Preliminary data suggests that vital physiologic support measures and free fraction of phenytoin are altered following head trauma. Therefore, we conducted a prospective, randomized controlled study to determine the correlation of albumin concentration and unbound phenytoin plasma concentration following head injury. Ten adult head trauma patients in the neurosurgical intensive care unit receiving phenytoin for the seizure prophylactic treatment were studied for their free and total plasma phenytoin concentration in peak and trough times and their respective albumin concentration. Free and total phenytoin levels were determined by both liquid chromatography and florescence polarization immunoassay (Eclair) of plasma samples after ultrafiltration and deproteinization. No significant difference was found in the plasma concentration measured with HPLC or FPIA while a marked correlation was noted between plasma albumin and free phenytoin concentration (r<sup>2</sup>=0.85). The total and free phenytoin concentrations were not significantly correlated (r<sup>2</sup>=0.60). A remarkable difference (P<0.05) was noticed when doses in patients were adjusted on the basis of total plasma phenytoin and calculated plasma phenytoin adjusted for serum albumin. Therefore, therapeutic monitoring in neurosurgical patients receiving phenytoin should be performed on the basis of pharmacologically active component (free fraction), rather than total phenytoin which is presently performed in the clinics.