The comparative risk of acute kidney injury of vancomycin relative to other common antibiotics
Martina Gaggl, Virginia Pate, Til Stürmer, Abhijit V. Kshirsagar, J. Bradley Layton
Abstract
The glycopeptide antibiotic vancomycin is a mainstay in the treatment of Gram-positive infection. While its association with acute kidney injury (AKI) has waxed and waned, recent data suggest nephrotoxicity, even as mono-therapy. Our study aimed to evaluate the 2-week risk of AKI after at least 3 days of intravenous vancomycin mono-therapy initiated within 5 days of hospitalization compared to other intravenous antibiotics used for similar indications. We used a new user-active comparator study design and identified patients with a first hospitalization during which they received vancomycin or comparator, from commercial claims based in the United States. We estimated incidence rates, hazard ratios using adjusted cox-regression models, and standardized mortality/morbidity ratio weighted cox-regression models. In the 32,997 patients vancomycin was used in 17% of patients and 129 cases of AKI were observed. Overall incidence of AKI was 9.3 (95% CI 0.78-1.22) per 100 person-years. The adjusted hazard ratio for vancomycin versus all other comparators was 0.74 (95% CI 0.45-1.21). Separate models for respective comparators resulted in hazard ratios below the null, except for vancomycin vs. cefazolin. Intravenous vancomycin mono-therapy does not increase the risk of AKI compared to other intravenous antibiotics used for similar indication in this cohort of hospitalized patients.