OBJECTIVE The aim of the study was to evaluate the effect of chronic treatment by amiodarone on beta adrenoceptor density and adenylate cyclase response in human myocardium. METHODS Density of beta 1 and beta 2 adrenoceptors was measured by radioligand binding assay. beta Adrenoceptor stimulated production of cAMP was measured by adenylate cyclase assay. METHODS Right auricular tissue from five patients on chronic amiodarone treatment was compared with that from nine patients in similar clinical and haemodynamic state undergoing coronary bypass surgery. RESULTS beta 1 and beta 2 adrenoceptor subtypes were quantified using the highly beta 1 selective antagonist Sandoz 204 545. The total beta adrenoceptor density was 28% lower in the amiodarone treated group than in the controls (42.0 v 58.3 fmol.mg-1 protein, p less than 0.02), beta 1 adrenoceptors were 25% lower (33.1 v 44.3 fmol.mg-1 protein, p less than 0.02), and beta 2 adrenoceptors were 36% lower (8.9 v 14.0 fmol.mg-1 protein, p less than 0.02). The cAMP production following non-selective beta adrenoceptor stimulation (isoprenaline 5 mumol.litre-1) was reduced by 38% in the amiodarone treated group (14.2 to 8.7 pmol.min-1.mg-1 protein, p = 0.05). Terbutaline stimulated cAMP production was reduced by 49% (8.3 to 4.3 pmol.min-1.mg-1 protein, p = 0.03). Fluoride stimulated cAMP production was not significantly different (9.4 v 8.4 pmol.min-1.mg-1 protein, p = 0.15). CONCLUSIONS Chronic treatment with amiodarone is associated with a non-selective downregulation of beta adrenoceptors. beta Adrenoceptor stimulated cAMP production was also reduced. The "beta blocking effect" of amiodarone is probably related to downregulation of beta adrenoceptors.