same age.