Robert Wolk; Virend K. Somers; Raymond J. Gibbons; Thomas Olson; Kathy O'Malley; Bruce D. Johnson 

Med Sci Sports Exerc.  2006;38(8):1367-1373.  ?2006 American College of Sports Medicine

Abstract and Introduction

Abstract

Purpose: Heart failure (HF) is associated with blunted HR recovery after exercise. The determinants of altered HR recovery in HF are unknown. The aim of this study was to investigate clinical correlates of HR recovery in HF patients.
Methods: Echocardiography, pulmonary function tests, exercise testing, and neurohormonal measurements were performed in 98 HF patients. HR recovery was calculated as the difference between heart rate at peak exercise and at 1 min into a recovery period. Study subjects were divided into three groups based on HR recovery tertiles: group 1 (HR recovery ≤ 6 bpm), group 2 (7 ≤ HR recovery ≤ 12), and group 3 (HR recovery ≥ 13).
Results: There were significant differences between the groups in multiple parameters. Compared with group 3, patients in group 1 had greater E/A ratios (1.81 ? 0.26 vs 0.98 ? 0.12, P = 0.011), shorter deceleration time (170 ? 11 vs 223 ? 11 ms, P = 0.016), and higher plasma atrial natriuretic peptide levels (207 ? 32 vs 101 ? 12 pg?mL^-1, P = 0.008), indicating higher left ventricular filling pressures and elevated left atrial pressures. Pulmonary function tests were suggestive of greater restrictive changes in the lungs. Finally, subjects in group 1 had impaired exercise capacity, as evidenced by shorter exercise duration (5.2 ? 0.2 vs 8.3 ? 0.4 min, P < 0.001), lower peak VO₂ (14.6 ? 0.6 vs 22.2 ? 1.0 mL?kg^-1?min^-1, P < 0.001), higher VE/VCO₂ ratios (36.4 ? 1.1 vs 31.1 ? 0.9, P = 0.001), and reduced chronotropic responses to exercise (39 ? 3 vs 69 ? 4 bpm, P < 0.001).
Conclusion: HR recovery may be a clinically useful index identifying HF patients with distinct echocardiographic, neurohormonal, and hemodynamic characteristics. This may have implications for our understanding of the pathophysiology of impaired HR recovery in HF as well as for the clinical evaluation of such patients.

Introduction

HR recovery is defined as the decline in HR immediately following exercise. Decreased HR recovery is a predictor of worse cardiovascular prognosis, independent of and in addition to other risk scores.^{[5-7,14,17,24,26]} Although the exact mechanisms of impaired HR recovery are not known. it may indicate a disturbance in automatic balance leading to delayed reactivation of parasympathetic tone.^[3]

Although studies showing the prognostic association of impaired HR recovery included subjects both with and without a history of cardiovascular disease, HR recovery in patients with chronic heart failure (HF) has not been specifically addressed, and in many studies HF was, in fact, one of the exclusion criteria.^{[5-7,14,17,24,26]} However, preliminary reports suggest that HF in and of itself may be associated with blunted HR recovery.^[10,21] The determinants of any impaired HR recovery in patients with HF have not yet been established. The aim of the present study was to investigate pathophysiological characteristics of HR recovery in a well-characterized cohort of patients with stable HF.

Methods

Patient Recruitment

Consecutive HF patients (N = 98) with a history of ischemic or nonischemic cardiomyopathy were recruited from the heart failure clinic and from the cardiovascular health clinic (a preventive and rehabilitative center) at the Mayo Clinic, Rochester, MN. Inclusion criteria included stable HF and echocardiographic ejection fraction ≥ 40%. Patients with primary lung disease, morbid obesity (body mass index (BMI) > 40 kg?m^-2), musculoskeletal diseases, valvular heart disease, peripheral vascular disease, chest pain, pacemaker dependency, atrial fibrillation, or history of complex ventricular arrhythmias, as well as current smokers or those with a heavy smoking history (> 10 packs per year), or patients who were otherwise unable to perform exercise testing, were excluded. All subjects underwent anthropometric measurements, including assessment of body fat by the three-site skinfold caliper method.^[20] In addition, based on evidence that exercise training and physical activity can affect HR recovery,^[25] each participant's exercise habit (as a categorical variable and as the amount of routine exercise in terms of minutes of exercise per week) was also obtained by questionnaire.

The study was approved by the Mayo institutional review board. Written informed consent was obtained from each patient prior to participation.

Cardiac and Pulmonary Function Measurements

Left ventricular function (ejection fraction, diastolic and systolic dimensions) was assessed while in the resting supine position using two-dimensional echocardiography according to recommendations of the American Society of Echocardiography.^[23] Ventricular filling pattern was evaluated at rest by pulsed wave Doppler flow of the mitral valve.^[18] The variables obtained from the inflow pattern of the mitral valve included peak velocity of early diastolic filling (E wave), peak velocity of late diastolic filling (A wave), E/A ratio, and deceleration time (DT).

Pulmonary function measurements were performed at rest prior to exercise testing and included an assessment of lung volumes (slow and forced vital capacities, VC and FVC, respectively; forced expiratory volume in 1 s, FEV₁; alveolar volume, V_A) and an assessment of expiratory flows (maximal forced expiratory flow, FEF_max; the mean forced expiratory flow between 25 and 75% of FVC, FEF_25-75). A single breath diffusion capacity of the lung for carbon monoxide (DLCO) was also measured. Spirometry and DLCO data were collected in accordance with American Thoracic Society standards.^[1]

Exercise Testing and HR Recovery

All patients performed the same incremental treadmill exercise test according to a standard protocol. Gas exchange, HR, and oxygen saturation were measured during a graded treadmill testing to volitional fatigue. Oxygen consumption, carbon dioxide production, and ventilation parameters were measured using a metabolic measurement system (MedGraphics CPX/D/Medical Graphics, St. Paul, MN). Participants were fitted with a nose clip and a standard mouthpiece attached to a PreVent? pneumotach (Medical Graphics, St. Paul, MN) and worn throughout the testing procedure, including the recovery period. Manual flow calibrations were performed with a 3-L syringe, and gas calibrations were performed with manufacturer-recommended gases of known concentration. All calibration procedures were accomplished immediately prior to each testing protocol. The protocol consisted of an initial treadmill speed and grade of 2.0 mph and 0%, respectively. The speed and/or grade were subsequently adjusted every 2 min to yield an appromimately 2-MET increase per work level. Participants were encouraged verbally to continue the exercise protocol to maximal exertion identified as a RPE > 18 on the Borg 6-20 scale and an RER of > 1.15. After achieving peak workload, the subjects were brought immediately to 1.7 mph and a 0% grade for a cool-down. They maintained this for 3 min, followed by 3 min of seated rest. The HR recovery was defined as the difference between HR at peak exercise and at 1 min into a recovery cool-down period.^[5,7,17,26]

Neurohormonal Measurements

All blood draws for biochemical measurements were taken in the morning after 30 min of quiet rest in the supine position. Plasma catecholamines and angiotensin-converting enzyme activity were measured using high-performance liquid chromatography and a spectrophotometric method, respectively. Angiotensin II, brain natriuretic peptide, and atrial natriuretic peptide (ANP) were assessed according to previously published standard methods.^[4,27]

Statistical Analysis

Statistical analysis was performed using STATISTICA software (Stat Soft Inc., Tulsa, OK). A chi-square test was used for comparison of categorical parameters. For between-group comparisons, ANOVA and ANCOVA were used, followed by Neuman-Keuls post hoc tests. Age, gender, BMI, smoker status, packs smoked per year, exercise habit, the amount of routine exercise, ischemic chronic HF etiology, and medications were used as covariates. A P value of ≤ 0.05 was considered statistically significant. All data are expressed as mean ? SEM.

Results

The HR recovery was calculated for all study subjects and ranged from -5 bpm (indicating an increase in HR during recovery after exercise) to +51 bpm (consistent with a substantial decrease in HR).

The study cohort was then divided into three groups based on HR recovery tertiles. There were 34 subjects with HR recovery ≤ 6 bpm (group 1), 34 subjects with HR recovery ≥ 7 and ≤ 12 bpm (group 2), and 30 subjects with HR recovery ≥ 13 bpm (group 3). The clinical characteristics of the three groups are shown in Table 1 . The subjects with lower HR recovery (group 1) had slightly greater BMI, but percentage of body fat was similar in all groups. Cardiovascular medications were similar in all groups, but diuretics were used more often in subjects with lower HR recovery. Lastly, more than 50% of patients with the lowest HR recovery had HF of ischemic etiology, in contrast to patients with the highest HR recovery, in whom the prevalence of ischemic HF was only 7% (Fig. 1).

HF patients with the lowest HR recovery had echocardiographic features of greater left ventricular and left atrial dysfunction ( Table 2 ). Specifically, the E/A ratio was greater and deceleration time was shorter in group 1, indicating higher left ventricular filling pressures and elevated left atrial pressures in subjects with lower HR recovery (Fig. 2). The presence of left ventricular and atrial pressure/volume overload was further confirmed by higher plasma ANP levels in group 1 ( Table 2 ) (Fig. 3).

Pulmonary function tests ( Table 3 ) were suggestive of greater restrictive changes in group 1 as compared with groups 2 and 3.

Subjects with the lowest HR recovery also had impaired exercise capacity ( Table 4 ) (Fig. 4). In spite of similar RPE and RER in all groups, exercise duration and peak VO₂ were both lower in group 1, and the VE/VCO₂ ratio was higher, indicating impaired ventilatory efficiency. Also, subjects with lower HR recovery had reduced chronotropic responses to exercise ( Table 4 ).

Discussion

In the present study, we investigated echocardiographic, neurohormonal, and hemodynamic correlates of HR recovery in clinically stable patients with HF. HR recovery was defined as maximum HR during exercise minus HR after exercise, which measures the actual decrease in HR relative to peak exercise (ΔHR). Although several other indices of HR recovery can also potentially be used (such as a relative decrease in HR, expressed as a percentage of peak HR or as a percentage of the increase from baseline to peak exercise), the relative decrease in HR is highly correlated with the actual decrease (ΔHR), and ΔHR seems to be the most informative with regard to outcome.^[14] In fact, our findings using the relative decrease in HR (a percentage of peak HR) were similar to those using the actual decrease (data not shown). In the present study we measured HR recovery as the difference between HR at peak exercise and HR at 1 min after exercise-the index that has been consistently shown to be an independent prognostic factor.^{[5,7,9,14,17,24,26]}

In previous studies on HR recovery in non-HF patients, the optimal prognostic cutoff Values for HR recovery (defining normal vs abnormal HR recovery) were based on specific statistical considerations or were arbitrarily selected. The most commonly chosen cutoff value was 12 bpm.^[9] If one considers the range of HR recovery values found in the present study (from -5 to +51 bpm (Fig. 1)), it appears that a substantial number of HF patients have HR recovery that would be considered normal by those arbitrary standards. The broad range of HR recovery values in our HF population raises an important question, namely, whether those HF patients with attenuated HR recovery represent a pathophysiologically distinct clinical cohort, different from those HF subjects with preserved HR recovery.

As mentioned earlier, the previously used HR recovery cutoff value of 12 bpm was selected for non-HF populations.^[9] The best clinically useful cutoff value for HR recovery in patients with HF has not been established. Therefore, in our study, we divided the entire study group based on HR recovery tertiles for that population. The main finding of our study is that those with the lowest HR recovery (≤ 6 bpm) represent a HF cohort with distinct echocardiographic, neurohormonal, and hemodynamic characteristics. The echocardiographic measurements are suggestive of greater impairment of left ventricular function, with greater left ventricular and left atrial pressure/volume overload, in HF patients with blunted HR recovery and are consistent with the presence of higher ANP levels in these patients. Natriuretic peptides are released from the atria and ventricles of the heart in response to wall stretch,^[15] and thus their elevated levels indicate atrial and ventricular pressure overload and dysfunction. The results of pulmonary function tests showed features of restrictive changes in the lungs. Although the cause for the latter cannot be directly inferred from the present study, it may be due to a greater degree of pulmonary congestion. Finally, the HF subjects with lower HR recovery displayed decreased exercise capacity, as evidenced by shorter exercise duration, lower peak oxygen consumption, lower ventilatory efficiency, and attenuated chronotropic responses to exercise. The finding of a graded association between HR recovery and various echocardiographic, neurohormonal, and hemodynamic measurements (such that there is a gradual change in a given measurement when going from the lowest to the highest tertile (Figs. 2-4) suggests a functional and pathophysiological link between these measurements and HR recovery.

These observations have several important pathophysiological and clinical implications. First, stretch of atrial receptors and alterations in venous return have been implicated in the regulation of chronotropic responses in the heart.^[11,19,22] Although we did not measure cardiac pressures at peak exercise or during recovery, elevated rest indices of cardiac pressure/volume overload in HF patients with lower HR recovery suggest that increased atrial stretch may underlie the blunted deceleration of HR after exercise in these patients. Secondly, our data indicate that in the clinical setting, impaired HR recovery may be an easily measured clinical index of greater functional and hemodynamic compromise requiring more aggressive therapy. In fact, in the present study, use of diuretics was significantly greater in those HF patients with lower HR recovery.

Patients with lower HR recovery were also characterized by a reduced chronotropic reserve during exercise, in that both the absolute and relative increases in HR were significantly lower compared with those HF patients with higher HR recovery. It is therefore unclear whether, from the pathophysiological perspective, reduced HR recovery is only a function of reduced chronotropic responses to exercise or whether it has a different pathophysiological mechanism. However, although chronotropic incompetence is predictive of cardiovascular outcome,^[12] the predictive value of HR recovery has been shown to be independent of changes in HR during exercise.^[5]

The association between HR recovery and peak VO₂ (Fig. 4) is also of interest and may indicate a pathophysiological link, although the exact significance of this finding remains to be elucidated. Because peak VO₂ has been consistently shown to be predictive of cardiovascular outcome in HF,^[16] this association suggests that HR recovery may be prognostically important in HF patients. Further studies are needed to confirm this possibility and to better define the overlap in the prognostic information provided by HR recovery and peak VO₂.

Finally, in the present study the patient group with the lowest HR recovery had the highest prevalence of HF of ischemic etiology (group 1; Table 1 and Fig. 1). Similarly, HF patients with ischemic cardiomyopathy had lower HR recovery (4 ? 1 bpm) than those with nonischemic cardiomyopathy (12 ? 1 bpm; P < 0.001). Differences in autonomic responsiveness between ischemic and nonischemic cardiomyopathy^[2,13] may potentially explain blunted HR recovery in group 1, and at the same time, ischemic cardiomyopathy may also contribute to more impaired left ventricular diastolic function and higher atrial pressures in this group. However, we observed statistically significant differences in echocardiographic, neurohormonal, and hemodynamic characteristics between HR recovery tertiles after adjusting for HF etiology, suggesting that these structural and functional associations of HR recovery are, at least in part, independent of HF etiology. Nevertheless, because ischemic HF may have a worse long-term prognosis than nonischemic HF,^[8] HF etiology should be considered in any future studies investigating the prognostic impact of HR recovery in HF.

Study Limitations and Future Directions

A potential limitation of this study is the relatively small sample size available for analysis. Furthermore, the study group consisted of carefully selected patients, recruited based on specific inclusion and exclusion criteria. As such, our findings need to be confirmed in a larger, community-based, nonselective HF cohort.

There are a number of potential areas in which the investigation of HR recovery in the HF population could be expanded in future studies. For example, the only available cutoff values for HR recovery are based on the prevalent literature in the CAD population without HF. Thus, there is a need to define the normal versus abnormal HR recovery responses specifically in the HF population. Also, the value of HR recovery as an independent prognostic tool in HF has yet to be determined. Further, little is known about the mechanisms underlying potential differences in HR recovery between ischemic and nonischemic HF etiologies, as suggested by our results. Finally, as the importance of HR recovery in HF is further confirmed in future studies, understanding how various treatment protocols (such as pharmaceutical regimens or exercise training/cardiac rehabilitation) affect HR recovery will be of potential clinical significance.

Conclusion

HR recovery may be a clinically useful index for identifying HF patients with distinct echocardiographic, neurohormonal, and hemodynamic characteristics. This may have implications for our understanding of the pathophysiology of impaired HR recovery in HF as well as for the clinical evaluation of such patients.

American Thoracic Society. Standardization of spirometry, 1994 update. Am. J. Respir. Crit. Care Med. 152:1107-1136, 1995. Anthonio, R. L., J. Brouwer, P. Lechat et al. Different effects of bisoprolol on heart rate in patients with ischemic or idiopathic dilated cardiomyopathy (a 24-hour Holter substudy of the Cardiac Insufficiency Bisoprolol Study [CIBIS]). Am. J. Cardiol. 83:1286-1289, 1999. Arai, Y., J. P. Saul, P. Albrecht, et al. Modulation of cardiac autonomic activity during and immediately after exercise. Am. J. Physiol. 256:H132-H141, 1989. Burnett, J. C. Jr., P. C. Kao, D. C. Hu, et al. Atrial natriuretic peptide elevation in congestive heart failure in the human. Science 231:1145-1147, 1986. Cole, C. R., E. H. Blackstone, F. J. Pashkow, C. E. Snader, and M. S. Lauer. Heart-rate recovery immediately after exercise as a predictor of mortality. N. Engl. J. Med. 341:1351-1357, 1999. Cole, C. R., J. M. Foody, E. H. Blackstone, and M. S. Lauer. Heart rate recovery after submaximal exercise testing as a predictor of mortality in a cardiovascularly healthy cohort. Ann. Intern. Med. 132:552-555, 2000. Diaz, L. A., R. C. Brunken, E. H. Blackstone, C. E. Snader, and M. S. Lauer. Independent contribution of myocardial perfusion defects to exercise capacity and heart rate recovery for prediction of all-cause mortality in patients with known or suspected coronary heart disease. J. Am. Coll. Cardiol. 37:1558-1564, 2001. Follath, F., J. G. Cleland, W. Klein, and R. Murphy. Etiology and response to drug treatment in heart failure. J. Am. Coll. Cardiol. 32:1167-1172, 1998. Gibbons, R. J. Abnormal heart-rate recovery after exercise. Lancet 359:1536-1537, 2002. Imai, K., H. Sato, M. Hori, et al. Vagally mediated heart rate recovery after exercise is accelerated in athletes but blunted in patients with chronic heart failure. J. Am. Coll. Cardiol. 24:1529-1535, 1994. Keatinge, W. R. The effect of increased filling pressure on rhythmicity and atrioventricular conduction in isolated hearts. J. Physiol. 149:193-208, 1959. Lauer, M. S., G. S. Francis, P. M. Okin, F. J. Pashkow, C. E. Snader, and T. H. Marwick. Impaired chronotropic response to exercise stress testing as a predictor of mortality. JAMA. 281:524-529, 1999. Malfatto, G., G. Branzi, S. Gritti, et al. Different baseline sympathovagal balance and cardiac autonomic responsiveness in ischemic and non-ischemic congestive heart failure. Eur. J. Heart Fail. 3:197-202, 2001. Morshedi-Meibodi, A., M. G. Larson, D. Levy, C. J. O'Donnell, and R. S. Vasan. Heart rate recovery after treadmill exercise testing and risk of cardiovascular disease events (The Framingham Heart Study). Am. J. Cardiol. 90:848-852, 2002. Munagala, V. K., J. C. Burnett, and M. M. Redfield. The natriuretic peptides in cardiovascular medicine. Curr. Probl. Cardiol. 29:707-769, 2004. Myers, J., and L. Gullestad. The role of exercise testing and gas-exchange measurement in the prognostic assessment of patients with heart failure. Curr. Opin. Cardiol. 13:145-155, 1998. Nishime, E. O., C. R. Cole, E. H. Blackstone, F. J. Pashkow, and M. S. Lauer. Heart rate recovery and treadmill exercise score as predictors of mortality in patients referred for exercise ECG. JAMA. 284:1392-1398, 2000. Oh, J. K., C. P. Appleton, L. K. Hatle, R. A. Nishimura, J. B. Seward, and A. J. Tajik. The noninvasive assessment of left ventricular diastolic function with two-dimensional and Doppler echocardiography. J. Am. Soc. Echocardiogr. 10:246-270, 1997. Pathak, C. L. Autoregulation of chronotropic response of the heart through pacemaker stretch. Cardiology 58:45-64, 1973. Pollock, M. L., D. H. Schmidt, and A. S. Jackson. Measurement of cardiorespiratory fitness and body composition in the clinical setting. Compr. Ther. 6:12-27, 1980. Racine, N., M. Blanchet, A. Ducharme, et al. Decreased heart rate recovery after exercise in patients with congestive heart failure: effect of beta-blocker therapy. J. Card. Fail. 9:296-302, 2003. Savin, W. M., D. M. Davidson, and W. L. Haskell. Autonomic contribution to heart rate recovery from exercise in humans. J. Appl. Physiol. 53:1572-1575, 1982. Schiller, N. B., P. M. Shah, M. Crawford, et al. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. J. Am. Soc. Echocardiogr. 2:358-367, 1989. Shetler, K., R. Marcus, V. F. Froelicher, et al. Heart rate recovery: validation and methodologic issues. J. Am. Coll. Cardiol. 38:1980-1987, 2001. Tiukinhoy, S., N. Beohar, and M. Hsie. Improvement in heart rate recovery after cardiac rehabilitation. J. Cardiopulm. Rehabil. 23:84-87, 2003. Watanabe, J., M. Thamilarasan, E. H. Blackstone, J. D. Thomas, and M. S. Lauer. Heart rate recovery immediately after treadmill exercise and left ventricular systolic dysfunction as predictors of mortality: the case of stress echocardiography. Circulation 104:911-916, 2001. Yasue, H., M. Yoshimura, M. Jougasaki, et al. Plasma levels of BNP in normal subjects and patients with chronic heart failure: measurement by immunoradiometric assay (IRMA). Horm. Clin. 42:397-403, 1993.

Acknowledgements

The authors thank Renee Blumers for help with the manuscript submission and Angela J. Heydmann for technical assistance. Funding sources: NIH grants HL71478 and HL07111.

Reprint Address

Bruce D. Johnson, Ph.D., Mayo Clinic, 200 First Street NW, Gonda 5-369, Rochester, MN 55905; E-mail: johnson.bruce@mayo.edu .