Comparative Incidence of Acute Kidney Injury in Critically Ill Patients Receiving Vancomycin with Concomitant Piperacillin‐Tazobactam or Cefepime: A Retrospective Cohort Study

STUDY OBJECTIVE: The combination of vancomycin and piperacillin-tazobactam has been associated with an increased risk of acute kidney injury (AKI) in non-critically ill patient populations, but it is still unknown if this association exists in critically ill patients. The objective of this study was to compare the incidence of AKI development during therapy or within 72 hours after completion of therapy in adult critically ill patients who received vancomycin with concomitant piperacillin-tazobactam or cefepime. DESIGN: Retrospective cohort study. SETTING: Medical, surgical, and neuroscience intensive care units (ICUs) within a single tertiary care hospital. PATIENTS: A total of 122 critically ill patients who received at least 48 hours of combination therapy with vancomycin and piperacillin-tazobactam (49 patients) or vancomycin and cefepime (73 patients) during an ICU admission between September 2012 and December 2014. MEASUREMENTS AND MAIN RESULTS: The primary outcome was AKI development, as determined by the Acute Kidney Injury Network criteria, during combination therapy or within 72 hours of completion of combination therapy. The inverse probability of the treatment-weighting (IPTW) approach was used to account for potential treatment selection bias. AKI incidence was assessed in the unadjusted and propensity score-weighted cohorts. Of the 122 patients, 37 patients (30.3%) developed AKI. In the unadjusted analysis, the incidence of AKI was similar in the piperacillin-tazobactam group compared with the cefepime group (32.7% vs 28.8%, p=0.647). The average treatment effect between the groups was not significant, showing no association between β-lactam choice and AKI (β = -0.004, p=0.958). Secondary outcomes were ICU length of stay, hospital length of stay, AKI duration, and need for renal replacement therapy. The choice of β-lactam was not a significant predictor of any of these outcomes: ICU length of stay (β = 0.436, p=0.780), hospital length of stay (β = 3.819, p=0.125), AKI duration (β = -4.027, p=0.283), and need for renal replacement therapy (β = 2.828, p=0.161). CONCLUSION: After adjusting for propensity to receive each of the treatment choices, no significant difference was found in the incidence of AKI development or other outcomes between the groups. The previously described finding that concomitant vancomycin and piperacillin-tazobactam increases AKI in non-critically ill patients may not be generalizable to the critically ill population. Prospective evaluation of this hypothesis is warranted.