Abstract

A rapid radio immunoassay technique has been developed for the determination of serum digitoxin concentration in the range encountered clinically. Unlabeled digitoxin in the unknown sample competes with a tritiated digitoxin tracer for binding sites of high affinity rabbit antibodies to a human serum albumin-digoxin conjugate (average intrinsic affinity constant for digitoxin, 5.3 x 10⁸ M^-1). Free labeled digitoxin is separated from the antibody-bound fraction by adsorption to dextran-coated charcoal. The entire procedure is carried out in whole serum, obviating the problems of extraction of the drug prior to assay. The method is sensitive (2 ng/ml or less), precise (standard deviation for replicate samples ±5%), specific (no false positives in patients not receiving cardiac glycosides) and requires only one hour for completion. Six patients with electrocardiographic evidence of digitoxin intoxication had a mean serum digitoxin concentration of 34 ng/ml (S.D.6, range 26-43) which differed significantly (P ≤.001) from the mean of 17 ng/ml (S.D. 8, range 3-39) in 52 nontoxic patients. The mean age of 71 years in toxic patients tended to be greater than that of the nontoxic group (P ≤.05), whereas sex, weight, serum potassium concentration, daily maintenance digitoxin dosage and renal function as judged by blood urea nitrogen values were not significantly different. A higher incidence of coronary artery disease was found in the toxic compared with the nontoxic group (P ≤.05). A reliable estimate of serum digitoxin concentration, considered in the clinical context, has been found to be useful in arriving at acute management decisions.