Cardiopulmonary exercise testing in mild heart failure: impact of the mode of exercise on established prognostic predictors.

OBJECTIVES In patients with heart failure (HF), peak oxygen consumption (peak VO(2)), the relationship between minute ventilation and carbon dioxide production (VE/VCO(2) slope) and heart rate recovery (HRR) are established prognostic predictors. However, treadmill exercise has been shown to elicit higher peak VO(2) values than bicycle exercise. We sought to assess whether the VE/VCO(2) slope and HRR in HF also depend on the exercise mode. METHODS Twenty-one patients with mild HF on chronic beta-blocker therapy underwent treadmill and bicycle cardiopulmonary exercise testing for measurement of peak VO(2) and the VE/VCO(2) slope. In patients with sinus rhythm (n = 16), HRR at 1 (HRR-1) and 2 min (HRR-2) after exercise termination was assessed. RESULTS Peak VO(2) was higher during treadmill as compared with bicycle testing (21.7 +/- 4.6 vs. 19.6 +/- 3.4 ml/kg/min; p = 0.006). HRR-1 tended to be slower (15 bpm, interquartile range 8-19, vs. 18 bpm, interquartile range 11-22; p = 0.16), and HRR-2 was significantly slower after treadmill exercise (26 bpm, interquartile range 20-39, vs. 31 bpm, interquartile range 22-41; p = 0.04). In contrast, VE/VCO(2) slope values did not differ between the test modes (32.9 +/- 5.5 vs. 32.3 +/- 5.0; p = 0.56). CONCLUSIONS In contrast to peak VO(2) and HRR, the VE/VCO(2) slope is not affected by the exercise mode in patients with mild HF.