Decreasing Vancomycin Utilization in the NICU by Optimizing Treatment Decisions in Suspected Late Onset Sepsis

Abstract

BackgroundLate Onset Sepsis (LOS) is frequently suspected in NICU patients in the setting of nonspecific clinical symptoms. Based on institutional antibiogram data, empiric treatment of LOS in our NICU is vancomycin and amikacin with a plan to deescalate or discontinue based on culture results and symptomatology. Baseline data in our NICU revealed vancomycin overuse where vancomycin was continued past 48 hours of culture negativity, after Gram-negative urinary tract infection (UTI) was diagnosed, or for urine cultures reported with multiple organisms or < 10,000 CFU/mL. Our objective was to eliminate inappropriately prolonged empiric use of vancomycin for suspected LOS or UTI.MethodsTo institute timely discontinuation of vancomycin when cultures are negative at 48 hours, group education sessions were conducted for physicians, nurse practitioners, and nurses that included evidence-based criteria for diagnosing “true UTI”. Vancomycin indication for use and duration were added to the rounding script for the night shift.ResultsAt baseline over a 6 month period, extra vancomycin doses were administered in 39% of LOS courses, typically because late-night doses (past 48 hours culture negativity) preceded the decision to discontinue empiric therapy on morning rounds. After intervention, during a 6 month period, extra vancomycin doses were reduced to 3%. A baseline anonymous survey revealed that some prescribers advocated continuing vancomycin in the setting of urine cultures with < 10,000 CFU/mL (18%) or for Gram-negative UTI until sensitivities are reported (24%). After intervention, these were both reduced to 7%. At baseline over a 9 month review of UTI data, the use of vancomycin past 48 hours occurred in 86% of patients with negative or contaminated urine cultures (< 10,000 CFU/mL or > 1 organism) and in 48% of patients with Gram-negative UTI. In the post intervention phase, this occurred in 0% and 50% (N = 2) of cases respectively.ConclusionA high incidence of overtreatment with vancomycin was found to be related to inconsistent system processes and knowledge deficiencies. Through improved documentation, staff education and creation of evidence based guidelines for the diagnosis and management of UTI, we successfully minimized vancomycin overutilization for suspected LOS and UTI in the NICU.Disclosures All authors: No reported disclosures.