Hemodynamic determinants of exercise capacity in chronic atrial fibrillation

To evaluate the response of patients with chronic atrial fibrillation (AF) to exercise, 79 male patients (mean age 64 ± 1 years) with AF underwent resting two-dimensional and M-mode echocardiography and symptom-limited treadmill testing with ventilatory gas exchange analysis. Patients were classified by underlying disease into five subgroups: no underlying disease (LONE: n = 17), hypertension (HT: n = 11), ischemic heart disease (n = 13), cardiomyopathy or history of congestive heart failure (CHF: n = 26), and valvular disease (n = 12). A higher maximal heart rate than expected for age was observed (175 vs 157 beats/min), which was most notable in the LONE and HT subgroups. Maximal oxygen uptake (VO2 max) was lower than expected for age in all groups. Patients with CHF had a lower resting ejection fraction than all other patients (p < 0.001), a lower VO2 max, and a lower maximal heart rate than LONE and HT patients (p < 0.001). Stepwise regression analysis demonstrated that echocardiographic measurements at rest were poor predictors of VO2 max and VO2 at the ventilatory threshold. Among clinical, morphologic, and exercise variables, maximal systolic blood pressure accounted for the greatest variance in exercise capacity, but it explained only 35%. In patients with AF the higher than predicted maximal heart rates may be a compensatory mechanism for maintaining exercise capacity after the loss of normal atrial function. However, even in the absence of underlying disease, it does not appear to compensate fully for a compromised exercise capacity. Although the exercise response was not strongly influenced by cardiac function at rest, the response of patients with AF without morphologic heart disease differed markedly from those with AF and underlying CHF. Although AF is associated with a reduced exercise capacity, the response to exercise in patients with AF is related more to the underlying heart disease than to AF itself.