OBJECTIVE This study was conducted to determine the relations among exercise capacity and pulmonary, peripheral vascular, cardiac and neurohormonal factors in patients with chronic heart failure. BACKGROUND The mechanisms of exercise intolerance in heart failure have not been fully clarified. Previous studies have indicated that peripheral factors such as regional blood flow may be more closely associated with exercise capacity than cardiac function, whereas the role of pulmonary function has received less attention. METHODS Fifty patients with stable heart failure underwent a comprehensive assessment that included a symptom-limited maximal cardiopulmonary exercise test, right heart catheterization, pulmonary function tests, neurohormonal levels, radionuclide ventriculography and forearm blood flow at rest and after 5 min of brachial artery occlusion. Univariate and stepwise linear regression analyses were used to relate peak exercise oxygen uptake to indexes of cardiac, peripheral vascular, pulmonary and neurohormonal factors both alone and in combination. RESULTS The mean ejection fraction was 19% and peak oxygen uptake was 16.5 ml/min per kg in this group of patients. By univariate analysis, there were no significant correlations between peak oxygen uptake and rest cardiac output, pulmonary wedge pressure, ejection fraction and pulmonary or systemic vascular resistance. In contrast, even in the absence of arterial desaturation during exercise, the forced expiratory volume in 1 s (r = 0.55, p < 0.001), forced vital capacity (r = 0.46, p < 0.01) and diffusing capacity for carbon monoxide (r = 0.47, p < 0.01) were all significantly associated with peak oxygen uptake. Peak postocclusion forearm blood flow (r = 0.45, p < 0.01), the corresponding minimal forearm vascular resistance (r = -0.56; p < 0.01) and plasma norepinephrine level at rest (r = -0.45; p < 0.01) were also significantly correlated with peak oxygen uptake. By multivariate analysis, minimal forearm vascular resistance and forced expiratory volume in 1 s were shown to be independently related to peak oxygen uptake, with a combined R value of 0.71. Other two-variate models included forced expiratory volume and plasma norepinephrine (R = 0.67) and forced expiratory volume and diffusing capacity (R = 0.65). Because forced vital capacity was highly correlated with forced expiratory volume in 1 s, it could be combined with the same variables to yield similar R values. Addition of any third variable did not improve these correlations. CONCLUSIONS In comparison with rest indexes of cardiac performance, measures of pulmonary function and peripheral vasodilator capacity were more closely associated with peak exercise oxygen uptake in patients with heart failure. Furthermore, the associations were independent of each other and together accounted for 50% of the variance in peak oxygen uptake. These data suggest that pulmonary and peripheral vascular adaptations may be important determinants of exercise intolerance in heart failure.