Monitoring phenytoin therapy using citric acid-stimulated saliva in infants and children.

Two factors have limited the use of saliva in monitoring phenytoin therapy: availability of adequate volume of clear saliva and lack of a sensitive phenytoin assay. The applicability of citric acid-stimulated saliva and of a sensitive analytical assay (fluorescence polarization immunoassay, "TDx" Abbott) was evaluated in this study. Phenytoin was measured in paired plasma-saliva specimens from epileptic children during the long-term or the initial phase of phenytoin therapy. Analysis was carried out in plasma and in the clear supernatant of saliva (following centrifugation). Pooled-estimate SD of the analytical assay variability was 0.175 micrograms/ml for plasma total phenytoin, 0.063 for plasma free phenytoin, and 0.009 for saliva phenytoin. Recovery measurements of phenytoin spiked into saliva samples gave a coefficient of variation of less than 5%. Correlations between saliva and total plasma phenytoin levels and between saliva and free plasma phenytoin levels were strong and highly significant (r = 0.99, p less than 0.01). The percentage of temporal fluctuation (as determined by saliva phenytoin profiles) during 10-24 h ranged between 25.5-177 (mean, 58.3; SD, 47.3). Ratios of plasma total phenytoin/saliva phenytoin and of plasma free phenytoin/saliva phenytoin levels were 9.54 +/- 1.05 and 0.71 +/- 0.09, respectively. Dialysis experiments showed no binding of phenytoin to saliva supernatant. The greater saliva phenytoin concentrations as compared to plasma free phenytoin concentrations could be due to active transport of phenytoin from plasma to saliva. Measurement of phenytoin in citric acid-stimulated saliva by fluorescent polarization immunoassay is a reliable, noninvasive, and convenient method for monitoring phenytoin therapy in children.