Abstract
The pharmacokinetics of voriconazole shows large intra-individual and inter-individual variability and is affected by various factors. Recently, inflammation has been focused as a significant factor affecting the variability. This study aimed to compare the influence of C-reactive protein (CRP) and other clinical laboratory parameters on intra-individual variability in trough voriconazole concentration and examine the impact of inflammation in patients with hematological malignancies. We conducted a retrospective, single-center, observational cohort study. Forty-two patients with hematological malignancy who received oral voriconazole for prophylaxis against deep mycosis and underwent multiple measurements of trough plasma voriconazole concentration were recruited. Quantitative changes in pharmacological and clinical laboratory parameters (Δ) were calculated as the difference between the current and preceding measurements. Voriconazole concentration/maintenance dose per weight (C/D) was found to correlate positively with CRP level (n = 202, rs = 0.314, p < 0.001). Furthermore, ΔC/D correlated positively with ΔCRP level (n = 160, rs = 0.442, p < 0.001), and ΔCRP showed the highest correlation coefficient among the laboratory parameters. Univariate and multivariate analyses identified ΔCRP (p < 0.001) and Δgamma-glutamyl transpeptidase (γGTP) (p = 0.019) as independent factors associated with ΔC/D. Partial R2 were 0.315 for ΔCRP and 0.024 for ΔγGTP, suggesting markedly greater contribution of ΔCRP to ΔC/D. In conclusion, since clinical laboratory parameters other than CRP had little influence on trough plasma voriconazole concentration, therapeutic drug monitoring and dose adjustment considering fluctuation in CRP level would be important for proper use of voriconazole in patients with hematological malignancies.
