Objective: To assess the performance of the Sheiner-Tozer equation when used to evaluate total phenytoin concentrations in patients with pronounced hypoalbuminemia.
Design: Patients with phenytoin concentrations drawn at least 5 days after initiation of therapy and with serum albumin concentrations of less than 25 g/L were identified during routine daily monitoring. Phenytoin samples were frozen at -30 degrees C, batched, and later thawed for total and free phenytoin assay. Separation of free phenytoin concentration was performed at 37 degrees C using a Centrifree micropartition filter.
Setting: A 400-bed university teaching hospital.
Patients: Twenty-nine adults with hypoalbuminemia receiving phenytoin therapy. Patients receiving drugs known to displace phenytoin, or those with renal failure, abnormal liver enzymes, or increased bilirubin concentrations were excluded.
Main outcome measure: Precision and bias of the normalized result using the Sheiner-Tozer equation were assessed with respect to observed phenytoin concentrations.
Results: The Sheiner-Tozer equation underpredicts the measured free concentration by approximately 12.4%. The equation provides a small but statistically significant bias (-2.7 mg/L) and a root mean squared error significantly different from 0. For most of our patients these departures appear small enough to warrant an initial empiric appraisal of hypoalbuminemia using the Sheiner-Tozer method.
Conclusions: The Sheiner-Tozer equation can normalize total phenytoin concentrations reliably in patients with low albumin concentrations at our institution.