Background:
Testosterone therapy improves symptoms in men with pathological androgen deficiency (1). Its role in ageing, obese men with Type 2 diabetes (T2D) and lowered testosterone levels, where symptoms may be confounded by comorbidities, is not well established.

Methods:
We assessed changes in androgen deficiency symptoms (using the Aging Men Symptom (AMS) score), sexual desire (Q17 of AMS) and erectile function (International Index of Erectile Function abridged version 5 [IIEF5]) in men with T2D participating in a randomised, double blind, placebo-controlled, 40-week trial of intramuscular testosterone undecanoate and a serum total testosterone (TT) of ≤ 12nmol/L (ClinicalTrials.gov, number NCT00613782).

Results:
88 men were randomised and 75 completed the study. Complete sets of data regarding AMS were obtained in 75 men and IIEF5 in 47 men. Baseline characteristics are shown in Table 1.
In the testosterone (T) group at 40 weeks, TT increased (+4.6nmol/L p<0.001 IQR [0.7,8.8]) while there was no significant change in the placebo group. Compared to placebo, testosterone had no effect on AMS total (adjusted mean difference between groups at 40 weeks; AMS total score -0.92 p=0.67 CI [-4.1,2.2]). In the T group at 40 weeks, no change in sexual desire was observed (-0.34 p=0.12 CI [-0.82,0.15]). There was no change in IIEF5 severity category at 40 weeks in either group (chi squared test, p= 0.67, p=0.32, respectively). While there was no correlation (p > 0.05) between baseline TT or calculated free testosterone with AMS, there was a significant association with depression (p=0.002) and positive correlation with baseline HOMA-IR (r=0.30, p=0.02).

Conclusions:
In this this RCT of moderately symptomatic men with T2D and lowered serum testosterone, testosterone treatment had no effect on symptoms of androgen deficiency or sexual function. This may be because these symptoms are a consequence of comorbidities, rather than of the gonadal status.