Voriconazole therapeutic drug monitoring and safety in HIV-infected patients with invasive fungal disease.

OBJECTIVE

To characterize the distribution of voriconazole (VRC) trough concentrations (C trough), identify factors associated with VRC exposure in HIV-infected patients complicated by invasive fungal disease (IFD), and determine the C trough threshold associated with an increased risk of adverse drug reactions (ADRs).

METHODS

A retrospective study was conducted at the First Hospital of Changsha from 1 January 2021, to 31 December 2025. Univariate and multivariate analyses were performed to identify factors associated with VRC C trough. Receiver operating characteristic (ROC) curve analysis was used to determine the optimal cutoff value for predicting ADRs.

RESULTS

A total of 44 patients contributed 55 VRC C trough measurements. The median C trough was 2.44 mg/L (range, 0.07-7.26 mg/L). Of these measurements, 8 (14.5%) were <1.0 mg/L, 39 (70.9%) were within the therapeutic range (1.0-5.5 mg/L), and 8 (14.5%) were >5.5 mg/L. The interindividual coefficient of variation for C trough was 63.6%. Univariate analysis showed that albumin-bilirubin (ALBI) grade, Child-Pugh classification, hypoalbuminemia, concomitant glucocorticoid use, C-reactive protein levels, and CD4+/CD8+ ratio were associated with VRC C trough. The mixed-effects model identified ALBI grade, concomitant glucocorticoid use, and the CD4+/CD8+ ratio as factors significantly associated with VRC C trough. The incidence of ADRs was 20.5% (9/44). Patients who developed ADRs had significantly higher median VRC C trough than those who did not (5.15 mg/L [range, 0.88-5.93 mg/L] vs. 2.40 mg/L [range, 0.07-7.26 mg/L], P = 0.038). ROC analysis yielded an area under the curve (AUC) of 0.720 (95% confidence interval, 0.525-0.914; P = 0.038) for C trough in predicting ADRs. The optimal cutoff value was 4.43 mg/L, with a sensitivity of 66.7% and a specificity of 80.4%.

CONCLUSION

In patients with HIV-infected and IFD, VRC C trough showed substantial variability and was influenced by multiple factors, including ALBI grade, concomitant glucocorticoid use, and the CD4+/CD8+ ratio. Maintaining VRC C trough < 4.43 mg/L may help reduce the risk of ADRs in this population.