Comparison of an HPLC-UV system (LM1010) and UPLC-MS/MS for plasma voriconazole measurement in routine clinical practice

We aimed to evaluate the clinical utility of the LM1010 high-performance liquid chromatography with ultraviolet-visible detection instrument for measuring plasma concentrations of voriconazole (VRCZ) under routine clinical conditions, including in patients with diverse clinical backgrounds. In total, 213 samples were collected from two hospitals (groups A and B). The plasma samples were analyzed using LM1010 and ultra performance liquid chromatography (UPLC)-tandem mass spectrometry (MS/MS). Laboratory test values and information on concomitant medications were collected from electronic medical records. In total, 66 samples from group A and 135 samples from group B were included in the analysis. Passing–Bablok regression analysis showed that in group A (n = 66), the slope was 0.972 (95% confidence interval [CI], 0.929–1.000) and the intercept −0.024 (95% CI, −0.100–0.056), indicating no significant proportional or constant bias. In group B (n = 135), the slope was 0.956 (95% CI, 0.930–0.975) and the intercept −0.006 (95% CI, −0.053–0.025), indicating a slight proportional bias but no constant bias. The Bland–Altman analysis demonstrated proportional bias in both groups A and B, and the slopes of the regression lines were 0.045 (95% CI, 0.011–0.079) and 0.038 (95% CI, 0.025–0.052), respectively. Interfering peaks were observed in the chromatograms of eight samples in group B, but their effect on the intermethod difference was small. No clinical laboratory test values or concomitant medications were identified as factors affecting the intermethod difference. The measurement of VRCZ plasma concentration by LM1010 was robust under heterogeneous clinical conditions and was useful for clinical therapeutic drug monitoring of VRCZ.