OUP user menu

Assessment of heart rate recovery after exercise stress test: implications for cardiac resynchronization therapy

Bela Merkely, Attila Roka
DOI: http://dx.doi.org/10.1093/europace/euq452 458-459 First published online: 5 January 2011

This editorial refers to ‘Cardiac resynchronization therapy improves exercise heart rate recovery in patients with heart failure’ by S. Okutucu et al., on page 526.

The autonomic nervous system plays a major role in the pathogenesis of chronic heart failure, which usually presents with excess sympathetic activity and a concomitant decrease in parasympathetic tone. Markers of neuroendocrine dysregulation, such as heart rate variability, baroreflex sensitivity, and serum levels of natriuretic peptides, can be measured and correlated with clinical outcomes.

Exercise increases the sympathetic tone and decreases parasympathetic activity. Measurement of post-exercise heart rate recovery (HRR) is a non-invasive method to assess parasympathetic function as heart rate deceleration (HRD) during the first minute after peak exercise is mediated primarily by activation of the parasympathetic nervous system.1 Impaired HRR was shown to be associated with an increased mortality in subjects referred for stress testing regardless of cardiovascular disease history.2 Epidemiological studies in stable heart failure patients have shown that the extent of autonomic dysfunction bears a strong correlation with heart failure severity.3 Blunted HRR has been shown to be associated with an increased mortality and, more specifically, an increased relative risk of sudden cardiac death.4 Heart rate recovery may provide independent prognostic information in patients with heart failure, even after accounting for previously identified clinical and exercise-derived prognostic variables.5

Patients eligible for cardiac resynchronization therapy (CRT) represent a heterogeneous heart failure group, and conventional parameters are insensitive to identify responders. Using criteria published in current guidelines [New York Heart Association (NYHA) functional class, QRS duration, and left ventricular ejection fraction], the ratio of non-responding patients after CRT device implantation still remains unacceptably high, up to 30%. Attempts have been made to evaluate the efficacy of non-invasive assessment of the autonomic nervous system to predict the response to CRT. Thomas et al. examined the relationship between HRR following exercise and the subsequent response to CRT in 37 patients. Functional responders demonstrated greater HRR than non-responders at 30, 60 and 90 s post-exercise. The differences in HRR between responders and non-responders were most pronounced at an early time point, 30 s from the cessation of exercise.6 On the other hand, in the study by Okutucu et al., HRR after exercise stress test performed before CRT implantation did not identify responders.7 Of note, the interpretation of HRR was different in the two studies.

Lack of standard definition of HRR makes it difficult to compare the results from different studies. Previous studies examining the role of HRR used different criteria to define abnormal HRR. Cole et al.8 defined an abnormal HRR as a decrease of <13 bpm. Thomas et al. used HRD gradients calculated with a formula that approximated the rate of HRR at 30, 60, 90, and 120 s after cessation of exercise. Their hypothesis was that this would be a more sensitive measure of HRR which would in part account for differences in the chronotropic response to exercise and allow investigating the autonomic influences independent of the absolute heart rate.6 They showed that the HRD correlated significantly more closely with echocardiographic and functional responses following CRT than the absolute HR reduction. The analysis also identified that HRD gradients at early, rather than late time points in recovery were the most sensitive indices to guide subsequent responses to CRT. Heart rate deceleration at 30 s had an 89% sensitivity, 88% specificity, 89% positive, and 78% negative predictive value to predict the functional response to CRT and a 64% sensitivity, 89% specificity, 95% positive, and 44% negative predictive value for the echocardiographic response.6 The differences in predictive values may be due to the partial independence of neurohormonal and cardiac structural reverse remodelling. Structural remodelling correlates with the outcome after CRT.9 The importance and correlation of neuroendocrine changes with outcome in these patients is less clear.

In the study by Thomas et al., there was no significant correlation between baseline echocardiographic and functional parameters and HRD, whereas HRD showed a good correlation with echocardiographic and functional improvements following CRT. Investigation of autonomic dysfunction using HRD analysis and other conventional markers (i.e. electrical and mechanical dyssynchrony) could, therefore, possibly be used to identify responders before CRT implantation. This still needs to be validated in a prospective study. The study by Okutucu et al. found that the degree of left ventricular remodelling, as defined by left ventricular end-systolic volume 6 months after CRT device implantation, showed a good correlation with the degree of HRR in the first minute after exercise stress test.7

Different aspects of neuroendocrine ‘reverse remodelling’ with CRT have been addressed in other studies. The role of autonomic function has been investigated in a subgroup analysis in 50 patients from the MIRACLE study.10 Cardiac resynchronization therapy was associated with an increase in heart rate variability in the absence of a measurable effect on circulating catecholamine, leading the authors to conclude that the beneficial effects of CRT on autonomic function are predominantly mediated through an increase in parasympathetic influences. Other investigators have also demonstrated the ability of CRT to partially reverse autonomic dysfunction through acute modulation of the arterial baroreflex.11

In the study by Okutucu et al., the ratio of patients with ischaemic cardiomyopathy was significantly higher in the non-responder group, than in the responders.7 This may also have affected the differences seen in HRR in the responder vs. non-responder group after CRT implantation as previous data suggest that the amount of myocardial damage correlated with an abnormal HRR. A large cohort study in 1296 patients who underwent cardiac stress test with SPECT imaging found that the abnormal HRR was more often found in older, diabetic, and hypertensive patients, and in those who had previous myocardial infarction and myocardial revascularization, higher heart rate at rest, worse perfusion defect quantification scores with SPECT, lower left ventricular ejection fraction, and larger left ventricular volumes. In multivariable analysis, age, heart rate at rest, left ventricular ejection fraction, and the extent and severity of the perfusion defect at rest were independent predictors of abnormal HRR. There was no correlation with gated SPECT markers of myocardial ischaemia.12

Heart failure patients represent a heterogeneous population whose response to biventricular pacing is mediated through a complex interplay of mechanisms. Evaluation of autonomic function may provide a valuable contribution to the pre-implant assessment of these patients. Early HRR (30 s to 1 min), when parasympathetic influences predominate, may carry the greatest predictive power. The role of autonomic nervous system in functional and structural remodelling after CRT still needs to be elucidated. The clinical utility of neuroendocrine assessment (such as HRR after exercise stress) to identify the potential responders to CRT in addition to current selection criteria will need to be evaluated in a prospective study.

Conflict of interest: none declared.


  • The opinions expressed in this article are not necessarily those of the Editors of Europace or of the European Society of Cardiology.