Tenofovir disoproxil fumarate (TDF) is a very effective antiviral drug that has been associated with tubular dysfunction. The aim of this study was to analyze the demographic, pharmacokinetic, and pharmacogenetic variables associated with TDF discontinuation for renal outcomes in stable HIV-positive patients using multivariable analyses. Three hundred and four patients were included (73% male, with median age and eCrCl of 45.3 years and 90.9 mL/min, respectively). After a median follow-up of 28.3 months, 27 patients discontinued TDF for renal adverse events [persistent urinary abnormalities (n = 21) or eCrCl < 60 mL/min (n = 6)] providing an incidence of 3.77 events per 100 patient-year. The probability of TDF discontinuation was higher with several features (male gender, older age, not Caucasians ancestry, absence of intravenous drug abuse, protease inhibitors, previous indinavir, HCV-positivity, lower CD4 cell count, detectable HIV-RNA, lower eCrCl, spot-urine proteinuria) and higher tenofovir concentrations but not genetic variants. Tenofovir plasma concentrations were prognostic of TDF discontinuation for renal adverse events suggesting that dose-adjustment may be warranted for long-term safety.