Combining Resynchronization and Defibrillation Therapies for Heart Failure

Despite major advances in pharmacological therapy during the past 2 decades,^{1 - 2} heart failure is associated with more than 280 000 deaths annually in the United States.³ The most common modes of death among patients with symptomatic heart failure are sudden cardiac death and death from progressive heart failure.² Two classes of implantable cardiac devices, implantable cardioverter defibrillators (ICDs) and cardiac resynchronization devices, separately address these modes of death in heart failure. Implantable cardioverter defibrillators can terminate potentially lethal ventricular arrhythmias by pacing or shocking the heart. In high-risk patients, ICDs reduce sudden cardiac death by 57% relative to usual care and improve overall survival.⁴ Implantable cardioverter defibrillators, however, generally do not improve quality of life.⁵

Cardiac resynchronization therapy (CRT), also called atrial synchronized biventricular pacing, is a more recently developed pacemaker-based technology that corrects the underlying dyssynchronous left ventricular contraction that is common in failing hearts.^{6 - 7} Unlike traditional dual-chamber pacemakers that pace only the right atrium and the right ventricle, CRT devices use an additional pacing lead that stimulates the lateral wall of the left ventricle.^{6 ,8} A recent meta-analysis⁹ of randomized trials found that CRT reduces death from progressive heart failure by 51% relative to controls and shows a trend toward reducing all-cause mortality. In addition, CRT improves cardiac function and quality of life.^{8 ,10} Cardiac resynchronization therapy has no apparent effect on sudden cardiac death.⁹

Given the complementary benefits of CRT and ICDs, a single device incorporating both resynchronization and defibrillation capabilities has great appeal. Coinciding with their approval by the Food and Drug Administration last year, the number of combined resynchronization-defibrillation (CRT-ICD) devices implanted in the United States increased dramatically from 1000 in 2001 to 14 000 in 2002 to an estimated 33 200 this year. More than 25% of all ICDs implanted in the United States in 2003 are likely to be CRT-ICD devices (G. Reicin, Morgan Stanley, written communication, May 8, 2003). Despite the increasing use of these combined devices, evidence of their merits from rigorous clinical trials is limited.

In this issue of THE JOURNAL, Young and colleagues¹¹ present the first published report of a completed randomized trial (The Multicenter InSync ICD Randomized Clinical Evaluation [MIRACLE ICD]) evaluating the safety and efficacy of devices combining resynchronization and defibrillation therapy. In their study, 369 patients with moderate-to-severe chronic heart failure, left ventricular ejection fraction of 35% or less, and left ventricular dyssynchrony (as measured by a QRS interval ≥130 ms), who were at high risk for life-threatening ventricular arrhythmias underwent successful CRT-ICD device implantation. Patients were then randomly assigned to receive CRT in the on vs off mode. All patients had the cardioverter-defibrillator function of their devices turned on throughout the trial. After 6 months of follow-up, quality of life, New York Heart Association functional class, and exercise duration were all improved to a greater degree among patients treated with active CRT vs controls. These findings confirm benefits of CRT previously shown in other trials in which patients did not have ICDs.^{8 ,10 ,12} In contrast with previous studies,^{8 ,10 ,12} however, MIRACLE ICD did not find a statistically significant association between CRT and improvements in 6-minute walk distance, left ventricular ejection fraction, or hospitalization rate.

Importantly, MIRACLE ICD shows that CRT does not interfere with cardioverter-defibrillator function. The time required for the device to detect ventricular fibrillation did not differ significantly between the 2 treatment groups.^{11 ,13} In addition, 99.9% of episodes of spontaneous ventricular tachycardia or ventricular fibrillation among patients with active CRT were terminated within a preset time by the patients' cardioverter-defibrillators vs 98.3% of episodes of spontaneous ventricular tachycardia or ventricular fibrillation among controls. Cardiac resynchronization therapy also had no appreciable effect on the percentage of patients with ventricular tachycardia or ventricular fibrillation, or the percentage of patients with inappropriate defibrillator shocks.

If CRT-ICD devices have a potential weakness, it is the left ventricular lead. Transvenous left ventricular leads can be difficult to implant. Implantation requires threading the lead from the right atrium through the coronary sinus to a cardiac vein along the lateral wall of the left ventricle. Transvenous left ventricular lead implantation was unsuccessful in 12% of patients in MIRACLE ICD^{11 ,13} and in 13% of patients in another CRT-ICD randomized trial.¹⁴ In this latter trial, the most common reason for failed left ventricular lead implantation was inability to locate or cannulate the coronary sinus.¹⁴ Only 63% of patients undergoing attempted device implantation in MIRACLE ICD had a left ventricular lead tip positioned along the lateral wall of the left ventricle,¹³ a site that likely maximizes the hemodynamic benefit of cardiac resynchronization.⁸ Cardiac perforation is an uncommon but potentially serious complication of left ventricular lead implantation, occurring in 1% of cases.¹¹ Once implanted, left ventricular leads are prone to dislodgement. In MIRACLE ICD, dislodgement led to an additional invasive lead repositioning or replacement procedure in 12% of patients.¹¹ Without a well-positioned left ventricular lead, CRT-ICD devices lose their ability to perform cardiac resynchronization and, in turn, their ability to improve quality of life and decrease death from progressive heart failure.

There is good reason to be cautiously optimistic that CRT-ICD device implantation success rates and complication rates will improve. With greater physician experience, implantation success rates approach 95%.¹⁴ Efforts are under way to improve left ventricular lead delivery systems and lead stability. For those patients who cannot have a transvenous left ventricular lead successfully placed, new surgical techniques allow epicardial lead placement via thoracotomy with minimal complications.^{15 - 16}

Mortality within 30 days of attempted device implantation in MIRACLE ICD was 1.2%, which is not unexpected given the severity of the patients' underlying heart failure. This mortality rate is comparable with that of conventional ICD implantation among patients with a left ventricular ejection fraction of 30% or less¹⁷ and is substantially less than the 30-day mortality rate of 5.3% for patients with NYHA class III or IV functional status and a left ventricular ejection fraction of 35% or less undergoing coronary artery bypass graft surgery (E. D. Peterson, Society of Thoracic Surgeons Adult Cardiac Database, written communication, April 18, 2003). These comparisons are unadjusted for differences in baseline patient characteristics and other confounders.

How many patients might benefit from a CRT-ICD device? Approximately 10% of patients with heart failure have a reduced ejection fraction, left ventricular dyssynchrony, and moderate-to-severe heart failure symptoms.¹⁸ Assuming that roughly half of these patients also have an indication for an ICD,¹⁹ perhaps 5% or 250 000 of the approximately 5 million patients with heart failure in the United States³ might be eligible for CRT-ICD therapy. The MIRACLE ICD trial excluded patients with atrial fibrillation or a traditional indication for a pacemaker. Whether CRT-ICD therapy should be extended to these groups of heart failure patients, as well as those with mild heart failure or mechanical dyssynchrony with a narrow QRS interval, is the subject of active investigation.^{8 ,20 - 21}

Should all patients meeting the MIRACLE ICD inclusion criteria of moderate-to-severe heart failure, left ventricular ejection fraction of 35% or less, left ventricular dyssynchrony, and an indication for an ICD be referred for CRT-ICD device implantation? Although the trial results are encouraging, several caveats must be considered when attempting to apply the findings of MIRACLE ICD to individual patients.

First, MIRACLE ICD was not designed to directly address the question of whether CRT-ICD therapy decreases mortality, and no statistically significant difference in mortality between the 2 treatment groups was observed. Definitive information about CRT-ICD therapy and mortality should be available soon from the Comparison of Medical Therapy, Pacing, and Defibrillation in Chronic Heart Failure (COMPANION) trial.²² Preliminary results from COMPANION show that CRT-ICD therapy decreases all-cause mortality by 43% relative to optimal pharmacological therapy.²² COMPANION will also address the effect of CRT-ICD therapy on survival among patients with nonischemic heart failure, a group that has not previously been shown to benefit from primary prevention ICD therapy alone.^{4 ,19}

Second, MIRACLE ICD randomized 369 patients and followed up these patients for 6 months. This is not a long enough follow-up period for detecting late or uncommon complications that may be associated with a relatively new technology like CRT-ICD therapy. Physicians, industry, and regulators must remain vigilant for such complications.^{23 - 26}

Third, MIRACLE ICD shows that, as with most therapies,²⁷ CRT-ICD therapy benefits some patients but not others. A total of 52% of patients in the combined treatment group had an improvement in their composite heart failure clinical response vs a worsening response in 33%. Comparable numbers for controls were 43% and 34%. These findings underscore the importance of studies aimed at optimizing patient selection, left ventricular lead placement, and device programming.

Fourth, patients in MIRACLE ICD were randomized only after undergoing successful CRT-ICD device implantation. The trial findings do not directly reflect the 12% rate of unsuccessful implantation. As a result, real-world outcomes for patients referred for CRT-ICD device implantation may be less favorable than those described for randomized patients in MIRACLE ICD.

Fifth, the majority of patients in MIRACLE ICD were taking angiotensin-converting enzyme inhibitors and β-blockers. These drugs have many of the same benefits shown and postulated for CRT-ICD therapy.^{1 - 2} Combined resynchronization-defibrillation device implantation is indicated only for those patients who remain symptomatic from heart failure despite treatment with proven heart failure medications including angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers and β-blockers, if tolerated.

The MIRACLE ICD trial is an important step forward in understanding the role of device therapy for heart failure. History shows that when 2 beneficial therapies are combined, unanticipated outcomes and even harm can occur.²⁸ In the case of CRT-ICD devices, however, MIRACLE ICD demonstrates that, for a selected group of patients with heart failure, treatment with these devices can improve quality of life and functional class without compromising ICD function. Ongoing and future studies will refine and possibly expand the indications for this promising therapy.