Vancomycin-associated acute kidney injury in pediatric cardiac intensive care patients.

OBJECTIVE

Acute kidney injury (AKI) is a significant source of morbidity among critically ill pediatric patients, including those that have undergone cardiac surgery. Vancomycin may contribute to AKI in pediatric patients admitted to a cardiac intensive care unit.

DESIGN AND SETTING

Patients admitted to the cardiac intensive care unit at Texas Children's Hospital and received vancomycin over a 4-year period were included in a case-control study. Patients were excluded if they underwent renal replacement therapy during vancomycin therapy. Patient demographic and disease state variables, vancomycin therapy variables, and use of other nephrotoxic medications were collected. The overall incidence of AKI was calculated based on doubling of serum creatinine during or within 72 hours of vancomycin therapy (vancomycin-associated AKI [vAKI]). Patients who developed vAKI were matched with three patients who did not develop vAKI, and conditional logistic regression was used to determine independent risk factors for vAKI.

RESULTS

A total of 418 patients met study criteria (males 57.8%) and infants (31 days to 2 years) were the most populous age group (48.6%). Vancomycin-associated AKI occurred in 30 patients (7.2%), which resulted in a total of 120 patients (30 cases; 90 controls). No significant differences were noted in vancomycin dosing between groups. Vancomycin-associated AKI patients were less likely to have undergone cardiac surgery (P < .05), more likely to have undergone extracorporeal membrane oxygenation (P < .05), and had greater exposure to nephrotoxic medications (P < .05). A conditional logistic regression model identified extracorporeal membrane oxygenation as associated with vAKI (odds ratio 14.4, 95% confidence interval 1.02-203, P = .048) and patients with prior cardiovascular surgery (odds ratio 0.10, 95% confidence interval 0.02-0.51, P < .01) or an elevated baseline serum creatinine (odds ratio 0.009, 95% confidence interval 0.0002-0.29, P < .01) as less likely to develop vAKI.

CONCLUSIONS

Vancomycin-associated AKI occurs infrequently in the pediatric cardiac intensive care population and is strongly associated with patient critical illness.