Modern pacemakers offer many programming options regarding the AV interval including the ability to vary AV intervals depending on whether atrial activity is paced or spontaneous and to shorten AV intervals with increasing rates. To determine if optimization of these features improves exercise tolerance, 14 patients with intact sinus node function and AV block treated with dual chamber pacemakers were enrolled in a randomized double-blind crossover trial. Doppler echocardiographic measurements of cardiac index and mitral flow were assessed over a range of programmable AV intervals at rest to determine each patient's optimal AV interval. Eleven patients completed serial graded exercise tests with spiroergometry after randomly programming the AV interval three ways in a crossover manner: fixed AV interval = 150 ms without rate adaptation (150/Fixed), fixed AVinterval = 150 ms with rate adaptation (150/R), or optimized AV interval with rate adaptive AV interval shortening (optimized/R). Exercise capacity was determined by maximum oxygen uptake. Ten men and four women, age 64 +/- 8 years, were enrolled. At rest, optimization of the AVintervalimproved the cardiac index by 21% (P < 0.001) and mitral flow by 13.4% (P < 0.001) when compared to least-favorable AV intervals. During exercise, no differences in maximum heart rates were noted. Maximum oxygen uptake was increased in both groups with rate adaptive AVinterval shortening when compared tofixed AVinterval without rate adaptation: 13.9% (adjusted P < 0.04) and 14.6% (adjusted P < 0.02) in optimized/R and 150/R, respectively. No differences were noted between optimized/R and 150/R. In conclusion, rate adaptive AV interval shortening improved exercise tolerance independent of changes in heart rate. However, optimization of the AV interval with Doppler echocardiography at rest did not further improve exercise capacity.