Background: An increased minute ventilation (VE)/carbon dioxide production (VCO2) relationship, an expression of ventilatory inefficiency (VI), is associated with increased morbidity and mortality in patients with left ventricular systolic dysfunction (LVSD). A direct link between VI and a specific cardiac abnormality has not been established.
Methods: We analyzed cardiopulmonary exercise test (CPET) data from patients (N=83) with severe LVSD (ischemic and nonischemic; left ventricular ejection fraction [LVEF] 19%±7%) and at least moderate exercise intolerance. Subjects were stratified into two groups based on the (VE/VCO2 ratio at anaerobic threshold (VE/VCO2@AT) (group 1 VE/VCO2@AT≤34; group 2 VE/VCO2@AT>34). Clinical, CPET, echocardiographic, and hemodynamic data were compared between groups.
Results: Group 2 subjects had lower exercise capacity (peak (VO2, 45.7%±11.8% vs 50.4±8.9% predicted; P<.05), with a significantly lower oxygen pulse (71.6%±24.5% vs 85.4±18.5% predicted) and maximum systolic BP (122±19 mm Hg vs 138±22 mm Hg; P<.001 for both), suggesting a more blunted stroke volume to exercise vs group 1. There were no differences in left ventricular (LV) size, LVEF, or mitral regurgitation between the two groups. In sharp contrast, group 2 had larger right ventricular (RV) dimensions (4.5±1.1 cm vs 3.9±0.8 cm) and more severe RV systolic dysfunction (RV fractional area change 26%±11% vs 33%±12%; tricuspid annular plane systolic excursion [TAPSE] 1.6±0.5 cm vs 2.0±0.5 cm; all P<.001) vs group 1. Multivariable analysis revealed that only TAPSE and Doppler-estimated pulmonary artery systolic pressure were independently associated with VE/VCO2@AT and the (VE/VCO2slope. The VE/VCO2@AT, VE/VCO2 slope, and TAPSE had nearly identical predictive value for death or transplant.
Conclusions: The present study suggests that VI is a functional, noninvasive marker of more advanced right-sided heart dysfunction in patients with severe LVSD.