OBJECTIVE: To determine the prevalence and factors associated with high-grade ED in Lima, Peru. MATERIAL AND METHODS: Analytical cross-sectional study in tuberculosis patients treated in an outpatient clinic of a public hospital in Lima, Peru in 2018. High-grade SD was the dependent variable, using the IIEF-5 questionnaire and the independent variables were sexual orientation, history of previous pathology, location of tuberculosis, type of treatment scheme and presence of hemoptysis. Prevalence ratios (PR) were estimated using simple and multiple regression models. RESULTS: Of 189 patients, the majority presented pulmonary tuberculosis (98.9%), overweight (25.9%) and just over half had high-grade ED (52.9%). In the simple regression it was found that the factors associated with presenting high-grade SD were age in years (RP=1.11), reporting diabetes (RP=2.09), having multi-drug resistant TB (RP=1.51) and presenting a treatment time from 1 year to more (RP=1.87). In the multiple regression, the variables that were associated with a higher frequency of high-grade TB were age in years (RP=1.11), a history of diabetes (RP=1.66) and having MDR TB (RP=2.04). CONCLUSIONS: This study suggests a high prevalence of ED in patients with TB, where four out of ten presented high-grade ED. Being older, having a history of diabetes and using an MDR treatment were positively associated with the development of high-grade ED. Studies with designs close to causality are required to know the real magnitude of the influence of clinical and therapeutic characteristics on the development of high-grade ED.