OBJECTIVES The primary objective of this retrospective study was to determine if implementation of vancomycin monitoring criteria could reduce the number of serum vancomycin concentrations obtained without adversely affecting patient outcomes.
BACKGROUND Controversy regarding the correlation between serum vancomycin concentrations and its efficacy and/or toxicity persists. Little evidence has shown a correlation between vancomycin peak concentration (20–40 mg/L) and toxicity. Likewise, there is little information that supports an association between trough (5–15 mg/L) serum concentrations and clinical cure or in vitro killing rates. For these reasons, many question the clinical utility and cost-effectiveness of monitoring serum vancomycin concentrations.
METHODS We reviewed medical records of 193 patients (1 d-19 yrs) who received vancomycin during a 2-month period before (Group 1; n = 100) and after (Group 2; n = 93) implementation of vancomycin monitoring criteria.
RESULTS There was no difference (P > 0.05) in baseline age, weight, white blood cell count, temperature, serum creatinine, and blood urea nitrogen between Groups 1 and 2. Although 49.5% of all patients had vancomycin serum concentrations performed, significantly (P < 0.005) fewer patients in Group 2 (32%) were monitored when compared to Group 1 (65%). Peak serum vancomycin concentrations were within the reference range (20–40 mg/L) in 48% of patients in Group 1 compared to 80% in Group 2 (P = 0.03). The mean duration of vancomycin therapy was greater (P = 0.004) for patients in Group 1 (7 ± 7.5 days) compared to Group 2 (4 ± 4.4 days) and the medians for the two groups were also different. Mean ending temperatures (P = 0.23), white blood cell counts (P = 0.71), serum creatinine (P = 0.3) and BUN (P = 0.24) were not different for the two groups.
CONCLUSIONS Implementation of criteria to decrease unnecessary serum vancomycin concentration monitoring does not adversely affect patient outcome and may decrease cost to the institution, healthcare system, and patient.