鈥擠r Reider's comments on our recent article in the Archives1 raise a fundamental question of the reliability of digoxin radioimmunoassay (RIA). Discrepancies in the measurements made with different kits are well recognized.2,3 Analytical recovery studies have shown that RIA consistently overestimates serum digoxin levels.2 Whether this is due to the endogenous digoxinlike substance (endoxin) has not been conclusively demonstrated. In our study, the calculated digoxin level underestimated the serum levels in five patients and overestimated it in two (patients 5 and 14). This observation indicates that the presence of endoxin cannot be the sole explanation for the discrepancy between the measured and calculated serum digoxin levels. Digoxin concentration measurements done by commercially available RIA methods may not be optimal.2,3 However, the purpose of our study was not to establish the reliability of digoxin RIA, but to determine the predictability of immunoassayable serum digoxin concentrations measured by