Influence of assay methods on serum concentrations of digoxin during FAB fragment treatment

The treatment of digoxin intoxication has been revolutionized by digoxin specific antibody fragments (Fab). Serum digoxin concentrations may be inaccurate after this treatment. We report a case of digoxin intoxication where the results of serum concentration determinations were strikingly disparate depending on the assay used. To investigate this discrepancy we compared serum samples spiked with digoxin from 0-50 ng/mL in the presence of increasing concentrations of digoxin specific Fab-fragments. Samples were measured using the Abbott TDx assay with and without ultrafiltration of the sample and the Dade-Stratus radial partition assay. The TDx assay was statistically reduced by the Fab-fragments although the magnitude of the effect was small. The radial partition assay was dramatically affected by the addition of Fab-fragments. The predicted non-Fab bound concentration correlated highly with the measured concentration. When samples were ultrafiltered prior to TDx assay, the measured concentration was dramatically depressed but the regression of predicted non-Fab bound concentration versus observed had a significantly lower slope than for the radial partition assay. We hypothesize that this difference is due to serum protein binding in addition to Fab-fragment binding. We conclude that the radial partition assay gives the best approximation of digoxin concentration remaining unbound to Fab-fragments. Ultrafiltration followed by TDx assay gives an acceptable approximation.