Reproducibility of standard preparation in digoxin radioimmunoassay in plasma and serum

We studied the reproducibility of standard preparations in digoxin radioimmunoassay in a randomized trial using serum and plasma as matrices. The errors expressed relative to the observed counts per minute (cpm) attributable to each of the procedures involved in the preparation of standards were as follows: preparation of stock solutions and dilutions, 1.3%; addition of diluted solutions to the medium, 1.2%; and residual error due to the assay procedure, 3.7%. No error caused by mixing, portioning, and storage was detected. Heparinized plasma gave lower cpm values than serum at 4.0 ng/ml (p less than 0.01), an effect that would give over- or underestimations by about 5% at that level, depending on the medium used. This suggests that standard and sample matrices should be similar. Our procedure for preparing the standards seems to be reasonably reliable; this is necessary for satisfactory monitoring of patients on digitalis therapy.