Objective We aimed to investigate (1) the effects of aerobic interval training (AIT) and aerobic continuous training (ACT) on (sub)maximal exercise measures and its determinants including endothelial function, muscle strength and cardiac autonomic function, and (2) the relationship between exercise capacity and these determinants. Methods Two-hundred coronary artery disease (CAD) patients (58.4 +/- 9.1 years) were randomized to AIT or ACT for 12 weeks. All patients performed a cardiopulmonary exercise test and endothelial function measurements before and after the intervention; a subpopulation underwent muscle strength and heart rate variability (HRV) assessments. Results The VO(2), heart rate and workload at peak and at first and second ventilatory threshold increased (P-time <0.001); the oxygen uptake efficiency slope (P-time <0.001) and half time of peak VO(2) (P-time <0.001) improved. Endothelial function and heart rate recovery (HRR) at 1 and 2 min improved (P-time <0.001), while measures of muscle strength and HRV did not change. Both interventions were equally effective. Significant correlations were found between baseline peak VO(2) and (1) quadriceps strength (r = 0.44; P < 0.001); (2) HRR at 2 min (r = 0.46; P < 0.001). Changes in peak VO(2) correlated significantly with changes in (1) FMD (rho = 0.17; P < 0.05); (2) quadriceps strength (r = 0.23; P < 0.05); (3) HRR at 2 min (rho = 0.18; P < 0.05) and Total power of HRV (rho = 0.41; P < 0.05). Conclusions This multicentre trial shows equal improvements in maximal and submaximal exercise capacity, endothelial function and HRR after AIT and ACT, while these training methods seem to be insufficient to improve muscle strength and HRV. Changes in peak VO(2) were linked to changes in all underlying parameters.