A pilot study evaluating mid-point vancomycin concentrations to estimate pharmacokinetics and area under the curve: a prospective study.

Abstract

Background: Guidelines recommend monitoring vancomycin by estimating the area under the curve (AUC₂₄) with two levels, peak and trough, preferably using Bayesian software. However, such an approach is not feasible or cost-effective in all settings. In this study, we evaluated the accuracy and precision of estimating AUC₂₄ using a single mid-point level in critically ill cancer patients. We hypothesized that concentrations at the midpoint of the dosing interval may provide insight into the volume of distribution (Vd) and clearance (Cl), capturing pharmacokinetic parameters assessed by the peak and trough concentrations, respectively.

Methods: A prospective study that included critically ill cancer patients with stable kidney function who received vancomycin during their ICU stay. Trough, peak, and midpoint concentrations were measured at steady-state. Pharmacokinetic equations from the short infusion model was used to calculate the AUC₂₄, Cl, and Vd using a single midpoint. We also determined AUC₂₄, Cl, and Vd using Bayesian software with peak and trough concentrations. Accuracy and precision of the pharmacokinetic estimations utilizing the mid-point level were determined, with a comparison to the Bayesian approach.

Results: We included 91 patients, with a mean age of 53 years ± 17 (SD), 54% were males, and 77% had solid tumors. The mean prescribed vancomycin dose was 15 mg/kg/dose ± 3 (SD). Median AUC₂₄ was 426 mg.dl/hr (IQR 326-559) and 464 mg.dl/hr (IQR 345-664), using the midpoint and Bayesian approach, respectively. The mean Vd was 48 L (± 11) and 45 L (± 13) and Cl was 5 L/hr (± 3) and 5 L/hr (± 2) for both approaches, respectively. The accuracy and precision for the midpoint, against the Bayesian, were: AUC₂₄ (11%, 22%, respectively), Cl (17%, 24%, respectively), and Vd (6%, 9%, respectively).

Conclusions: Utilization of mid-point vancomycin levels to calculate pharmacokinetic parameters demonstrated promising findings. Further research with larger and more diverse patient populations is needed to evaluate this approach further.