The results of the present study show that the reduction of the total number of beta adrenoceptors affected the beta 1-adrenoceptor subpopulation, whereas the beta 2 adrenoceptors were not detectably altered in the failing heart. Dopexamine had a 9.8-fold greater affinity to beta 2 adrenoceptors than to beta 1 adrenoceptors. In nonfailing myocardium, dopexamine increased force of contraction concentration-dependently. However, dopexamine alone had no effect in papillary muscle strips from moderately (NYHA II-III) and severely (NYHA IV) failing myocardium. However, in the presence of milrinone, it concentration-dependently increased force of contraction. Under this condition, the effectiveness was slightly less pronounced in NYHA IV than in NYHA II-III. Dopexamine concentration-dependently stimulated adenylate cyclase activity. Experiments with the beta 1-selective antagonist CGP 207.12 A and the beta 2-selective antagonist ICI 118.551 showed that both stimulation of adenylate cyclase and the increase of force of contraction are mediated by beta 2 adrenoceptors. It is concluded that although the number of beta 2 adrenoceptors is preserved in the failing myocardium, dopexamine alone does not increase force of contraction. However, the positive inotropic effect of dopexamine, which is observed in the presence of milrinone and the stimulation of adenylate cyclase activity by dopexamine are mediated by beta 2 adrenoceptors. Therefore, beta 2 adrenoceptors exist in the human myocardium, are coupled to adenylate cyclase, and are capable of increasing force of contraction. These results provide evidence for an impaired coupling of beta 2 adrenoceptors to mechanisms beyond receptor occupation in terminal heart failure.