Comparison of the Beckman TL-100 ultracentrifuge with the Amicon Centrifree Micropartition filters to determine protein-free phenytoin concentrations.

Non-protein-bound phenytoin levels obtained by ultracentrifugation (UC) with the Beckman TL-100 desk-top ultracentrifuge were compared with those obtained by ultrafiltration (UF) using the Amicon Centrifree Micropartition device. Total phenytoin concentrations were measured by an enzyme multiplied immunoassay technique (EMIT) adapted to a Cobas Bio centrifugal analyzer, whereas free levels were estimated by a radiometric assay. The following conditions for UC were established: 350,000 g for 1 h, sample volume 250 microliter. From three pooled human plasmas containing 18.8, 44.4, and 73.9 microM total phenytoin, the following percentages of free phenytoin levels were obtained with UF: 10.4 +/- 0.1 (SD), 10.7 +/- 0.6, and 11.1 +/- 0.1%. The corresponding figures with UC were 10.8 +/- 0.9, 11.4 +/- 0.4, and 12.0 +/- 0.6%, respectively. Within-sample precision showed a coefficient of variation between 3.5 and 8.0% for UC, with UF corresponding values between 1.0 and 5.6%. Free phenytoin levels in 19 patients undergoing phenytoin therapy were between 6.8 and 19.3% with UF (mean 10.1 +/- 3.0%) and 8.7 and 23.9% (mean 11.9 +/- 3.5%; p less than 0.001 by paired t test) with UC. Least-squares linear regression analysis resulted in the following equation: percentage free phenytoin by UC = 1.11 x % free phenytoin by UF - 0.70 (r = 0.935, p less than 0.001). The results indicate that within certain reservations UC under the conditions tested may be an alternative to UF or equilibrium dialysis.