Introduction

Large studies examining the relationship between testosterone levels and depressive symptoms (DS) have shown conflicting results. We investigated the association between testosterone levels (tT) and mood impairment in a homogeneous cohort of Caucasian-European men seeking medical help for either solitary erectile dysfunction (ED) or premature ejaculation (PE) as primary reasons for office evaluation, without any known psychiatric disorders.

Methods

Complete data from the last 224 consecutive patients aged 18-70 years were analysed. Patients completed the 21-item Beck's inventory for Depression (BDI), the International Index of Erectile Function (IIEF) and the Premature Ejaculation Diagnostic Tool (PEDT). DS were considered for BDI≥11. Clinical depression (CD) was defined as a BDI ≥16. IIEF-EF was categorized according to Cappelleri's criteria. Testosterone deficiency (TD) was defined as tT <3.5 ng/mL. Patients were categorized according to their PEDT score (PEDT <11 vs. ≥11). Comorbidities were scored with the Charlson Comorbidity Index (CCI, categorized 0 vs ≥1). Descriptive statistics and logistic regression models tested the association between tT and DS and/or CD.

Results

Overall, mean (SD) age was 45.7 (14.1) yrs. Mean tT value and BDI were 4.7 (2.4) ng/mL and 8.59 (7.9), respectively. According to IIEF-EF domain scores, 37 (16.5%), 21 (9.4%), 33 (14.7%) and 71 (31.7%) men reported mild, mild-to-moderate, moderate and severe ED, respectively. Solitary PE was complained by 62 (27.6%) patients. Of all, obesity (NIH class≥1) and TD were observed 31 (14%) and in 86 (38.4%) patients, respectively. BDI scores suggestive for DS and CD were observed in 64 (28.6%) and 41 (18.3%) patients, respectively. Patients with TD showed higher rate of obesity (23.5% vs. 8.0%; p<0.001), BDI scores suggestive for either DS (41.9% vs. 20.3%; p=0.001) or CD (27.9% vs. 12.3%; p=0.003) compared to eugonadal patients. No differences in terms of age, CCI, IIEF-EF and PEDT were observed between the two groups. At univariate analysis tT was negatively correlated BDI scores (Rho -0.23; p<0.001). At multivariate analysis, both BMI (OR 1.1; p=0.013) and TD (OR 2.1; p=0.03) achieved independent predictor status for DS, after accounting for age, CCI, IIEF-EF and PEDT. Conversely, only TD (OR 2.34; p=0.03) achieved independent predictor status for CD.

Conclusions

These findings showed that both DS and unacknowledged CD are highly prevalent in men with concomitant TD and sexual dysfunction. Testosterone deficiency emerged as an independent predictor for CD in this cohort of patients without known psychiatric disorders.