Vancomycin vs. Vancomycin/Piperacillin-Tazobactam-Associated Acute Kidney Injury in Noncritically Ill Patients at a Tertiary Care Military Treatment Facility

Broad-spectrum antibiotics are often used as initial empiric therapy in patients at risk for infections by multidrug-resistant organisms. Emerging literature and anecdotal reports within Tripler Army Medical Center indicate an increased incidence of vancomycin-associated acute kidney injury when used in combination with piperacillin-tazobactam. This is a retrospective, single-center study comparing the incidence of acute kidney injury in noncritically ill patients receiving either vancomycin or vancomycin in combination with piperacillin-tazobactam in a 206-bed tertiary care military training facility. Data were collected from electronic medical records between May 2012 and October 2014 and evaluated via multivariable logistic regression models. Patients included for analysis were 17 years of age and older, were admitted to medical/surgical wards, and received vancomycin or vancomycin in combination with piperacillin-tazobactam for at least 48 hours. A vancomycin trough level, baseline serum creatinine level, and at least two follow-up serum creatinine levels were required for inclusion. Patients were excluded if they were pregnant, admitted to an intensive care unit while on antimicrobial therapy, or their baseline serum creatinine was equal to or greater than 1.5 mg/dL. Of 1,133 patients evaluated retrospectively, 455 were included for analysis. Of 202 patients, 49 (24%) taking vancomycin in combination with piperacillin-tazobactam developed acute kidney injury in contrast to 28 of the 253 patients (11%) given vancomycin without piperacillin-tazobactam (unadjusted odds ratio 2.57 [95% confidence interval (CI) 1.55-4.28], p < 0.001). Dual therapy remained significant after adjusting for age, sex, body mass index, concomitant nephrotoxic agents, and preexisting comorbid status as evaluated by Charlson comorbidity index (adjusted odds ratio 2.14 [95% CI 1.26-3.6], p = 0.005). Contrast administration (p < 0.001), fluoroquinolone administration (p < 0.001), and Charlson Comorbidity Index > 6 (p = 0.008) were also found to be independent risk factors for acute kidney injury. Significant increased incidence of nephrotoxicity was noted with vancomycin and piperacillin-tazobactam as compared to vancomycin within Tripler Army Medical Center. This finding influenced our institution's decision to add ceftaroline as a formulary agent in the treatment of skin and soft-tissue infections and further supported the need for rapid de-escalation of antibiotics within our mi