Asymptomatic left ventricular dysfunction and cardiopulmonary exercise testing for mortality risk prediction

This editorial refers to ‘Cardiopulmonary exercise testing in patients with asymptomatic left ventricular dysfunction: lack of prognostic predictive power of ventilatory variables’, by U. Corrà et al., pp. 1158–1163. ‘Things are not always what they seem; the first appearance deceives many; the intelligence of a few perceives what has been carefully hidden’. —Gaius Julius Phædrus, (c. 15 BC–AD 50), Roman fabulist. The development and application of cardiopulmonary exercise test (CPET) for the evaluation of chronic heart failure has had a long and rich history.1–4 Unlike other modalities of provocative stress testing, CPET can provide valuable insight into cardiopulmonary interactions, pulmonary vascular response, and neurocardiac signal integration.3,5,6 Peak oxygen uptake (VO2) remains the gold-standard measure of exercise capacity and the body’s integrative responses to a defined regimen of incremental exercise stress. However, full exertion may not be possible in some patients with comorbidities and advanced cardiopulmonary disease. Over the past two to three decades, alternative parameters have been developed and explored as alternatives to maximal or peak VO2. By logarithmic transformation of serial measurements of VO2, Baba et al.7 first derived a correlation coefficient termed oxygen uptake efficiency slope (OUES) that enabled effort-independent assessment of exercise capacity. Likewise, the slope of minute ventilation/carbon dioxide production (VE/VCO2) representing ventilatory efficiency was found to be of prognostic significance in patients with chronic heart failure,8–10 and provided even greater discriminatory power for predicting survival than peak VO2 in advanced heart failure.11 At the other end of the spectrum, in a community-based population of mostly pre-frail, ambulatory older adults with few comorbidities, maximal VO2 was evidently superior to OUES in stratification and prediction of mortality hazard.12 With an increasing number of CPET parameters and disease subtypes, the clinician must therefore exercise judgement and wisely decide on the appropriate parameter(s) to use for assessment and disease monitoring, depending on patient factors including body habitus, functional capabilities and disabilities, and the clinical and physiological contexts. In this issue of the Journal, Corrà et al.13 reported the prognostic utility of ‘primary’ CPET variables [i.e. VE/VCO2 slope, peak VO2, and exercise oscillatory ventilation (EOV)]10 in 1280 consecutive patients with a range of different cardiovascular diseases who underwent CPET for prognostic exertional stratification after completing cardiac rehabilitation. Among them, 585 were classified as having asymptomatic left ventricular dysfunction (ALVD) [mean left ventricular ejection fraction (LVEF) 30 ± 7%], and 695 had heart failure with reduced ejection fraction (HFrEF) (mean LVEF 25 ± 3%). Asymptomatic left ventricular dysfunction was defined as LVEF ≤40% in the absence of heart failure symptoms (i.e. ACC/AHA Stage B) and typically without positive physical exam findings. At a mean follow-up of 906 days, 142 patients died. By multivariable Cox regression, the authors found that peak VO2 was the only statistically significant predictor of the primary endpoint of death at 3 years in the ALVD group, whereas in patients with HFrEF, VE/VCO2, peak VO2, and EOV remained informative predictors of adverse outcomes. Given the lack of prognostic value of the ventilatory parameters in patients with ALVD, the authors proposed dropping VE/VCO2 and EOV from the algorithm for prognostic exertional stratification of these patients. Asymptomatic left ventricular dysfunction is an important condition that remains understudied. The prevalence of ALVD (systolic dysfunction subtype) has been estimated at 3–6%, depending on the study.14 A meta-analysis including over 25 000 participants with an average follow-up of 7.9 years found that the relative risks for progression to heart failure was 4.6 (95% confidence interval 2.2–9.8) for the systolic dysfunction subtype, and 1.7 (1.3–2.2) for the diastolic dysfunction subtype.15 The findings of Corrà et al.13 demonstrated yet again the usefulness of specific CPET parameters in refining the risk stratification of patient subsets to inform prognosis for the ALVD systolic dysfunction subtype. The study highlighted differences in the exercise stress response and underlying pathophysiology of ALVD compared with HFrEF, in that the former had predominantly ventricular dysfunction with limited ventilatory (and neurocardiac) impairment, whereas the latter encompassed more extensive systemic physiological deficits beyond the heart and cardiopulmonary circulation. Availability of natriuretic peptide levels for comparative analysis could have provided additional clues in this regard. While the study has advanced our understanding of prognostic exertion stratification in ALVD, external validation beyond a single centre is needed. With a larger sample size and multi-centre contribution, a classification and regression tree (with machine learning) could be developed. Further characterisation of ventricular dysfunction and its associated circulatory disorder based on functional systemic physiology may lead to improved disease classification and management and patient care. This could potentially be explored in the next step using CPET with concurrent echocardiography.10 Conflict of interest: The authors declare they have no conflict of interest. No new data were generated or analysed in support of this research. The opinions expressed in this article are not necessarily those of the Editors of the European Journal of Preventive Cardiology or of the European Society of Cardiology.