Propranolol and the sodium-channel-blocking antiarrhythmic agent propafenone share structural features. Although propafenone's beta-blocking actions are readily demonstrable in vitro, clinically significant beta-blockade occurs inconsistently in vivo. In this study, we tested the hypothesis that genetically determined variations in the biotransformation of propafenone to its 5-hydroxy metabolite account for variations in the drug's beta-blocking action. We assessed beta-blockade by measuring the reduction in tachycardia produced by boluses of isoproterenol and treadmill exercise in 14 normal subjects during treatment with placebo and with 150, 225, and 300 mg of propafenone every eight hours for five days each. Nine subjects (with the extensive-metabolizer phenotype) metabolized most of the propafenone to 5-hydroxy propafenone, and five (with the poor-metabolizer phenotype) did not produce this metabolite. At the lower dosages, beta-blockade was present in both groups but was significantly greater in the subjects with poor metabolism, in whom deficient 5-hydroxylation was associated with higher plasma propafenone levels. At the highest dose, a similar degree of beta-blockade was observed in the two groups. Propafenone also had a higher affinity for beta 2 receptors in vitro than either of its major metabolites. We conclude that the degree of beta-blockade during propafenone therapy reflects genetically determined variations in the metabolism of the parent drug, which is necessary for beta-blockade, and that this action of propafenone is considerably enhanced in patients with deficient 5-hydroxylation of propafenone.