Are there symptom‐specific testosterone thresholds in aging men?

Study Type – Prognosis (retrospective cohort) Level of Evidence 2b What’s known on the subject? and What does the study add? Late onset hypogonadism is a syndrome combining low testosterone levels and clinical symptoms. But specific testosterone levels as well as clinical symptoms (e.g. sexual dysfunction) are poorly defined and still under debate. This study explores the association between testosterone levels and clinical symptoms with regard to co‐morbidities such as obesity, smoking and age. A main focus in this study is to investigate the influence of age on late onset hypogonadism, as many symptoms, which are correlated to low testosterone, might reflect just a normal process of aging. Finally the results show that only few symptoms (mainly psychosomatic symptoms) are associated with low testosterone and all other symptoms are significantly associated with age. OBJECTIVE • To study the association between specific clinical symptoms (e.g. low libido and erectile dysfunction) and testosterone levels and age in order to define symptom‐specific testosterone thresholds. MATERIALS AND METHODS • Serum samples for testosterone determination were obtained from 675 healthy men. • Participants underwent urological examination and completed the Aging Males Symptoms scale, the Beck Depression Index and the International Index of Erectile Function. Overall scores and those from individual questions from the questionnaires were evaluated. • Testosterone levels in men with symptoms were compared with those in men without symptoms. • The risks of clinical symptoms were evaluated using univariate, multiple multinomial regression analyses and Bonferroni correction. RESULTS • Significant associations between testosterone levels and a number of androgen deficiency symptoms were seen at testosterone levels of 13.5–14.4 nmol/L, but multiple logistic regression analysis revealed confounding effects with age. • Symptoms such as loss of libido, lack of vigour and sexual dysfunction were associated with age rather than with testosterone. • Erectile dysfunction was reported at testosterone levels between 14.65 nmol/L and 14.8 nmol/L, but was again significantly associated with age rather than testosterone levels. • The severity of symptoms significantly increased with decreasing testosterone levels using univariate analysis, but only the relationship with psychological symptoms remained significant after Bonferroni correction. CONCLUSION • In aging males, androgen deficiency sy