Abstract
Aim
Current guidelines recommend the combination of vancomycin with either piperacillin-tazobactam (PT) or a third generation cephalosporin (3GC) as empirical antimicrobial therapy of PJI, immediately after surgery. However, clinical and biological safeties of such high dose-combinations are poorly known.
Method
All patients managed in a reference center in France between 2011 and 2016 receiving an empirical antimicrobial therapy for PJI were included in a prospective cohort study. Antimicrobial-related AE upcoming during the empirical treatment phase were describe according to the Common Terminology Criteria for Adverse Events (CTCEA), and severe ones (grade ≥ 3) were reported to pharmacovigilance. AE determinants were assessed using univariate logistic regression.
Results
Three hundred and thirty-one patients (166 males, 50.2%; median age, 70.1 (IQR, 59.4–79.1) years) with empirically-treated PJI were included. Vancomycin (n=228; 68.9%), teicoplanin (n=33; 10.0%), antistaphylococcal penicillin (n=29; 8.8%) and daptomycin (n=4; 1.2%) were the most commonly used anti-Gram positive antimicrobials. Most common combinations were vancomycin-PT (n=122;36.9%) and vancomycin-3GC (n=33; 10.0%). Forty-two (12.7%) patients experienced 49 AE in a median delay of 8 (IQR, 5–13) days. They included 25 acute kidney injuries (AKI; 7.6% of patients) including 16 (4.8%) without vancomycin overdose, 4 drug reactions with eosinophilia and systemic symptoms, isolated fevers, rashes or pruritus (1.2% each), 3 eosinophilia (0.9%), 2 hepatitis (0.6%), and one febrile neutropenia, injection site reaction or vomiting (0.3% each). Ten AE were considered as severe (3.0% of patients). Treatment has to be stopped in most cases (n=38; 95.0%). All AE had a favorable outcome. In univariate analysis, the use of vancomycin (OR 6.878; p=0.026) and/or PT (OR 3.667; p<10–3), and consequently the vancomycin-PT combination (OR 4.149; p<10–3) were found to be determinants of empirical antimicrobial therapy-related AE. Moreover, vancomycin-PT combination was found as an AKI risk-factor (OR 8.000; p<10–3).
Conclusions
Empirical antimicrobial therapy of PJI is associated with a high rate of AE. These results reinforce recent data suggesting an increased risk of AKI when using vancomycin in combination with PT and encouraging the preferential use of 3GC or cefepim in this indication.
