Reviewed research

Authors Marberger M, Wilson TH, Rittmaster RS.

Review Date Nov 2010

Citation BJU International 2010; Oct 18. doi: 10.1111/j.1464-410X.2010.09766.x.



Testosterone supplementation in older men with age-related reduced testosterone levels (as opposed to established clinical androgen deficiency) to help with various ailments, including sexual dysfunction, is controversial and not backed by good evidence. The association between testosterone levels and sexual dysfunction in older men remains poorly defined and other factors such as chronic diseases including diabetes and obesity may be more important predictors of sexual problems.



To identify predictors of sexual dysfunction using baseline data from the reduction by dutasteride of prostate cancer events (REDUCE) study.



REDUCE was a 4-year randomized, double-blind, placebo-controlled study evaluating the efficacy and safety of once-daily dutasteride 0.5 in over 8000 men aged 50-75 years with a prostate-specific antigen (PSA) level of 2.5-10 ng/mL (50-60 years) or 3.0-10 ng/mL (>60 years) and a negative prostate biopsy within 6 months of enrolment.

Baseline values (mean serum testosterone, age, International Prostate Symptom Score [IPSS, a measure of lower urinary tract symptoms], total prostate volume [TPV], body mass index [BMI], and presence of diabetes/glucose intolerance) were compared in subjects with and without sexual dysfunction (sexual inactivity, impotence, decreased libido or a Problem Assessment Scale of the Sexual Function Index [PAS-SFI] score <9 indicating a ‘small problem’ in at least one of sexual inactivity, impotence or decreased libido).



Multivariate logistic regression showed that baseline age and IPSS were significant predictors of all four sexual function criteria examined (P<0.0001). BMI was a significant predictor of decreased libido, impotence and a PAS-SFI score <9, while diabetes/glucose intolerance was a significant predictor of sexual inactivity, impotence and a PAS-SFI score < 9. Testosterone and TPV were not significant predictors of any sexual function criterion examined.



Age, IPSS, BMI and diabetes/glucose intolerance, but not serum testosterone or TPV, were significant independent predictors of sexual dysfunction in the REDUCE study population. The lack of association between sexual dysfunction and serum testosterone questions the value of modestly reduced or low normal testosterone levels as criteria for choosing testosterone replacement in older men with sexual dysfunction.


Points to Note
  1. This was a cross-sectional analysis and thus the authors’ use of the word ‘predictor’ is questionable. This study design can only identify ‘associations’ not causes and effects.
  2. The use of a single blood sample to measure testosterone (as was used in this study) is not very reliable at an individual patient level but may indicate trends at a population level.
  3. Much remains to be elucidated with respect to the mechanisms underlying male sexual dysfunction, but it is clear that BMI and other aspects of the ‘metabolic syndrome’ are important factors in addition to age and lower urinary tract symptoms.
  4. The results of this study do not support the use of testosterone therapy in older men with sexual dysfunction.
  5. Large-scale RCTs of testosterone administration in older men are warranted to define risk-benefit profiles and to determine the testosterone threshold at which intervention may be warranted for a variety of health outcomes.


Website: http://www.ncbi.nlm.nih.gov/pubmed/20955266

Reviewed by
Updated on