Effectiveness of dual-chamber pacing in patients with dilated cardiomyopathy is still controversial. Our study was performed: to select the most favorable individual atrioventricular (AV) delay; to compare hemodynamic short-term effects in each patient after 2 periods of DDD pacing and sinus rhythm (AV spontaneous); to assess hemodynamic long-term (1 year) effects after DDD pacing at optimum AV delay. In 1996, 9 patients (7 men, 2 women; mean age 69 +/- 5 years) with dilated cardiomyopathy (5 idiopathic, 4 ischemic), NYHA functional class III-IV, ejection fraction < 30%, end-diastolic volume > 60 ml/m2, mitral regurgitation +2/+3, PR interval > or = 200 ms, were enrolled. All patients were implanted with DDD pacemakers and monitored for: ejection fraction and end-diastolic volume (measured by echocardiography and radionuclide angiography); clinical conditions; exercise tolerance and maximum oxygen consumption (by Weber exercise protocol); neurohormonal activity (plasma renin, aldosterone, atrial natriuretic factor). Data were recorded: before DDD implantation; after 2 randomized, single-blind periods of 3 months in VVI mode (at ventricular "sentinel" rate of 50 b/min) and in DDD mode with the optimum AV delay, corresponding for each patient to the minimum end-diastolic volume measured by radionuclide angiography and to the highest cardiac output recorded by echocardiography; after 6 months of DDD pacing with most favorable AV delay. Three more patients died 6 months after (between sixth and twelfth month of follow-up), due to refractory heart failure; 1 patient dropped out because his pacemaker was programmed in VVI mode at low rate, due to intolerance of DDD pacing. Among the other 4 patients no clinical and laboratory parameters were significantly different after 1 year of follow-up. In conclusion, DDD pacing in selected patients with dilated cardiomyopathy showed disappointing results, despite a strict and laboratory monitoring; DDD pacing could be of major benefit in larger populations, according to Doppler mitral flow pattern: those patients with a larger A-wave amplitude could be more sensitive to DDD pacing than those with evidence of poor atrial systole. Moreover, biatral and/or biventricular pacing could also play a significant role.