Abstract
Background: A tertiary healthcare institution, located in New Zealand, switched from targeting trough vancomycin concentrations to 24-h area under the concentration-time curve (AUC 24), facilitated by pharmacists using a Bayesian software-guided approach.
Aim: To evaluate the AUC 24 obtained for initial and ongoing prescribed vancomycin doses, time to attain target AUC 24 (400–600 mg/L per hr [mg/L.h]), the relationship between trough concentrations and AUC 24 , and the adherence to local vancomycin prescribing guidelines.
Method: Electronic patient records were used and retrospective data were compiled for adult patients given vancomycin intermittent infusions for ≥48 h during an 8-month period from 1 March 2021–31 October 2021. Descriptive summary statistics were calculated with values reported as median and range. Categorical variables were reported as counts and percentages. Linear regression analysis was performed to measure correlation between steady-state AUC 24 and serum trough concentrations. Ethical approval was granted by the Health New Zealand Canterbury Research Office (Reference no: RO#20311) and the study conforms with the National Ethics Advisory Committee National ethical standards: health and disability research and quality improvement.
Results: A total of 119 patients were included. The median (range) age was 63 (18–95) years, and creatinine clearance was 72 (13–205) mL/min. Following the first maintenance dose, the median (range) predicted steady-state AUC 24 was 648 (253–1730) mg/L.h; with 31% (n = 37) of patients attaining the target range. Of the 119 patients, after iterative dosing informed by Bayesian dose prediction, 100 (84%) patients attained the target AUC 24 in a median (range) of 2 (1–12) days. In these 100 patients, the median (range) AUC 24 and trough concentration was 519 (409–600) mg/L.h and 13 (4–23) mg/L respectively. The linear correlation between AUC 24 and trough concentrations was R 2 = 0.48. Most (n = 71) vancomycin courses adhered to the loading and first maintenance dosing guidelines.
Conclusion: The proportion of patients achieving the target AUC 24 range with the first maintenance dose was low but substantially improved with vancomycin concentration measurement and Bayesian dose prediction. The modest correlation between trough concentrations and AUC 24 supports the recommendation to target AUC 24 directly. There was a high degree of prescribing guideline adherence in our study.