Comparison of Mw\Pharm 3.30 (DOS) and Mw\Pharm ++ (Windows) Versions of Pharmacokinetic Software for PK/PD Modelling of Vancomycin in Continuous Administration

Abstract

Objective: For a long time, the Mw\Pharm software suite (MEDI\WARE, Prague, Czech Republic/Groningen, Netherlands) has been used for PK/PD modelling in therapeutic drug monitoring (TDM). The aim of this study was to find the best model in the newer Windows version of Mw\Pharm++ 1.3.5.558 (WIN) for continuous administration of vancomycin. Patients: Twenty adult patients with a mean age of 66 ± 12 years, body weight 85 ± 16 kg, and median dose 1,625 g/24 h were repeatedly examined for vancomycin. Methods: Concentrations predicted by “#vancomycin_adult_k_C2”, “#vancomycin_adult_C2”, “vancomycin_adult_C2”, “van-comycin_C1” WIN models and “vancomycin (cont.inf.) %ahz” (DOS1) and “vancomycin adult” DOS models were compared with the measured values and with the DOS1 model. Statistics: Percentage prediction error (%PE) calculated as (predicted-measured)/measured or (predicted-DOS1)/DOS1, RMSE, Bland-Altman bias, Pearson’s coefficient of rank correlation (R), Student’s t-test. Statistical analysis was performed using the GraphPad Prism version 5.00 for Windows. Results: %PE values varied between −3.2 ± 33.0% and −7.4 ± 36.7%, with the exception of “vancomycin_C1” , the only one--compartment model, where it was −20.8 ± 39.4%. The best outcomes were achieved with “vancomycin adult” . The “#vancomy-cin_adult_k_C2” model produced the lowest %PE, RMSE, and Bland-Altman bias among the WIN models, but its correlation (Pearson’s R) was less tight. RMSE was the same in “vancomycin_adult_C2” while %PE and Bland-Altman bias were similar, with slightly better correlation when compared to “#vancomycin_adult_k_C2” . The %PE value between the two DOS models was 4.1 ± 13.9% (NS); “vancomycin adult” produced slightly better outcomes than DOS1. Conclusion: “vancomycin_adult_C2” and “#vancomycin_adult_k_C2” produced the best outcomes between WIN models. Both DOS models produced lower bias and their prediction was comparable.