Background Voriconazole (VRCZ) is commonly used as oral and intravenous (IV) formulations. Few studies have comprehensively analysed the variation factors for the weight-corrected VRCZ serum concentration/dose (C/D) ratio based on the administration route. We retrospectively investigated the risk factors that influence the VRCZ C/D ratio in patients treated with oral or IV formulations.

Methods A total of 325 patients were divided into two groups (IV and oral groups). Propensity score matching was performed and linear regression analyses were used to identify the risk factors that affect the VRCZ C/D ratio according to the administration route. Receiver operating characteristic (ROC) curves were also used to assess the predictive potential for VRCZ trough concentration >5 µg/mL.

Results The VRCZ C/D ratio in the oral group was significantly lower than that in the IV group (p<0.001). Propensity score matching resulted in 65 in the IV group matched with 65 in the oral group. Multivariate analysis showed that age (p=0.039), aspartate aminotransferase (AST) (p=0.016) and total bilirubin (TBIL) (p=0.041) levels were independent influencing factors of the VRCZ C/D ratio in the oral group. ROC curves showed that the predicted probability of combined age, AST and TBIL had maximal area under the curve (AUC) of 0.901 for VRCZ trough level >5 µg/mL. Meanwhile, the ratio of TBIL (p=0.005) and single dose (p=0.015) were independent factors in the IV group with ROC_AUC of 0.781.

Conclusions To obtain optimal VRCZ efficacy and safety, dose adjustment is required based on multiple factors that may cause the observed difference in the VRCZ C/D ratio and trough levels between oral and IV administration.