The aim of this study was to compare AAIR and DDDR pacing at rest and during exercise. We studied 15 patients (10 men, age 65 +/- 6 years) who had been paced for at least 3 months with activity sensor rate modulated dual chamber pacemakers. All had sick sinus syndrome (SSS) with impaired sinus node chronotropy. The patients underwent a resting echocardiographic evaluation of systolic and diastolic LV function at 60 beats/min during AAIR and DDDR pacing with an AV delay, which ensured complete ventricular activation capture. Cardiac output (CO) was also measured during pacing at 100 beats/min in both pacing modes. Subsequently, the oxygen consumption (VO2AT) and VO2AT pulse at the anaerobic threshold were measured during exercise in AAIR mode and in DDDR mode with an AV delay of 120 ms. The indices of diastolic function showed no significant differences between the two pacing modes, except for patients with a stimulus-R interval > 220 ms, for whom the time velocity integral of LV filling and LV inflow time were significantly lower under AAI than under DDD pacing. At 60 beats/min, CO was higher under AAI than under DDD mode only when the stimulus-R interval was below 220 ms. For stimulus-R intervals longer than 220 ms, and also during pacing at 100 beats/min, the CO was higher in DDD mode. The stimulus-R interval decreased in all patients during exercise. The time to anaerobic threshold, VO2AT, and VO2AT pulse showed no significant differences between the two pacing modes. Our results indicate that, at rest, although AAIR pacing does not improve diastolic function in patients with SSS, it maintains a higher CO than does DDDR pacing in cases where the stimulus-R interval is not excessively prolonged. On exertion, the two pacing modes appear to be equally effective, at least in cases where the stimulus-R interval decreases in AAIR mode.