Cardiopulmonary exercise parameters in relation to all-cause mortality in patients with chronic heart failure

Abstract

In this study we analysed the all-cause mortality ⋅over a period of maximal 6 years in 60 male patients (age: 63.4±8.3 years, mean±S.D.), suffering from chronic heart failure with resting left ventricular ejection fraction and ^⋅E/^⋅O2 slope as independent factors. We assessed functional NYHA class (II: n=36, III: n=24), radionuclide left ventricular ejection fraction (29.2±10.4%) and peak values of heart rate, ^⋅O2, ^⋅CO2, ^⋅E, anaerobic threshold and exercise duration with an incremental work load test on the treadmill. ^⋅O2 relative to ^⋅E was based on the individual slopes of the regression of ^⋅O2 on ^⋅E during the first 6 min of exercise. These slopes with other exercise-related variables and factors such as etiology, medication, and NYHA class were analysed with a Cox’s Regression Method. A survival time analysis (Kaplan-Meier survival curve) was done to establish the influence of ^⋅E/^⋅O2 slope and left ventricular ejection fraction (both split into above and below median values), as well as their interaction, on survival. From all investigated exercise-related variables, ^⋅E/^⋅O2 slope is the most powerful variable regarding prediction of all-cause mortality in our group of chronic heart failure patients. Concerning risk stratification, the subgroup (n=18) with a relatively high left ventricular ejection fraction (>28%) and flat ^⋅E/^⋅O2 slope (<27.6) had most survivors (77.8%) after about 3 years, while the subgroup (n=12) with a relatively high left ventricular ejection fraction (>28%), but a steep ^⋅E/^⋅O2 slope (>27.6) had least survivors (33.3%). This difference in percentage is highly significant (P=0.0025). The fact that ^⋅E/^⋅O2 slope and left ventricular ejection fraction show comparable main and interaction effects between measures of exercise tolerance (e.g., anaerobic threshold, peak ^⋅O2, exercise duration) on the one hand, and all-cause mortality on the other, suggests the existence of common sources of variance. Based on our analysis, it is unlikely that effects on all-cause mortality are mediated through phenomena related to exercise tolerance. Therefore, we hypothesize that the effects on exercise tolerance and all-cause mortality both depend on common factors, which cause both cardiac and peripheral organ (c.q. muscular) dysfunctions. Moreover, this study clearly shows that ^⋅E/^⋅O2 slope during incremental exercise is an important prognostic marker for risk stratification in chronic heart failure patients, NYHA class II and III.

Introduction

Chronic heart failure patients compared to normal subjects have relatively low left ventricular ejection fraction at rest and steep ^⋅E/^⋅O2 slopes during exercise [1], [2], [3], [4], [5]. Left ventricular ejection fraction has been reported as an independent predictor of mortality or heart transplantation in chronic heart failure patients [6]. Compared to normal left ventricular ejection fraction, a severe restriction in left ventricular ejection fraction (<30%) may lead to a reduced cardiac output during exercise, as resting left ventricular ejection fraction is critical to the magnitude of change during incremental exercise, whereas limitations in heart rate response in chronic heart failure patients reduce possible compensation [7]. A steep ^⋅E/^⋅O2 slope during exercise may be caused by either an excessive ventilation with oxygen consumption within ‘normal’ ranges or a ‘normal’ ventilation with a relatively low oxygen consumption. If a steep ^⋅E/^⋅O2 slope is based on excessive ventilation with a ‘normal’ oxygen consumption, then the oxygen cost of breathing will be high. This possibly induces a competition for oxygen between respiratory- and locomotor muscles, causing a reduced oxygen supply to the locomotor muscles [8], [9]. If, however, a steep ^⋅E/^⋅O2 slope is based on ‘normal’ ventilation and a relatively low oxygen consumption, then the slope may reflect a reduced oxygen supply to the skeletal muscles (peripheral supply) and/or limitations in peripheral oxygen extraction or utilization [4], [5], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26]. Therefore, a steep ^⋅E/^⋅O2 slope combined with a low baseline left ventricular ejection fraction may indicate limitations in both peripheral oxygen supply, oxygen extraction and/or utilization. In the case that a steep ^⋅E/^⋅O2 slope is combined with a mild restriction in baseline left ventricular ejection fraction, we hypothesize that such a combination may primarily reflect peripheral limitations in oxygen extraction and/or utilization, rather than limitations in peripheral supply. It is well known that steep ^⋅E/^⋅CO2 slopes and high ^⋅E/^⋅O2 ratios during exercise are associated with an increased mortality in chronic heart failure patients [1], [5], [27], [28]. In the light of these different pathophysiological phenomena, we were interested in the main and interaction effects of left ventricular ejection fraction and ^⋅E/^⋅O2 slope on several exercise-related cardiopulmonary variables and their relationship to survival. We analysed the all-cause mortality over a period of maximal 6 years in 60 chronic heart failure patients with resting left ventricular ejection fraction and exercise-related ^⋅E/^⋅O2 slope as independent factors. To collect relevant cardiopulmonary data, all patients underwent an incremental work load test on a treadmill.