Plasma protein binding (PPB) can possibly alter the already variable pharmacokinetics of voriconazole. Voriconazole PPB was determined only once, being 58%, according to equilibrium dialysis (ED). We investigated voriconazole PPB more in detail, with a convenient and newer high‐throughput ED assay (HT‐ED), in human blank plasma spiked with voriconazole and in plasma from intensive care unit (ICU) patients treated with voriconazole. HT‐ED was conducted in a 96‐well plate, setup against phosphate‐buffered saline. Voriconazole concentrations were measured by liquid chromatography–tandem mass spectrometry. The median PPB was 47.6% [interquartile range (IQR) 45.3%–50%] in vitro, and 49.6% (IQR 42.5%–52.5%) in ICU samples (p = 0.35), and is not depending on total voriconazole concentration (0.7–11.2 mg/L, p = 0.65). The drug mainly binds to albumin (25.5 ± 5.1%), and to a lesser extent to α‐1‐acid glycoprotein (AAG; 4.8 ± 1.2%). The HT‐ED assay can be performed at 37°C or 25°C (p = 0.44) and in batch: PPB variations during freeze–thaw cycles (p = 0.13) and during frozen storage up to 12 months (p = 0.10) were not clinically relevant. Voriconazole PPB is approximately 50%, according to HT‐ED. As albumin and AAG only account for approximately 30% of total voriconazole PPB, other plasma components could influence PPB and therefore efficacy or toxicity because of variations in unbound fractions. © 2014 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm Sci 103:2565–2570, 2014