Practical value of a new serum digitoxin assay.

The purpose of this study was to evaluate a new fluorescence polarization immunoassay, TDx, for digitoxin by comparing the results of this assay with those of a radioimmunoassay (RIA). Thirty-three serum samples were obtained from 15 patients during, and for 4 weeks after, a 4-week course of digitoxin therapy. Each sample was separated by centrifugation, coded, and frozen until analysis. At the time of analysis, each sample was divided and analyzed simultaneously by TDx and RIA. Nine samples yielded results less than 2 ng/ml (limit of assay sensitivity) by one or both methods and were excluded from further data analysis. Linear regression analysis of the results of the remaining 24 paired samples (x = TDx, y = RIA) revealed a strong correlation coefficient of r2 = 0.95, slope = 0.95, and a y intercept of -0.99 (y = -0.99 + 0.95x). Additionally, the TDx results were lower than the RIA values in only five of 33 paired samples; and these occurred in four patients who had a significantly lower mean estimated creatinine clearance than that of the other 11 patients (39.0 +/- 9.1 ml/min/1.73 m2 vs. 63.3 +/- 11.8 ml/min/1.73 m2, p less than 0.01). The TDx system is a comparable alternative to the RIA method, but differences in specificity and sensitivity may exist and should be evaluated more thoroughly.