Phenytoin as a probe of drug metabolism. Predicting clearance with a salivary sample.

Phenytoin (PHT) was administered in single 300-mg doses to each of 12 healthy, male subjects. Serial blood samples and salivary samples were collected for the next 48 h, and concentrations of plasma total PHT, unbound plasma PHT (plasma ultrafiltrates), and salivary PHT (salivary ultrafiltrates) were measured by immunofluorescence polarization. The following parameters were estimated: CLint/F, CL'int/F, V/F, and total and unbound PHT plasma disappearance half-lives. Estimates of CL'int/F (CL'int/F) were calculated from single 48-hour salivary PHT measurements. Mean (+/- SD) values were 0.026 +/- 0.009 1 . h-1 . kg-1, 0.385 +/- 0.148 1 . h-1 . kg-1, 12.6 +/- 3.5 l/kg (referenced to unbound drug), 28.0 +/- 12.1 h, and 25.9 +/- 13.5 h, respectively. When V/F referenced to unbound PHT was set at 14 l/kg, CL'int/F estimates were good predictors of the actual CL'int/F values demonstrating a mean prediction error of -9.012 1.h-1.kg-1. These data demonstrate that under specified conditions, intrinsic unbound PHT clearance can be estimated from a single PHT measurement in saliva, thereby permitting PHT to be used safely and sampled simply and noninvasively as a probe of hepatic mixed function oxygenase activity in humans.