![[ Visit Client Website ]](images/banner.gif)
Purpose: Initial vancomycin dose is often calculated from population-based pharmacokinetic parameters. The serum creatinine (S.Cr) is often utilized to estimate creatinine clearance (eCrCl) to calculate elimination. It is recognized that creatinine may be inaccurate in patients with reduced muscle mass. Cystatin C has recently been identified as a correlate to renal function with less reliance on muscle mass. Equations using cystatin C are available to estimate eGFR in lieu of S.Cr.
Methods: Since November, 2008 the Minneapolis VA Medical Center Pharmacy pharmacokinetic service has obtained cystatin C levels to calculate dosing regimens for patients with significantly reduced muscle mass. S.Cr. is also collected. Vancomycin trough levels are obtained at steady-state conditions to confirm achievement of target concentrations. We retrospectively compared the accuracy of S.Cr. versus cystatin C in population-based pharmacokinetic equations to predict vancomycin trough concentrations. Each prediction was then compared versus the true vancomycin trough concentration for bias and precision. The study population consisted of patients with spinal cord injuries and/or a BMI less than 18.5.
Results: 32 patients met inclusion criteria. The mean error of predicting the true trough using cystatin C was -3.4mcg/ml (95%CI=-5.1 to -1.7) versus -8.0mcg/ml (95%CI =-10.6 to -5.3) using S.Cr. (p=0.001). Both methods under predicted the true trough. Precision, measured by mean absolute error was 4.6mcg/ml (95%CI=3.4 - 5.9) using cystatin C compared to 8.9mcg/ml (95%CI=6.5–11.3) using S.Cr. (p=0.002). The ability to predict within 50% of the true trough (P50) was 91% using Cystatin C versus 56% using S.Cr. (p=0.007).
Conclusion: In this patient population, use of cystatin C to predict vancomycin elimination was superior to use of serum creatinine.