A novel method for predicting the unbound valproic acid concentration.

Masayuki Ishikawa, Masashi Uchida, Takahiro Asakawa, Shota Suzuki, Shingo Yamazaki, Yuki Shiko, Yohei Kawasaki, Takaaki Suzuki, Itsuko Ishii

Drug metabolism and pharmacokinetics 2023 Jun

filter terms:

In this study, we constructed a prediction formula for unbound valproic acid (VPA) concentration that was more accurate and widely applicable than previously reported formulae. A total of 136 datasets from 75 patients were analyzed retrospectively. The median of free fraction of VPA was 0.16 (interquartile range: 0.07; range: 0.07-0.45). The parameter that combined total VPA concentration (CtVPA) and serum albumin (SA), (CtVPA [μM] - 2 × SA [μM]), was significantly related to the free fraction of VPA (r = 0.76, p < 0.001). We constructed a combined parameter-based prediction formula for unbound VPA concentration. Analysis using external datasets from patients without severe renal failure showed that the prediction errors of the unbound VPA concentration were lower than those of previously reported formulae. Although the previous formulae showed large prediction errors, especially in the specific range of CtVPA values, the constructed formula showed a weak trend with CtVPA or SA. The formula based on (CtVPA [μM] - 2 × SA [μM]) had high prediction accuracy and wide applicability in predicting the unbound VPA concentration in patients without severe renal failure. Copyright © 2023 The Japanese Society for the Study of Xenobiotics. Published by Elsevier Ltd. All rights reserved.

Citation

Masayuki Ishikawa, Masashi Uchida, Takahiro Asakawa, Shota Suzuki, Shingo Yamazaki, Yuki Shiko, Yohei Kawasaki, Takaaki Suzuki, Itsuko Ishii. A novel method for predicting the unbound valproic acid concentration. Drug metabolism and pharmacokinetics. 2023 Jun;50:100503