After decades focused on pharmacologic management, the year 2001 ushered in a new era of implantable device therapies for the management of heart failure (HF) with the U.S. Food and Drug Administration (FDA) approval of the first cardiac resynchronization therapy (CRT) device. Soon thereafter, randomized controlled trials (RCTs) were published that supported the routine use of implantable cardioverter-defibrillators (ICDs) and combined CRT-ICD devices in the management of HF. Although ICDs had already been indicated for the management of resuscitated cardiac arrest, ventricular fibrillation, and hemodynamically destabilizing ventricular tachycardia in HF patients, these newer studies demonstrated mortality reduction with the prophylactic use of an ICD, substantially enlarging the population of patients eligible for an ICD. By 2005, the strength of evidence supporting the use of an ICD or CRT with or without a defibrillator in the management of HF was sufficiently strong to recommend use of these therapies in all eligible patients. Despite the overwhelming evidence in support of device therapy for heart failure, ICDs and CRT remain underutilized.
In addition to the therapeutic benefits of CRT and ICDs, implantable devices that monitor HF clinical status and/or hemodynamics have been developed and continue to be under investigation. One such device has been approved by the FDA. By using measurement of intrathoracic impedance, an available CRT-ICD device can track intrathoracic fluid volume changes. Another investigational approach to managing HF is through the use of implantable hemodynamic monitoring systems. These systems enable the day-to-day management of cardiac filling pressures and other physiologic parameters not otherwise available to the clinician. Initial reports suggest a substantial opportunity to reduce HF morbidity rates (e.g., HF hospitalizations), with the use of these devices. This chapter reviews the use of ICDs and CRT for the management of HF, previews the promise of implantable HF monitoring devices, and mentions other investigational device therapies for HF.
Implantable Cardioverter-Defibrillators in the Management of Heart Failure
ICDs were initially applied to survivors of sudden cardiac death (SCD) to treat second episodes of ventricular tachycardia or ventricular fibrillation (see Chapter 22 for a complete discussion of ICD use in secondary prevention of SCD). Patients with left ventricular dysfunction, either from ischemic or nonischemic etiologies, are at increased risk for SCD, the leading cause of mortality in patients with HF, which occurs at a rate six to nine times that seen in the general population. The Metoprolol CR/XL Randomized Intervention Trial in Heart Failure (MERIT-HF) showed that patients with New York Heart Association (NYHA) functional class II or III symptoms die most frequently as a result of SCD. This study estimated the proportion of total mortality attributable to SCD at 64% and 59% for NYHA class II and III patients, respectively. In contrast, the major cause of death in class IV patients in MERIT-HF was progressive or end-stage HF. Thus, all but the sickest HF patients are more likely to die suddenly rather than from worsening HF.
With this background, a series of studies published between 1996 and 2005 expanded the use of ICDs as prophylactic therapy in at-risk subjects (see Table 22-2 ). These studies have focused on patients with coronary artery disease, usually after myocardial infarction (MI), and more recently on those with left ventricular (LV) systolic dysfunction of any cause. In the HF/LV dysfunction population, the Multicenter Automatic Defibrillator Implantation Trial (MADIT) was the first primary prevention study to show the benefit of prophylactic ICD implantation. High-risk patients (n = 196) with prior MI, an LV ejection fraction (LVEF) of 35% or less, and nonsustained ventricular tachycardia at a rate of 3 to 30 beats higher than 120 beats/min underwent electrophysiologic testing and were randomized to an ICD versus conventional antiarrhythmic therapy, primarily amiodarone. Compared with conventional therapy, the ICD arm demonstrated an impressive reduction in all-cause mortality at 2 years (15.8% vs. 38.6%; P = .009). However, significantly more patients in the ICD group were receiving treatment with a β-blocker, which confounded the results of the trial.
MADIT was followed by a number of other encouraging studies of LV systolic dysfunction patients, such as the Coronary Artery Bypass Graft (CABG)-Patch trial and the Multicenter Unsustained Tachycardia Trial (MUSTT). However, the landmark trials establishing a role for ICDs as primary prevention of mortality in HF patients are MADIT II and the National Institutes of Health (NIH)–sponsored Sudden Cardiac Death–Heart Failure Trial (SCD-HeFT). Although underpowered to demonstrate a significant difference for its primary endpoint, the Prophylactic Defibrillator Implantation in Patients with Nonischemic Dilated Cardiomyopathy (DEFINITE) trial also contributed substantially to the burden of proof supporting prophylactic intervention with an ICD in the management of HF.
MADIT II was a prospectively designed RCT powered to assess the survival benefit of ICDs in a population of post-MI patients with reduced ejection fraction (<30%). Importantly, this trial included no arrhythmic markers, such as nonsustained or inducible ventricular tachycardia, for inclusion. A total of 1232 patients were randomly assigned in a 3 : 2 ratio to receive an ICD (742 patients) or conventional medical therapy (490 patients). During an average follow-up of 20 months, the all-cause mortality rates were 19.8% in the conventional therapy arm and 14.2% in the ICD group (31% relative risk reduction [RR]; P = .016; Figure 13-1 ). The effect of ICD therapy on survival was similar in subgroup analyses stratified according to age, gender, ejection fraction, NYHA class, and the QRS interval. Moreover, β-blocker use was 72% in these patients and was well balanced between the ICD and conventional therapy groups.
Of note, the majority of patients enrolled in MADIT II fell into NYHA class II or III. NYHA class IV patients were excluded, and the NYHA class I cohort was relatively small. The average LVEF was 23%, and these findings suggest that HF patients with mild to moderate symptoms and a moderate to severe reduction in LVEF may benefit the most from a prophylactic ICD. Moreover, in contrast to MADIT I, in which the survival benefit of ICD therapy was seen early after randomization, the survival benefit observed in MADIT II began approximately 9 months after the device was implanted. The authors suggested that this difference might be due to a lower risk population being enrolled in MADIT II, the absence of arrhythmia as risk stratification for entry, or the use of more aggressive medical treatment. Regardless of the explanation, this observation may be important when considering the timing of device placement in eligible patients.
Whereas MADIT II enrolled exclusively post-MI patients with an ischemic cause of LV systolic dysfunction and HF, the DEFINITE trial was the first randomized trial of primary prevention therapy with an ICD in nonischemic cardiomyopathy patients. Such patients also exhibit high rates of SCD, but consensus had not been reached regarding the management of SCD risk in such patients. This was due, in part, to limitations in objective risk assessment, in that no invasive or noninvasive testing procedure had been shown to accurately determine which nonischemic HF patient is likely to die suddenly. Also clouding the picture were older observations that suggested prophylactic administration of an antiarrhythmic agent such as amiodarone might prolong survival in nonischemic cardiomyopathy patients.
The DEFINITE trial was a prospective evaluation of 458 patients with nonischemic cardiomyopathy. Entry criteria included an ejection fraction of 35% or less, a history of symptomatic HF, and the presence of ambient arrhythmias defined as episodes of nonsustained ventricular tachycardia or at least 10 premature ventricular contractions in 24 hours on continuous ambulatory electrocardiographic (ECG) monitoring. Patients (n = 458) were equally randomized to each arm of the study to receive either an ICD and standard medical therapy or standard medical therapy alone. Compliance with medical therapy was excellent and included an angiotensin-converting enzyme inhibitor (ACEI) in 86% of the cohort and a β-blocker in 85%. The patients were followed for a mean of 29 ± 14 months with a primary endpoint of all-cause mortality.
There were 68 deaths reported in the DEFINITE trial: 28 in the ICD group and 40 in the standard therapy group. The implantation of an ICD yielded a nonsignificant 35% reduction in death from any cause (hazard ratio [HR], 0.65; 95% confidence interval [CI], 0.40 to 1.06; P = .08) and significantly reduced the risk of sudden death by a remarkable 80% (HR, 0.20; 95% CI, 0.06 to 0.71; P = .006). In the subgroup of NYHA class III patients, all-cause mortality was significantly decreased in the ICD arm (HR, 0.37; 95% CI, 0.15 to 0.90; P = .02). Although this study was underpowered and did not reach statistical significance with respect to the primary endpoint of all-cause mortality for the entire randomized cohort, the results demonstrated a strong trend toward a survival advantage for patients receiving an ICD. It is worth mentioning that the all-cause mortality reduction seen in DEFINITE was 35%—a value strikingly similar to the 31% relative risk reduction observed in the ischemic population studied in MADIT II. The statistical power of DEFINITE was affected by a low rate of SCD in both groups, which may be related to aggressive use of ACEI and β-blockade in this trial.
The results of the SCD-HeFT trial were published in 2005 and have had a substantial impact on current practice and reimbursement guidelines for ICDs. This landmark RCT enrolled 2521 patients from 148 mostly U.S. centers between 1997 and 2001. Patients with NYHA class II (70%) or III (30%) HF and reduced LVEF (≤35% or less, mean ~25%) of either ischemic or nonischemic etiology were eligible for the study. SCD-HeFT was a three-arm trial, comparing treatment with an ICD with amiodarone and placebo. Thus, SCD-HeFT addressed at least two important issues in HF management: 1) whether empirical amiodarone therapy saves lives in well-treated NYHA class II and III heart failure patients with no arrhythmic indication for the drug and 2) whether a prophylactic ICD saves lives in such patients with HF from either an ischemic or nonischemic cause.
In SCD-HeFT, patients received standard HF therapy, if tolerated, which included an ACEI or angiotensin receptor blocker in 85%, β-blocker in 69%, and aldosterone antagonists in 19% (compatible with guideline recommendations at the time the study was conducted). The median follow-up was 45.5 months. Importantly, the cohort was equally divided between ischemic and nonischemic causes of HF, allowing an important subgroup analysis of these cohorts to be done. Mortality rates in the ICD, amiodarone, and placebo groups were 17.1%, 24%, and 22.3% at 3 years and 28.9%, 34.1%, and 35.9% at 5 years, respectively. The ICD was associated with a statistically significant 23% reduction in all-cause mortality compared with placebo (HR, 0.77; 97.5% CI, 0.62 to 0.96; P = .007). Outcomes on amiodarone were not significantly different from placebo across all subgroups (HR, 1.06; 97.5% CI, 0.86 to 1.30). Similar degrees of benefit were noted in patients with ischemic (21% mortality rate reduction) and nonischemic HF (27% mortality rate reduction), thus confirming the findings of MADIT II and DEFINITE, respectively. The SCD-HeFT trial provides the most robust evidence to date to support the prophylactic use of ICDs in patients with mild to moderate HF and reduced ejection fraction from virtually any cause.
Implantable Cardioverter-Defibrillators Early After Myocardial Infarction
Although many of the trials discussed above included patients with ischemic cardiomyopathy, the majority of those patients were months to years removed from their acute infarcts. The relative risk of SCD of those recovering from an acute ischemic event may be different from those with chronic infarct–related scar. In a subgroup analysis of patients enrolled in the Valsartan in Acute Myocardial Infarction Trial (VALIANT), the risk of sudden death was highest within the first 30 days after MI. In those who had an event (occurring a median of 180 days following their infarct), 19% had either sudden death or an aborted cardiac arrest within the first 30 days. Although these findings suggested that an ICD in the early postinfarct period would significantly reduce mortality, the results of prospective randomized trials have proved otherwise. The role of an ICD in the initial weeks following MI has been defined by the landmark Defibrillator in Acute Myocardial Infarction Trial (DINAMIT) and later confirmed by the Immediate Risk Stratification Improves Survival (IRIS) trial.
DINAMIT was the first trial to evaluate the role of prophylactic ICD implantation in the acute infarct period. The study randomized 674 patients who were within 6 to 40 days of MI, had an LVEF of 35% or less, and had impaired cardiac autonomic function, defined as either depressed heart rate variability or an elevated average 24-hour heart rate on Holter monitoring, to prophylactic ICD implant versus no ICD. Patients with sustained ventricular tachycardia more than 48 hours after MI, those with NYHA class IV heart failure, and those who were revascularized were excluded. After a mean follow-up of 30 months, no difference was observed in all-cause mortality between the two groups (HR for death in the ICD group, 1.08; 95% CI, 0.76 to 1.55; P = .66). The prespecified secondary endpoint, arrhythmic death, was more common in the control arm, whereas nonarrhythmic death was more common in the ICD arm. The results of this study provide the primary rationale for current guidelines that recommend deferring ICD implantation for at least 40 days following MI.
The lack of benefit of an ICD in the immediate postinfarct period was confirmed by the large, multicenter IRIS trial, which randomized patients who were within 5 to 31 days of MI to an ICD or best medical therapy alone. Patients with an LVEF of 40% or less and a heart rate of 90 beats/min or greater on the first available ECG and/or those with a nonsustained ventricular tachycardia 150 beats/min or greater were included in the study. A total of 898 patients were followed for a mean of 37 months. Overall, no mortality reduction was observed in the ICD group relative to control (HR, 1.04; 95% CI, 0.81 to 1.35; P = .78). Similar to the DINAMIT findings, although SCDs decreased in the ICD group compared with the control group (HR, 0.55; 95% CI, 0.31 to 1.00; P = .049), this decrease was paralleled by an increase in non-SCD mortality in the ICD group (HR, 1.92; 95% CI, 1.29 to 2.84; P = .001).
Indications for Prophylactic Cardioverter-Defibrillator Implantation in Heart Failure Patients
Based on these trials, the indication for an ICD was extended to NYHA class II and III heart failure patients with a reduced ejection fraction (see Chapter 22 ). The 2009 ACC/AHA guidelines for HF promote class I indications for the use of an ICD as primary prevention of all-cause mortality in well-treated NYHA class II and III patients with either ischemic or nonischemic cardiomyopathy and an LVEF of 35% or less. Those with ischemic cardiomyopathy, NYHA class I functional status, and an ejection fraction less than 30% are also assigned a class I indication. ICD implantation in those with a recent MI must be deferred until 40 days have passed from the date of acute MI. For those who have been surgically or percutaneously revascularized, the Centers for Medicare and Medicaid Services require waiting 3 months prior to ICD implant. In addition, the ACC/AHA guidelines qualify that ICD candidates should have a reasonable expectation of survival with a good functional status for more than 1 year.
Practical Considerations in Implantable Cardioverter-Defibrillator Therapy
Before ICD implantation, patients should have a thorough understanding of the risks and benefits of device therapy, as well as the need for routine follow-up and defibrillation threshold testing and the morbidity associated with appropriate and inappropriate ICD shocks. Patients should understand that ICDs have been shown to prolong survival, but they will not improve HF symptoms, nor will they slow disease progression. It should also be explained to the patient and family that the defibrillator function of an implantable device can be turned off as part of end-of-life care. Questions regarding exercise, driving, cellular telephones, and airport security, among others, should be anticipated and addressed in advance. The management of complications after ICD implantation is discussed in Chapter 22 . Monitoring for bleeding or infection is paramount during the early postimplant period, although seeding of the device from a distant infectious source may occur at any time. For patients who develop new or recurrent ventricular arrhythmias that cause frequent ICD shocks, therapeutic options include antiarrhythmic drugs such as amiodarone or mexiletine, catheter ablation of ventricular tachycardia, or, in selected cases, consideration of mechanical circulatory support or heart transplant. In patients with ischemic heart disease, the occurrence of polymorphic ventricular tachycardia that leads to ICD shocks should warrant reassessment for ischemia. Patients who develop significant anxiety related to ICD shocks may benefit from anxiolytic therapy and/or referral to a therapist or support group. Finally, it is important to remember that ICDs may aggravate HF and result in increased HF hospitalizations. This may result from mechanical dyssynchrony induced by right ventricular pacing.
Conduction Abnormalities in Heart Failure
A number of conduction abnormalities are seen in the setting of chronic HF. Approximately one third of patients with HF and reduced ejection fraction have a QRS duration greater than 120 ms, most commonly seen as a left bundle branch block (LBBB). Such conduction delays produce suboptimal ventricular filling, a reduction in LV contractility, extended mitral regurgitation, and paradoxical septal wall motion. These mechanical manifestations of ventricular conduction abnormalities have been referred to as ventricular dyssynchrony, especially because the septum and LV free wall no longer contract in a normal near-simultaneous fashion. This situation reduces the ability of the failing heart to eject blood and has been associated with increased mortality in HF patients.
In the mid-1990s, the application of pacing therapies to overcome ventricular dyssynchrony began to be explored. In particular, atrial-synchronized biventricular pacing—now known as cardiac resynchronization therapy (CRT)—emerged as the most promising approach for the treatment of ventricular dyssynchrony. A series of studies confirmed the benefits of CRT in NYHA functional class III and IV heart failure patients with ventricular dyssynchrony and led to a strong contemporary recommendation for the use of this therapy.
The first application of atrial-synchronized biventricular pacing was performed by Cazeau and colleagues, who used four-chamber pacing in a 54-year-old man with NYHA class IV heart failure and significant atrial-ventricular and ventricular conduction disturbances. Standard transvenous pacing leads were placed in the right atrium and right ventricle. The left atrium was paced by a lead placed in the coronary sinus, whereas the LV was paced by an epicardial lead located on the LV free wall. After 6 weeks of pacing, the patient’s clinical status improved markedly, with a weight loss of 17 kg and a disappearance of peripheral edema. His functional class improved remarkably to NYHA class II. Such favorable anecdotal experiences led to the inception of small studies to evaluate the acute effects of biventricular pacing on systemic hemodynamics. These studies provided additional proof of concept that CRT might reverse the deleterious consequences of ventricular dyssynchrony. Several studies soon followed to further evaluate the acute and longer-term effects of CRT in HF. The results were equally encouraging, with patients showing consistent, sustained improvement in exercise tolerance, quality of life (QOL), NYHA functional class, and cardiac output. With the advent of a transvenous, rather than an epicardial, approach for pacing the LV, larger-scale observational and randomized controlled trials of CRT were made possible.
Landmark Cardiac Resynchronization Therapy Clinical Trials
Substantial evidence supports the beneficial effects of CRT for the treatment of HF. Approximately 4000 patients have been evaluated in randomized single- or double-blind controlled trials, including large-scale morbidity and mortality studies. The most important of these trials are the Multisite Stimulation in Cardiomyopathy (MUSTIC) studies, the Multicenter InSync Randomized Clinical Evaluation (MIRACLE) trial, MIRACLE ICD, the Safety and Effectiveness of Cardiac Resynchronization Therapy with Defibrillation (CONTAK CD) trial, the Cardiac Resynchronization in Heart Failure (CARE-HF) trial, and the Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) trial.
The MUSTIC trials were designed to evaluate the safety and efficacy of cardiac resynchronization in patients with advanced HF, ventricular dyssynchrony, and either normal sinus rhythm or atrial fibrillation. They represent the first randomized single-blind trials of CRT for HF. The first study involved 58 randomized patients with NYHA class III heart failure, normal sinus rhythm, and a QRS duration of at least 150 ms. All patients were implanted with a CRT device, and after a run-in period, patients were randomized in a single-blind fashion to either active pacing or no pacing. After 12 weeks, patients were crossed over and remained in the alternate study assignment for 12 weeks. After completing this second 12-week period, the device was programmed to the patient’s preferred mode of therapy.
The second MUSTIC study involved fewer patients (only 37 completers) with atrial fibrillation and a slow ventricular rate, either spontaneously or from radiofrequency ablation. A VVIR biventricular pacemaker and leads for each ventricle were implanted, and the same randomization procedure just described was applied; however, biventricular VVIR pacing versus single-site right ventricular VVIR pacing, rather than no pacing, was compared in this group of patients with atrial fibrillation. The primary endpoints for MUSTIC were exercise tolerance assessed by measurement of peak oxygen consumption (VO 2 ) or the 6-minute walk test and QOL determined using the Minnesota Living with Heart Failure questionnaire. Secondary endpoints included rehospitalizations and/or drug therapy modifications for worsening HF. Results from the normal sinus rhythm arm of MUSTIC provided strong evidence of benefit. The mean distance walked in 6 minutes was 23% greater with CRT than during the inactive pacing phase ( P < .001). Significant improvement was also seen in quality of life and NYHA functional class, with fewer hospitalizations during active resynchronization therapy. The atrial fibrillation cohort evaluated in MUSTIC demonstrated similar improvements, although the magnitude of benefit was slightly less.
MIRACLE was the first prospective, randomized, double-blind, parallel-controlled clinical trial designed to evaluate the merits of CRT and to further elucidate potential mechanisms of action of CRT. Primary endpoints were NYHA class, QOL score (using the Minnesota Living with Heart Failure questionnaire), and 6-minute walk distance. Secondary endpoints included assessments of a composite clinical response, cardiopulmonary exercise performance, neurohormone and cytokine levels, QRS duration, cardiac structure and function (as determined by echocardiography), and a variety of measures of worsening HF and combined morbidity and mortality.
The MIRACLE trial was conducted between 1998 and 2000. Patients (n = 453) with moderate to severe symptoms of HF associated with an LVEF of 35% or less and a QRS duration of at least 130 ms were randomized (double-blind) to cardiac resynchronization (n = 228) or to a control group (n = 225) for 6 months while conventional therapy for HF was maintained. Compared with the control group, patients randomized to CRT demonstrated a significant improvement in QOL score (−18.0 vs. −9.0 points; P = .001), 6-minute walk distance (+39 vs. +10 meters; P = .005), NYHA class (−1.0 vs. 0.0 class; P < .001), treadmill exercise time (+81 vs. +19 seconds; P = .001), peak VO 2 (+1.1 vs. 0.1 mL/kg/min; P < .01), and LVEF (+4.6% vs. −0.2%; P < .001) ( Figure 13-2 ).
Patients randomized to CRT demonstrated a highly significant improvement in a composite clinical HF response endpoint compared with control subjects, suggesting an overall improvement in HF clinical status. In addition, when compared with the control group, fewer patients in the CRT group required hospitalization (8% vs. 15%) or intravenous medications (7% and 15%) for the treatment of worsening HF (both P < .05). In the resynchronization group, the 50% reduction in hospitalization was accompanied by a significant reduction in length of stay, resulting in a 77% decrease in total days hospitalized over 6 months compared with the control group. The major limitation of the therapy was caused by unsuccessful implantation of the device in 8% of patients. The results of this trial led to the FDA approval of the InSync system (Medtronic, Inc., Minneapolis, MN) in August 2001, the first approved CRT system in the United States, allowing the introduction of CRT into clinical practice.
The MIRACLE trial also provided persuasive evidence supporting the occurrence of reverse LV remodeling with chronic CRT. In the MIRACLE trial, serial Doppler echocardiograms were obtained at baseline and at 3 months and 6 months in a subset of 323 patients. CRT for 6 months was associated with reduced end-diastolic and end-systolic volumes (both P < .001), reduced LV mass ( P < .01), increased ejection fraction ( P < .001), reduced mitral regurgitation ( P < .001), and improved myocardial performance index ( P < .001) compared with controls. These effects are similar to those seen with β-blockade in HF, but in the MIRACLE trial, they were seen in patients already receiving β-blocker therapy.
The MIRACLE ICD study was designed to be almost identical to the MIRACLE trial. MIRACLE ICD was a prospective, multicenter, randomized, double-blind, parallel-controlled clinical trial intended to assess the safety and efficacy of a combined CRT-ICD system in patients with dilated cardiomyopathy (LVEF ≤35%, LV end diastolic dimension ≤55 mm), NYHA class III or IV heart failure, ventricular dyssynchrony (QRS ≥130 ms), and an indication for an ICD. Primary and secondary efficacy measures were essentially the same as those evaluated in the MIRACLE trial but also included measures of ICD function, including the efficacy of antitachycardia therapy with biventricular pacing.
Of 369 randomized patients who received devices, 182 were controls (ICD active, CRT inactive), and 187 were in the resynchronization group (ICD active, CRT active). At 6 months, patients assigned to active CRT had a greater improvement in median QOL score (−17.5 vs. −11.0; P = .02) and functional class (−1 vs. 0; P = .007) than controls but were no different than controls in the change in distance walked in 6 minutes (55 vs. 53 m; P = .36). Peak oxygen consumption increased by 1.1 mL/kg/min in the resynchronization group, versus 0.1 mL/kg/min in controls ( P = .04), and treadmill exercise duration increased by 56 seconds in the CRT group and decreased by 11 seconds in controls ( P = .0006). The magnitude of improvement was comparable with that seen in the MIRACLE trial, suggesting that HF patients with an ICD indication benefit as much from CRT as those patients without an indication for an ICD. The combined CRT-ICD device used in this study was approved in June 2002 by the FDA for use in NYHA class III and IV heart failure patients with reduced ejection fraction, ventricular dyssynchrony, and an ICD indication.
The CONTAK CD trial enrolled 581 symptomatic HF patients with ventricular dyssynchrony and malignant ventricular tachyarrhythmias, all of whom were candidates for an ICD. Following unsuccessful implant attempts and withdrawals, 490 patients were available for analysis. The study did not meet its primary endpoint of a reduction in disease progression, defined by a composite endpoint of HF hospitalization, all-cause mortality, and ventricular arrhythmia requiring defibrillator therapies, although the trends were in a direction favoring improved outcomes with CRT. However, the CONTAK CD trial did show statistically significant improvements in peak oxygen uptake and quality of life in the resynchronization group compared with control subjects, although quality of life was improved only in NYHA class III and IV patients without right bundle branch block (RBBB). LV dimensions were also reduced, and LVEF increased, as seen in other trials of CRT. Importantly, the improvement seen in peak oxygen consumption with cardiac resynchronization was again comparable with that observed in the MIRACLE trial. Improvements in NYHA functional class were not observed in this study.
Begun in early 2000, COMPANION was a multicenter, prospective, randomized, controlled clinical trial designed to compare drug therapy alone with drug therapy in combination with cardiac resynchronization in patients with dilated cardiomyopathy, an intraventricular conduction defect (IVCD), NYHA class III or IV heart failure, and no indication for a device. COMPANION randomized 1520 patients into one of three treatment groups in a 1 : 2 : 2 allocation. Group 1 (308 patients) received optimal medical care only, group II (617 patients) received optimal medical care and the Guidant (Indianapolis, IN) CONTAK TR (biventricular pulse generator), and group III (595 patients) received optimal medical care and the CONTAK CD (combined HF/bradycardia/tachycardia device). The primary endpoint of the COMPANION trial was a composite of all-cause mortality and all-cause hospitalization, measured as time to first event, beginning from time of randomization. Secondary endpoints included all-cause mortality and a variety of measures of cardiovascular morbidity.
When compared with optimal medical therapy alone, the combined endpoint of mortality or HF hospitalization was reduced by 35% for patients receiving CRT and by 40% for patients receiving CRT-ICD (both P < .001). For the mortality endpoint alone, CRT patients had a 24% risk reduction ( P = .060) and CRT-ICD patients had a 36% risk reduction ( P < .003) compared with optimal medical therapy. COMPANION confirmed the results of earlier resynchronization therapy trials in improving symptoms, exercise tolerance, and quality of life for HF patients with ventricular dyssynchrony. In addition, COMPANION showed, for the first time, the impact of CRT-ICD in reducing all-cause mortality.
The Cardiac Resynchronization–Heart Failure (CARE-HF) trial was designed to evaluate the effects of resynchronization therapy without an ICD on morbidity and mortality rates in patients with NYHA class III or IV heart failure and ventricular dyssynchrony. Patients (n = 813) with an LVEF of 35% or less and ventricular dyssynchrony—defined as a QRS duration of 150 ms or greater, a QRS duration between 120 and 150 ms, with echocardiographic evidence of dyssynchrony—were enrolled in this randomized, unblinded, controlled trial and were followed up for an average of 29 months. Optimal medical therapy alone was assigned to 404 patients, and 409 patients were randomized to optimal medical therapy plus resynchronization therapy. The risk of death from any cause or unplanned hospitalization for a major cardiac event, the primary endpoint analyzed as time to first event, was significantly reduced by 37% in the treatment group compared with control subjects (HR, 0.63; 95% CI, 0.51 to 0.77; P < .001). In the CRT group, 82 patients (20%) died during follow-up compared with 120 patients (30%) in the medical group, yielding a significant 36% reduction in all-cause mortality with resynchronization therapy (HR, 0.64; 95% CI, 0.48 to 0.85; P < .002; Figure 13-3 ). Resynchronization therapy also significantly reduced the risk of unplanned hospitalization for a major cardiac event by 39%, all-cause mortality plus HF hospitalization by 46%, and HF hospitalization by 52%.