Vancomycin and acute kidney injury following hip and knee surgery: A retrospective cohort study at five hospitals

Background: A previous study reported an association between vancomycin prophylaxis and postoperative acute kidney injury (AKI) following hip and knee arthroplasty. Questions/purposes: It was hypothesized that adults who received surgical infection prophylaxis containing vancomycin would experience a higher incidence of post-operative AKI within 7 days of knee and hip arthroplasty compared to patients who received surgical infection prophylaxis that did not contain vancomycin. Methods: This retrospective cohort included adults admitted to 5 hospitals from January 1, 2012 through December 31, 2014 with a principal procedure for knee or hip arthroplasty. Patients with baseline stage 5 chronic kidney disease, AKI prior to surgery, or lacking ≥2 or more serum creatinine values within 72 hours of each other were excluded. Antibiotics administered on the day of surgery were classified as surgical infection prophylaxis. The primary endpoint was incident postoperative AKI within 7 days of surgery defined as a 0.3 mg/dL or 50% increase in serum creatinine within a 72-hour rolling window. The primary analysis was multiple logistic regression. This study had 95% power to detect an OR≥=1.32 (≥=4.8% risk difference) for incident AKI using a two-sided α=0.05. Race was categorized as black vs. non-black. Results: This study included 4,632 patients who were predominately elderly (mean age of 68 ± 12), white (73%), female (60%), admitted to an academic medical center (70%), and undergoing a primary arthroplasty (81%). The incidence of postoperative AKI was 12.3% (n=568). The prevalence of vancomycin prophylaxis was 55.8% (n=2,584). The incidence of postoperative AKI was 12.3% (n=568) for all patients, 12.3% for those who received surgical infection prophylaxis containing vancomycin, and 12.3% for those who received surgical infection prophylaxis that did not contain vancomycin. Vancomycin prophylaxis was not associated with postoperative AKI (adjusted OR=1.07, 95%CI 0.88 to 1.30, P=0.473) compared to patients who did not receive vancomycin using logistic regression that adjusted for surgery type, age, race, chronic kidney disease, congestive heart failure, anemia, and angiotensin-converting enzyme inhibitors/angiotensin receptor II blockers. Conclusions: Among 4,632 elderly patients admitted to five hospitals over a 3-year period, vancomycin prophylaxis (at doses of 1g or 2g on day of surgery) was not associated with postoperative AKI within 7 days of knee or hip arthroplasty compared to patients who did not receive vancomycin prophylaxis