The influence of serum separator tubes (SSTs) on total and free phenytoin concentrations and phenytoin dosages was evaluated in patients treated with phenytoin. Thirty blood samples were obtained from 18 patients. Equal volumes of blood were placed into SSTs and plain tubes. Samples were centrifuged, and serum analyzed for total and free phenytoin by fluorescence polarization immunoassay (FPIA) at 2, 24, and 48 h. Total phenytoin concentrations collected in SSTs were significantly lower than those collected in plain tubes at 2 (12.3 vs 12.8 mg/L, p < 0.01), 24 (11.4 vs 12.9 mg/L, p < 0.01), and 48 h (10.9 vs 12.9 mg/L, p < 0.01). These differences resulted in significantly higher calculated maximum rates of metabolism (Vmax) and daily phenytoin dosages (R0) for SSTs at 24 (Vmax: 10.4 vs 10.1 mg/kg/day; R0: 8.2 vs 8.0 mg/kg/day, p < 0.01) and 48 h (Vmax: 10.8 vs 10.3 mg/kg/day; R0: 8.5 vs 8.1 mg/kg/day, p < 0.01). Free phenytoin concentrations from SSTs were significantly lower at 48 h (1.56 vs 1.61 mg/L, p < 0.05). However, there were no significant differences in calculated dosages. Observed statistical differences in total phenytoin concentrations can be clinically important for making dosage adjustments, especially in patients undergoing nonlinear elimination. Thus, SSTs should not be used to collect blood for total serum phenytoin determination.