Abstract
Background
Nephrotoxicity is a common adverse effect of polymyxin B (PMB) with reported acute kidney injury (AKI) rates of 20% to >60%. Data on PMB dosing to optimize efficacy while minimizing toxicity are limited. Previous studies suggest higher doses improve outcomes but are also associated with more AKI. Data are needed to evaluate optimal dosing and contributing factors to minimize AKI and to evaluate renal recovery.
Methods
Retrospective study evaluating PMB in adults at NewYork-Presbyterian Hospital from 2012 to 2016. Patients who received PMB dosed twice daily for ≥2 days were included. Patients on renal replacement therapy within 48 hours prior to PMB or with AKI at time of PMB initiation were excluded. A classification and regression tree (CART) analysis was performed to identify the PMB dose most predictive of AKI which defined the breakpoint for high- vs. low-dose PMB cohorts for all subsequent comparisons. The primary outcome was to determine whether high-dose PMB independently predicted AKI. Secondary outcomes included in-hospital mortality, time to AKI, and recovery of renal function.
Results
246 patients were included: majority were male (59%) with median age 41 years. Median PMB dose was 2.9 mg/kg/day or 180 mg/day for a median duration of 10 days. AKI occurred in 64% and 38% had recovery of renal function by hospital discharge. The breakpoint for high-dose PMB determined by CART was 160 mg/day, putting 104 in low-dose and 142 in high-dose groups. High-dose PMB was associated with AKI compared with low-dose PMB on univariable (75% vs.. 49%, P < 0.001) and multivariable (OR 3.43; 95% CI 1.68,6.99; P = 0.001) analyses. Concomitant vancomycin (OR 3.34; 95% CI 1.74,6.41;P < 0.001), history of transplant (OR 4.96; 95% CI 2.14,11.48;P < 0.001), and previous PMB exposure (OR 2.37; 95% CI 1.23,4.57; P = 0.01) were also identified as independent predictors of AKI. No significant differences were found for in-hospital mortality (28% vs. 21%, P = 0.326), renal recovery (37% vs. 41%, P = 0.723), time to AKI (median 5 vs. 6 days, P = 0.125) between groups.
Conclusion
High-dose PMB (>160 mg/day) was independently associated with AKI as well as concomitant vancomycin, history of transplant, and previous PMB exposure. High-dose PMB did not have a significant impact on in-hospital mortality, recovery of renal function, or time to development of AKI.
Disclosures
M. Yin, Gilead Sciences: Consultant, Consulting fee.