Tuberculous meningitis (TBM) is the most severe manifestation of tuberculosis. Pyrazinamide (PZA) is a pivotal antituberculous drug, but its dose has not been optimised for TBM. The aims of this study were to describe the pharmacokinetics of PZA during TBM treatment, to identify predictors of PZA exposure and to assess relationships between PZA dose and exposures in plasma and CSF. Plasma PZA pharmacokinetic data were assessed on Days 2 and 10 of treatment in 52 adult TBM patients. A CSF-to-plasma concentration ratio was determined on Day 2. Predictors of plasma PZA exposure, correlation between plasma and CSF exposures, and prediction of CSF concentrations based on dose and plasma exposure were evaluated. The geometric mean plasma PZA exposure (AUC0-24) and peak concentration (Cmax) on Day 2 were 709 h·mg/L and 59 mg/L following a median dose of 33.3 mg/kg/day; AUC0-24 on Day 10 (523 h·mg/L) was lower (P < 0.001). Dose and BMI correlated with AUC0-24 and Cmax. The CSF concentration at 3-6 h was 42 mg/L and the CSF-to-plasma ratio was 90%. AUC0-24, Cmax and CSF concentration were highly correlated. CSF concentration could be predicted based on dose and various plasma exposure measures with <5% bias and <21% imprecision. Exposure to PZA decreases during the first days of TBM treatment, possibly due to the evolving inductive effect of rifampicin. PZA penetrates well in CSF. The association between PZA dose and exposures in plasma and CSF provides a rationale to study higher PZA doses for TBM.
Keywords: Pharmacokinetics; Pyrazinamide; Tuberculous meningitis.
Copyright © 2019. Published by Elsevier B.V.