Exercise Oscillatory Breathing And Increased Ventilation To CO2 Production Slope In Heart Failure

Abstract

The minute ventilation-carbon dioxide production (VE/VCO2) slope and peak oxygen consumption (VO2) are established prognostic markers in heart failure (HF) patients. Recently, the occurrence of exercise oscillatory breathing (EOB) has emerged as an additional strong indicator of survival. PURPOSE: To compare the prognostic significance of EOV to established cardiopulmonary exercise test (CPX) variables. METHODS: Two hundred and eighty-eight stable HF patients (198 male/90 female, age: 55.4 ±13.3 years, ejection fraction: 32.5 ±12.5%) who underwent CPX participated in this study. EOB was denned as oscillatory fluctuations in VE for >60% of the exercise test at an amplitude >15% of the oscillatory fluctuations observed at rest. RESULTS: The mean tracking period was 20.8 ±13.0 months. Sixty-two subjects died from cardiac causes during the tracking period (annual mortality rate: 10.8%). One hundred and one subjects (35.1%) demonstrated EOB during CPX. The VE/VCO2 slope was significantly higher (39.4 ±8.5 vs. 33.0 ±7.8, p<0.001) while peak VO2 was significantly lower (13.2 ±3.9 vs. 16.8 ±5.1 mlO2· min-1·kg-1, p<0.001) in the EOB subgroup. Receiver operating characteristic curve analysis revealed the optimal prognostic threshold values for the VE/VCO2 slope and peak VO2 was 36.2 (Area: 0.68, 95% CI: 0.60-0.75, Sensitivity 71%/Specificity 64%, p<0.001) and 14.1 mlO2· min-1·kg-1 (Area: 0.63, 95% CI: 0.55-0.71, Sensitivity 63%/Specificity 63%, p=0.002), respectively. Univariate Cox regression analysis revealed EOB (Hazard ratio: 5.5, 95% CI: 3.2-9.5, p<0.001), the VE/VCO2 slope (Hazard ratio: 4.1, 95% CI: 2.4–6.9, p<0.001) and peak VO2 (Hazard ratio: 2.4, 95% CI: 1.4-4.1, p<0.001) were all significant predictors of mortality. In the multivariate Cox regression analysis, only EOB (Chi-square: 46.5, p<0.001) and the VE/VCO2 slope (Residual chi-square: 11.9, p=0.001) were retained. Peak VO2 did not add prognostic value and was removed from the regression (Residual chi-square: 1.6, p=0.21). CONCLUSIONS: These findings identify EOB as a strong predictor of cardiac mortality in patients with HF. The presence of EOB does not necessarily imply an elevated VE/VCO2 slope but the combination of both variables identifies patients at higher risk for adverse events.