The effect of altering heart rate on ventricular function in patients with heart failure treated with h-blockers Simon D.R. Thackray, MRCP, a Justin M. Ghosh, MRCP, Graham A. Wright, PhD, b Klaus K.A. Witte, MRCP, Nikolay P. Nikitin, PhD, Gerald C. Kaye, MD, FRCP, Andrew L. Clark, MA, MD, FRCP, Ann Tweddel, MD, FRCP, and John G.F. Cleland, MD, FRCP, FESC, FACC Kingston-upon-Hull, United Kingdom Background h-blockers are effective for the treatment of heart failure, but their mechanism of action is unresolved. Heart rate reduction may be a central mechanism or a troublesome side effect. Methods A randomized, double-blind, parallel group study comparing chronic higher-rate (80 pulses per minute) with lower-rate (60 pulses per minute) pacing in pacemaker-dependent patients with symptomatic left ventricular (LV) systolic dysfunction, receiving h-blockers. Gated radionuclide ventriculography (RNVG) was performed at baseline and after at least 9 months. The primary outcome was change in LV volumes, as a marker of beneficial reverse remodeling, from baseline to follow-up. Results Forty-nine patients were randomized. Mean age was 74 F 6 years and with LV ejection fraction of 26% F 9% at baseline. During 14 F 13 months of follow-up, 21 patients (43%) died and 25 (51%) completed the study protocol: 12 in the higher-rate and 13 in the lower-rate group. Mean LV end-diastolic (higher rate +20 F 104 ml vs lower rate 65 F 92 ml, P =.03) and systolic (higher rate +29 F 83 ml vs lower rate 60 F 74 ml, P =.006) volumes increased with higher-rate versus lower-rate pacing, whereas LV ejection fraction declined (higher rate 4.2% F 4.4% vs lower rate +2.2% F 5.4%, P =.002). Conclusion Reversal of h-blocker induced bradycardia has deleterious effects on ventricular function, suggesting heart rate reduction is an important mediator of their effects. The prognosis of patients with pacemakers and heart failure is poor. (Am Heart J 2006;152:713.e92713.e13.) h-blockers reduce morbidity and mortality in patients with heart failure, 1 but the mechanisms mediating this effect are uncertain. It is likely that improvement in left ventricular (LV) function is an important factor. 2,3 However, the mechanism by which h-blockers improve LV function is also uncertain. Possible mechanisms include blockade of adrenergic receptors, leading to improved cardiac myocyte metabolism and a reduction in apoptosis, 4,5 a reduction in heart rate, 6,7 a reduction in atrial and ventricular arrhythmias, 8 or a reduction in myocardial ischemia and/or hibernation. 9,10 The data supporting the hypothesis that h-blockers exert their effects merely by slowing heart rate are From the a Department of Academic Cardiology, University of Hull, Castle Hill Hospital, Kingston-upon-Hull, United Kingdom, and b Department of Nuclear Medicine, Hull Royal Infirmary, Kingston-upon-Hull, United Kingdom. No conflicts of interest exist for any of the authors. This work is entirely supported by a grant from the British Heart Foundation. Submitted October 15, 2005; accepted July 13, 2006. Reprint requests: Simon D.R. Thackray, MRCP, Department of Academic Cardiology, University of Hull, Castle Hill Hospital, Cottingham, East Yorkshire, United Kingdom. E-mail: simonthackray@hotmail.com 0002-8703/$ - see front matter n 2006, Published by Mosby, Inc. doi:10.1016/j.ahj.2006.07.007 inconclusive, and observational data are conflicting. 11-13 Even if the 2 phenomena are correlated, this does not prove cause and effect. Bradycardia could be a critically necessary mediator of the benefits of h-blockers or may just be an unwanted side effect. Prevention of bradycardia could enhance the benefits of h-blockade, for instance, by improving symptoms like breathlessness and fatigue or by allowing institution of therapy even in patients with low heart rates. Moreover, in patients with heart failure who have had a pacemaker implanted for conducting system disease, it is not known whether prevention of bradycardia would attenuate the benefits of h-blockade. Finally, if heart rate reduction is central to the effect of h-blockade, then nonadrenergic ratemodifying agents such as ivabradine 14 may become useful agents in treatment of chronic heart failure. Methods Setting The study was done in a regional tertiary referral cardiothoracic center serving a mixed urban/rural population of 650 000 for the purposes of pacemaker implantation. Written informed consent was obtained from all patients. The local research ethics committee approved the study.
713.e10 Thackray et al. American Heart Journal October 2006 Table I. Baseline characteristics and response to different paced rates High rate (80 ppm 1 ) (n = 12) Low rate (60 ppm 1 ) (n = 13) P Age (y) 70 F 7 75 F 6 P =NS Carvedilol dose (mg) 25 F 7 26 F 7 P =NS Ischemic heart disease 10 (83%) 11 (85%) P = NS Months paced at entry 30 F 22 35 F 34 P =NS Paced rate at entry 62 F 4.5 61 F 2.7 P =NS Paced ventricular beats during study (% of total) 95% F 4% 92% F 7% P =NS Mean heart rate during study period 83 F 2 beat/min 1 64 F 3 beat/min 1 P b.001 Follow-up (m) 13.6 F 2.9 13.9 F 2.7 P =NS DDD/VVI (R) (n) 6/6 7/6 P =NS SR/AF (n) 7/5 8/5 P =NS Baseline LVEF (%) 26.2 F 8.9 21.8 F 8.9 P =NS Change in LVEF (%) 4.5 F 4.4 +2.2 F 5.4 P =.002 Baseline LVEDV 317 F 127 ml 349 F 123 ml P =NS Change in LVEDV +8 F 104 ml 65 F 92 ml P =.03 Baseline LVESV 235 F 119 ml 285 F 117 ml P =NS Change in LVESV +23 F 83 ml 61 F 74 ml P =.006 Improved LV function4 1 of 12 (8%) 7 of 13 (53%) P b.001 Data are presented as mean F 1 SD. 4Defined as reduced LVEDVs (ml) and improved LV ejection fraction from baseline to follow-up RNVG. Patient selection Patients with symptoms of heart failure and LV systolic dysfunction (LV ejection fraction on biplane echocardiography of b40%) were identified by screening consecutive patients attending a routine pacemaker follow-up clinic. 15 Other inclusion criteria were permanent pacing for N3 months, age N18 years, and symptoms compatible with heart failure present for N3 months. Exclusion criteria were a recent (b1 month) myocardial infarction or cardiac surgery, biventricular pacing devices, contraindications to h-blocker therapy, or inability to give informed consent. Interventions Patients were randomized either to lower-rate ventricular pacing, with pacing rate set at 60 ppm and any rate response algorithm set to low, or higher-rate ventricular pacing, with pacing rate set at 80 ppm with any rate response algorithm set to medium or high. Patients were randomized to 1 of 2 groups on a 1:1 basis using random number blocks. Patients with single-chamber VVI pacemakers and dual-chamber DDD pacemakers were randomized separately to achieve an equal ratio in each group. Patient assessment In addition to a medical history, New York Heart Association classification, and physical examination, patients had an electrocardiogram, blood tests for renal function and hemoglobin, and an echocardiogram. Patients were assessed clinically at 3-month intervals for at least 9 months. Duration of pacing, current pacing mode, indication for pacing, and rhythm at implant were recorded. Number of paced beats as a percentage of total was derived from the internal pacemaker histograms or from 24-hour Holter monitoring if not available from pacemaker telemetry. Chronic atrial fibrillation (AF) was defined, for the purposes of this study, as the presence of AF, diagnosed on serial 12-lead electrocardiograms. Patients not already treated with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, h-blockers, and spironolactone had treatment initiated and titrated to optimal doses. Carvedilol was the standard h-blocker used in our clinical practice. Once titration was complete, doses of medication, other than loop diuretics, had to be unchanged for 3 months before randomization. The methodology for radionuclide ventriculography (RNVG) used locally has been published in detail previously. 16 For each patient, the following was measured: LV end-diastolic volume in ml (LVEDV), LV end-systolic volume (LVESV [ml]), and LV ejection fraction (LVEF [%]). Baseline scans were conducted only once patients had been stabilized on maintenance doses of medication, including carvedilol, for at least 3 months. A second scan was performed at the end of the study. Statistical analysis Analyses were done using commercially available software (SPSS; SPSS, Chicago, IL). Data are reported as mean F 1 SD. Continuous variables between groups were compared by the Student t test for unpaired observations; multiple groups were compared using analysis of variance with Tukey analysis, where appropriate. Nominal data were compared by the m 2 with A and Cramer s V analysis. Power calculations were based on the within-patient difference between the baseline scan and the follow-up; 22 patients per group gave 90% power to detect a change in LVEDV of 30 ml. The excellent repeatability and reproducibility of the RNVG locally has previously been reported. 16 Results Screening of 307 patients attending the pacing clinic identified 87 patients who fulfilled the entry criteria, of whom 62 agreed to participate. All patients were stabilized on optimal doses of angiotensin-converting
American Heart Journal Volume 152, Number 4 Thackray et al. 713.e11 Figure 1 Graph showing the change in LV ejection fraction (%) and change in LVEDVs by gated RNVG for both pacing rate groups (high rate and low rate) from baseline to follow-up assessment ( represents the mean). enzyme inhibitor or angiotensin receptor blocker and spironolactone. Seven patients (11%) were intolerant of carvedilol and 6 (9%) patients withdrew during the 3-month up-titration period. Forty-nine patients were eventually randomized. Baseline characteristics were similar between groups (Table I). Mean age was 73 F 8 years; most were in New York Heart Association functional class III/IV, with ischemic heart disease, marked LV dilatation, and a depressed LV ejection fraction. The maintenance dose of carvedilol was 26 F 10 mg. Both groups were predominantly ventricular pacing dependent (high rate, 95% paced beats vs 92% for low rate, P = nonsignificant [NS]), leading to a mean 19-beat/min 1 difference in heart rate between the groups over the mean of 13.8 F 2.8 months of follow-up. Annualized mortality was 35%, and only 28 (58%) patients survived to the second assessment. Ten patients
713.e12 Thackray et al. American Heart Journal October 2006 died in each group; there was no excess mortality in patients with VVI pacemakers vs DDD pacemakers. The cause of death was worsening heart failure in 7 patients; sudden or unwitnessed death in 10; and stroke, cancer, and renal failure (1 each). No patient required or requested change in paced rate from that assigned. Three patients randomized to high-rate pacing had a technically inadequate baseline or follow-up scan. Of the patients completing the study (n = 25), in the high-rate group, 6 patients had a VVI pacemaker and 6, a DDD pacemaker; in the low-rate group, 6 had a VVI, and 7, a DDD pacemaker. Compared with higher-rate pacing, pacing at a lower rate was associated with reductions in mean end-systolic and end-diastolic volumes ( P =.006 and P =.03, respectively, for the difference between groups) and maintenance of LV ejection fraction ( P =.002 for the difference between groups). Of the 13 patients, 9 (69%) randomized to lower-rate pacing but only 1 (8%) of 12 randomized to higher-rate pacing showed an increase in ejection fraction and a reduction in LVEDV (m 2 = 57, P b.01) (Figure 1). Multivariable analysis showed no association between the improvement in LV function during long-term h-blockade and age, duration of pacing at entry, baseline LV function, type of pacemaker, or underlying rhythm. Discussion This study suggests that the benefits of h-blockade on LV volume and systolic function are attenuated or reversed by increasing the heart rate to a modest extent. This effect was independent of any other clinical or pacing parameter and sustained over a mean of N13 months of treatment. Clearly, heart rate is an important mediator of h-blocker effect, although this study does not preclude the possibility of additional benefit mediated via adrenergic receptor blockade. The small heart rate difference observed in the COMET study is unlikely to account for the substantial difference in mortality observed in that study. 6 The study is not of adequate design to test whether the observed changes in structure and function translate into a change in symptoms or prognosis. However, the study does confirm the dismal prognosis of patients with heart failure and conventional right ventricular pacing. It is possible that biventricular pacemakers should be used in such patients, and relevant trials are underway. 17 The DAVID trial 18 reported an increase in morbidity and mortality in patients with DDD back-up pacing. This effect is likely to have been mediated by an increase in desynchronising right ventricular pacing rather than an increase in average heart rate. The study is of considerable practical importance. Many patients with conventional pacemakers have or will develop heart failure, and biventricular pacing will be used increasingly. In both settings, the physician has the capacity to control ventricular rate and prevent bradycardia. This study provides evidence that the resting ventricular pacing rate should not be set high. Indeed, the smaller benefit observed in MUSTIC AF, 19 where patients had atrioventricular node ablation and backup pacing at 70 beat/min, compared with other trials of cardiac resynchronisation, lends further support to the concept that, when it comes to pacing, rate as well as synchrony matters in patients with LV systolic dysfunction. Limitations This is a small study with a high mortality rate during the study, reflecting the high mortality rate seen in patients with heart failure and pacemakers. Selecting patients with less profound degrees of cardiac impairment may have lead to a higher survival and, hence, follow-up rate for this study. Most patients became pacemaker-dependent during the study, either under the influence of h-blockade or a higher baseline pacing rate; this maximizes the opportunity to cause ventricular dysynchrony. We feel that both heart rate and ventricular dysynchrony are important issues in this patient group and the larger heart failure population. A larger study will be required to dissect out the relative contribution of these to the prognosis of these patients. A period of 9 months was allowed between baseline and follow-up scans to allow for reverse remodeling occurring under the influence of h-blockade. It was unexpected that during this period, such a large mortality rate would occur this represents a major flaw of the study. We thank Dr Michael Cooklin, Dr John Caplin, Dr Mike Norell, the regional cardiac physiologists and, in particular, the nuclear medicine radiographers for their support. References 1. Cleland JGF, Bristow M, Erdmann E, et al. Beta-blocking agents in heart failure: should they be used and how? Eur Heart J 1996;17:1629-39. 2. Doughty RN, Whalley GA, Gamble G, et al. On behalf of the Australia and New-Zealand heart failure research collaborative group: left ventricular remodelling with carvedilol in patients with congestive heart failure due to ischaemic heart disease. JACC 1997;29:1060-6. 3. Dubach P, Myers J, Bonetti P, et al. Effects of bisoprolol fumarate on left ventricular size, function, and exercise capacity in patients with heart failure: analysis with magnetic resonance myocardial tagging. Am Heart J 2002;143:676-83. 4. Bristow MR, Ginsburg R, Minobe W, et al. Decreased catecholamine sensitivity and beta-adrenergic receptor density in failing human hearts. N Engl J Med 1982;307:205-11.
American Heart Journal Volume 152, Number 4 Thackray et al. 713.e13 5. Lowes BD, Gilbert EM, Abraham WT, et al. Myocardial gene expression in dilated cardiomyopathy treated with beta-blocking agents. N Engl J Med 2002;346:1357-65. 6. Poole-Wilson PA, Swedberg K, Cleland JG, et al. Comparison of carvedilol and metoprolol on clinical outcomes in patients with chronic heart failure in the Carvedilol Or Metoprolol European Trial (COMET): randomised controlled trial. Lancet 2003;362: 7-13. 7. Nagatsu M, Spinale FG, Koide M, et al. Bradycardia and the role of h-blockade in the amelioration of left ventricular dysfunction. Circulation 2000;101:653-9. 8. Cice G, Tagliamonte E, Ferrara L, et al. Efficacy of carvedilol on complex ventricular arrhythmias in dilated cardiomyopathy: doubleblind, randomized, placebo-controlled study. Eur Heart J 2000; 21:1259-64. 9. Cleland JG, Pennell DJ, Ray SG, et al. Carvedilol hibernating reversible ischaemia trial: marker of success investigators. Myocardial viability as a determinant of the ejection fraction response to carvedilol in patients with heart failure (CHRISTMAS trial): randomised controlled trial. Lancet 2003;362:14-21. 10. Jourdain P, Funck F, Fulla Y, et al. Myocardial contractile reserve under low doses of dobutamine and improvement of left ventricular ejection fraction with treatment by carvedilol. Eur J Heart Fail 2002;4:269-76. 11. Waagstein F, Swedberg K, Hjalmarson A, et al. Improvement after metoprolol in idiopathic dilated cardiomyopathy is predicted by baseline systolic blood pressure and change in heart rate. JACC 1996;27 (Suppl A):170A. 12. Sackner-Bernstein JD, Lukas MA, Wu B. Evidence against heart rate reduction as the primary mechanism of carvedilol in chronic heart failure. JACC 1998;31:32a. 13. Lechat P, Hulot JS, Escolano S, et al. Heart rate and cardiac rhythm relationships with bisoprolol benefit in chronic heart failure in CIBIS II Trial. Circulation 2001;103:1428-33. 14. Camm AJ, Lau CP. Electrophysiological effects of a single intravenous administration of ivabradine (s 16257) in adult patients with normal baseline electrophysiology. Drugs R D 2003;4:83-9. 15. Thackray SD, Witte KK, Nikitin NP, et al. The prevalence of heart failure and asymptomatic left ventricular systolic dysfunction in a typical regional pacemaker population. Eur Heart J 2003;24:1143-52. 16. Wright GA, Thackray S, Howey S, et al. Left ventricular ejection fraction and volumes from gated blood-pool SPECT: comparison with planar gated blood-pool imaging and assessment of repeatability in patients with heart failure. J Nucl Med 2003;44:494-8. 17. Cleland JG, Ghosh J, Khan NK, et al. Multi-chamber pacing: a perfect solution for cardiac mechanical dyssynchrony? Eur Heart J 2003;24:384-90. 18. Wilkoff BL, Cook JR, Epstein AE, et al. Dual Chamber and VVI Implantable Defibrillator Trial Investigators Dual-chamber pacing or ventricular backup pacing in patients with an implantable defibrillator: the Dual Chamber and VVI Implantable Defibrillator (DAVID) Trial. JAMA 2002;288:3115-23. 19. Linde C, Leclercq C, Rex S, et al. Long-term benefits of biventricular pacing in congestive heart failure: results from the MUltisite STimulation in cardiomyopathy (MUSTIC) study. J Am Coll Cardiol 2002;40:111-8.