The empirical use of vancomycin in combination with a broad-spectrum beta-lactam is currently recommended after the initial surgery of prosthetic joint infection (PJI). However, the tolerability of such high-dose intravenous regimens is poorly known. Adult patients receiving an empirical antimicrobial therapy (EAT) for a PJI were enrolled in a prospective cohort study (2011 to 2016). EAT-related adverse events (AE) were described according to the common terminology criteria for AE (CTCAE), and their determinants were assessed by logistic regression and Kaplan-Meier curve analysis. The EAT of the 333 included patients (median age, 69.8 years; interquartile range [IQR], 59.3 to 79.1 years) mostly relies on vancomycin (n = 229, 68.8%), piperacillin-tazobactam (n = 131, 39.3%), and/or third-generation cephalosporins (n = 50, 15%). Forty-two patients (12.6%) experienced an EAT-related AE. Ten (20.4%) AE were severe (CTCAE grade ≥ 3). The use of vancomycin (odds ratio [OR], 6.9; 95% confidence interval [95%CI], 2.1 to 22.9), piperacillin-tazobactam (OR, 3.7; 95%CI, 1.8 to 7.2), or the combination of both (OR, 4.1; 95%CI, 2.1 to 8.2) were the only AE predictors. Acute kidney injury (AKI) was the most common AE (n = 25; 51.0% of AE) and was also associated with the use of the vancomycin and piperacillin-tazobactam combination (OR, 6.7; 95%CI, 2.6 to 17.3). A vancomycin plasma overexposure was noted in nine (37.5%) of the vancomycin-related AKIs only. Other vancomycin-based therapies were significantly less at risk for AE and AKI. The EAT of PJI is associated with an important rate of AE, linked with the use of the vancomycin and the piperacillin-tazobactam combination. These results corroborate recent findings suggesting a synergic toxicity of these drugs in comparison to vancomycin-cefepime, which remains to be evaluated in PJI. (This study has been registered at ClinicalTrials.gov under identifier NCT03010293.).