Abstract
International guidelines suggest to use testosterone therapy (TTh) in hypogonadal men presenting symptoms of testosterone deficiency (TD), even if there is no fixed threshold level of T at which TTh should be started. We aimed to develop and validate a nomogram named TRACE (Testosterone ReplACEment) for predicting the need of TTh in patients with "low-normal" total testosterone levels. The following nomogram variables were used: serum T level; serum LH level; BMI; state of nocturnal erections; metabolic comorbidities; and IPSS total score. The nomogram has been tested by calculating concordance probabilities, as well as assaying the calibration of predicted probability of clinical testosterone deficiency and need for TTh, together with the clinical outcome of the TTh. A cohort of 141 patients was used for the development of the nomogram, while a cohort of 123 patients attending another institution was used to externally validate and calibrate it. Sixty-four patients (45.3%) received TTh. Among them, sixty patients (93.7%) reported a significant clinical improvement after TTh. The nomogram had a concordance index of 0.83 [area under the ROC curve 0.81 (95% CI 0.71-0.83)]. In conclusion, the TRACE nomogram accurately predicted the probability of clinical impairment related to TD, and resulted in a simple and reliable method to use to select hypogonadal patients with not clearly pathological testosterone values who will benefit from TTh.