All Patients With Heart Failure and Intraventricular Conduction Defect or Dyssynchrony Should Not Receive Cardiac Resynchronization Therapy
Case Presentation: J.S., a 72-year-old man, was diagnosed with heart failure 5 years ago. Before that, he had suffered a large anterior myocardial infarction and had undergone successful coronary artery bypass grafting surgery. An internal cardioverter-defibrillator (ICD) was implanted 3 years ago. He was hospitalized briefly for decompensated heart failure 9 months ago, and his medical regimen was intensified at that time. He subsequently resumed his usual activities, which included daily walks of up to several hundred yards around his company’s construction work site and 2 to 3 rounds of golf weekly. His only symptoms were fatigue after 36 holes of golf and shortness of breath while walking uphill rapidly. Medications included 0.125 mg digoxin, 40 mg furosemide, and target doses of an angiotensin-converting enzyme inhibitor, a β-blocker, and warfarin. An ECG demonstrated atrial fibrillation with a controlled ventricular response. The QRS duration was 130 ms with a left bundle-branch block pattern. On echocardiogram, the left ventricular (LV) ejection fraction (EF) was 23% with anterior wall akinesis; the LV internal diastolic dimension was 83 mm; and there was trace mitral regurgitation. Cardiac resynchronization therapy (CRT) was recommended. Placement of an LV pacing lead and “upgrading” his ICD to an ICD/biventricular pacemaker was performed uneventfully.
Response by Abraham p 2691
Shortly after the procedure, he experienced recurrent ventricular tachyarrhythmias, and his ICD, which had not discharged since implantation, delivered 4 shocks over a 24-hour period to terminate episodes of ventricular fibrillation. There was no evidence of acute myocardial infarction, worsening heart failure, or a pericardial effusion. The ventricular arrhythmias were responsive to intravenous amiodarone. After 7 days, he was discharged on oral amiodarone in addition to his other medications. Over the next several weeks, he had only 1 episode of ventricular tachycardia terminated by pacing. During this period, however, he complained of fatigue, malaise, and gastrointestinal symptoms. Amiodarone was discontinued, and over the next several weeks, these symptoms resolved. There were no further ventricular arrhythmias, and J.S. resumed his activities at work and on the golf course. Repeat echocardiogram showed no interval change in LV size or function.
In recognition of the fact that heart failure is an important and growing public health problem, a variety of new treatment approaches have been developed. Beginning in the 1980s, drugs that target maladaptive neurohormonal activation, such as angiotensin-converting enzyme inhibitors, β-blockers, angiotensin receptor blockers, and aldosterone antagonists, have been shown in large-scale clinical trials to produce impressive reductions in morbidity and mortality in heart failure patients.1 Additional agents such as the combination of hydralazine/isosorbide dinitrate also favorably affect the clinical course (at least in selected populations), and new drugs are in various stages of development.2 Although several of these drugs may yet prove to be effective in treating heart failure, there is concern that these newer agents (and particularly those that target neurohormonal activation) may not favorably alter the natural history nearly as well as the drugs that were incorporated into the therapeutic regimen over the past few years.3
Using CRT to Treat Heart Failure Patients
In view of the increased prevalence of heart failure and the persistent unacceptably high morbidity and mortality associated with this condition despite current pharmacological treatments, novel therapeutic approaches have been sought. Recent attention has focused on cardiac dyssynergy as a viable therapeutic target. Many patients with heart failure have abnormalities in the temporal pattern of cardiac contraction that impair both systolic and diastolic functions of the heart and worsen mitral valvular incompetence.4 Mechanical dyssynergy often is seen in association with the electrical dyssynergy that is manifest by a widening of the QRS duration on the surface ECG. The prevalence of a wide QRS in heart failure populations is estimated to be in the range of ≈25% (or greater in more severely ill patients) and when present has been associated with increased mortality.
Recognition of the adverse impact of cardiac dyssynergy in heart failure patients has stimulated the development of novel pacing techniques to correct the problem. The most widely accepted and studied of these has been the use of a biventricular pacemaker in which pacing leads are positioned in the right atrium and right ventricle and a third lead is introduced through the coronary sinus into the venous system of the heart so that LV stimulation can be obtained. This approach allows optimization of timing between atrial and ventricular contractions and enhanced synchronization of interventricular and intraventricular contractions. The results of clinical trials using CRT have now provided incontrovertible evidence of the efficacy of this approach in highly selected patients.5,6 Not only have these studies demonstrated increased exercise capacity, relief of symptoms, and improvement in quality of life, but also there has been evidence that CRT can reduce morbidity and mortality even beyond what is offered by optimal medical therapy of heart failure.6 As a result, CRT has emerged over the past few years as an important adjunct to medical therapy and has brought unquestionable benefits to many patients. Whether it helps and should be offered to all patients with heart failure and intraventricular conduction defect or dyssynchrony, however, is another question entirely.
Should CRT Be Used in All Heart Failure Patients?
The decision to use CRT in heart failure patients must be based on evidence from clinical trials demonstrating that it is both effective and safe to use. The heart failure population is highly heterogeneous; patients exhibit considerable variability in many aspects of their condition. Differences between patients in age, gender, race, origin of heart failure, extent of cardiac dysfunction (both systolic and diastolic), degree of functional and symptomatic impairment, extent of cardiac dyssnynergy, presence of comorbidities, and background therapy all influence the clinical course and the response to additional treatment modalities. Ideally, clinical trials assessing new therapies should include large, well-defined populations that will help determine which patients are most likely to benefit. In this way, their results can be applied in a more rational way to patients seen in everyday clinical practice.
It is instructive to examine the characteristics of the patients who were included in clinical trials assessing the efficacy of CRT (Table 1).7 First, it is important to recognize that heart failure patients with preserved EF, a group now estimated to make up approximately half of the heart failure population, would not be considered for CRT, at least according to the results of available clinical studies that excluded patients with LV EFs >0.35. Second, for the most part, the clinical trial patients had evidence of marked prolongation of their QRS (usually with a left bundle-branch block pattern), were in sinus rhythm, and had advanced (but not end-stage) heart failure while being treated with good background medical therapy (Table 1).
Are the patients who were included in the clinical trials demonstrating efficacy of CRT representative of the heart failure patients who are seen on a daily basis in clinical practice? This question can be answered by examining the clinical characteristics of the US heart failure population. A reasonable scenario is that 50% will have evidence of systolic dysfunction with an EF ≤0.35, 30% will have New York Heart Association (NYHA) class III or “ambulatory” class IV symptoms, and 30% will have a QRS duration ≥120 ms. In this case, only ≈4.5% of heart failure patients would be considered candidates for CRT.
Even this analysis, however, greatly overestimates the number of patients for whom there are convincing data supporting the use of CRT. Most of the patients in the clinical trials were in sinus rhythm and had a left bundle-branch block pattern on the surface ECG. Because synchrony between atrial and ventricular contractions and, in particular, improved synchronization of LV contraction both appear to contribute to the benefits of CRT, the efficacy of CRT in patients with atrial fibrillation and those with conduction abnormalities other than a left bundle-branch block is uncertain.8 Many potential candidates also are likely to be very elderly (ie, >80 years of age). Because the number of such patients in the clinical trials in this age range was limited, there is less certainty about the response to CRT than there is in a younger population, particularly in regard to improving exercise capacity and quality of life. Furthermore, important comorbidities and other issues (both medical and nonmedical) in these patients would diminish enthusiasm for proceeding with CRT and would raise questions about the ultimate cost-efficacy of this approach.9
In addition to the above caveats, a substantial number of the remaining potential patients will have a QRS duration between 120 and 150 ms. Although such patients were entered in the clinical trials assessing the efficacy of CRT, it is clear that the average QRS duration was considerably longer in the patients studied and on average was in the range of 160 to 170 ms (Table 1). Whether patients with moderate prolongation of the QRS interval will benefit to the same extent as those with more marked prolongation is debatable. In the Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) trial, patients who were randomized to CRT experienced a 19% reduction (P=0.014) in the primary end point of time to death or all-cause hospitalization compared with patients randomized to optimal medical treatment.10 However, when the population was analyzed according to extent of QRS prolongation, only the patients whose QRS was >168 ms experienced a significant reduction in risk. Patients whose QRS was ≤147 ms demonstrated essentially no benefit of treatment, whereas CRT had an intermediate effect in the patients whose QRS duration fell between these groups. Although these results should not be viewed as definitive evidence of the lack of efficacy of CRT in patients with less prolonged QRS duration, they do raise an important concern about this population. Perhaps this finding should come as no surprise because QRS duration is only a surrogate for the presence of dyssynergy. Although patients with markedly prolonged QRS duration would be most likely to demonstrate considerable amounts of dyssynergy and thus be more likely to respond to CRT, the converse also is likely to be true.11 This possibility was, in fact, recognized by the investigators in the Cardiac Resynchronization-Heart Failure (CARE-HF) study, who mandated that patients with a QRS duration between 120 and 150 ms have evidence of ventricular dyssynergy by echocardiographic examination to be included in the trial.6 The number of patients excluded because of an absence of demonstrable dyssynergy was not reported. That the favorable effects of CRT in CARE-HF seem to show less dependence on QRS duration than they do in the COMPANION study (which did not require evidence of mechanical dyssynchrony) is likely a manifestation of a lower prevalence of dyssynergy in the COMPANION patients with shorter QRS duration.
Finally, it is important to recognize that the database supporting the use of CRT included patients who were receiving optimal medical management. To the great credit of the investigators who participated in these studies, the percentage of patients receiving angiotensin-converting enzyme inhibitors (and/or angiotensin receptor blockers) and β-blockers was very high. This is relevant to this discussion because the use of these drugs in general practice appears to be considerably less than in the CRT studies. Besides being an indication of poor medical practice, the failure to optimize medical therapy in potential CRT candidates could have 2 potential consequences. The first is that the impact of CRT on the clinical course might be different (and either more or less effective) when initiated against a background of suboptimal medical treatment. The second is that many patients being treated with a suboptimal treatment regimen will experience significant improvement in either EF or NYHA functional class when the treatment regimen is optimized. In this case, CRT might not offer the same degree of benefit as it did to patients in the clinical trials.
It may well be that the indications for CRT will expand over the coming years and that the population that benefits will be larger than our estimate. For instance, CRT has been shown to reduce LV volumes and to increase EF in patients with NYHA functional class II symptoms of heart failure.12 This presumed reversal of cardiac remodeling argues well for the use of CRT in this population because reverse remodeling usually is associated with an improvement in the clinical course, including a reduction in mortality. However, reverse remodeling remains a surrogate for clinical outcomes, and it is noteworthy that these less severely impaired patients failed to achieve a significant improvement in exercise capacity (which was the primary end point of the study) or in quality of life. Thus, more definitive evidence of efficacy of CRT in this less symptomatic population is needed. Similarly, smaller clinical trials in patients with atrial fibrillation or QRS prolongation with other than left bundle-branch block pattern or with evidence of dyssynchrony in the absence of QRS prolongation on the surface ECG have shown trends which suggest that CRT may be helpful in these groups. If these results are confirmed by larger and more rigorous studies, the paradigm for the use of CRT will have to be broadened to include these patients. Similarly, studies using larger numbers of very old patients will help to determine the efficacy of CRT in this important subpopulation.
Issues Related to CRT Use in Practice
What then are the major problems encountered with the implementation of CRT in clinical practice? From our observations, they are largely related to physicians being “too quick to pull the trigger” for recommending CRT and extrapolation of the current database so that the indications for CRT are expanded well beyond those that can justified by currently available clinical trial results. Although such practices may serve to expand the population in which CRT is used, there is little justification for this approach.
The timing of the implementation of CRT should be based on an algorithm that incorporates evidence from the clinical trials with existing knowledge of the natural progression of the disease and its response to available treatments. Typically, clinical trials of CRT enrolled patients when they were clinically stable but still demonstrated substantial symptomatic limitation despite optimal medical treatment. Thus, patients in clinical practice presenting with recent-onset heart failure and those in whom there has been a recent decompensation or recent significant cardiac event should not be considered candidates for CRT until the treatment regimen has been optimized as outlined in Table 2. Many patients who at first appear to be candidates for CRT based on presence of low EF and limitation of activities as a result of symptoms may well show improvement in one or both of these parameters with appropriate treatment. Substantial improvement in EF (and even normalization in some cases) with treatment of ischemia, with improvement in loading conditions, or after administration of neurohormonal blocking agents may not be immediate and often becomes apparent only over time. Similarly, there is evidence that QRS changes may occur in response to treatment of heart failure.13 Although this may pose a dilemma in some cases, it argues for delaying CRT for a period (usually for 3 to 6 months in our practices) after initiation and optimization of treatment to determine whether the patient continues to manifest evidence of cardiac dysfunction and heart failure symptoms that would classify them as candidates for this therapy according to entry criteria in the relevant clinical trials. Thus, it is important for potential candidates to wait for a period of 3 months, during which time therapy is optimized and the response is assessed before CRT therapy is initiated.
The patient subgroups in which the efficacy of CRT remains uncertain are listed in Table 3. Although use of CRT in these subgroups might well prove to be beneficial, discriminating responders from nonresponders is difficult and currently uncertain.14 The benefits of CRT in patients at both ends of the spectrum of heart failure (ie, with either more or less severe symptoms) remain problematic. The uncertainties of extending indications into less severely ill populations (eg, “class creep”) are exemplified by a recent study of NYHA functional class II in which the primary end point of increased exercise capacity with CRT was not met, nor was there evidence of improved quality of life in these patients.12 Even more troublesome is the use of CRT as “bailout” therapy in patients with end-stage heart failure. Such patients were excluded from the clinical trials of CRT, and there is virtually no convincing evidence that they will improve at this late stage with resynchronization. Proceeding with CRT could expose these patients to an unnecessary procedure that often is taxing, compromises safety, and raises false hopes and expectations for both the patients and their families. Similarly, the concept of “drive-by” CRT during a planned ICD implant for sudden death prophylaxis remains unsupported. It is more prudent to place a programmable ICD set to a backup rate that discourages frequent pacing. Whether CRT is additive in benefit during an ICD implantation in those with demonstrable mechanical dyssynchrony or simply with a QRS lengthening remains unproved at this time.
Even when patients are selected for CRT on the basis of entry criteria used in the relevant clinical trials, it is clear that not all patients will respond to this form of therapy. In the pivotal Multicenter InSync Randomized Clinical Evaluation (MIRACLE) study, for instance, increases in distance covered during the 6-minute walk test averaged 39 m in the CRT group compared with 10 m in the control patients (P=0.005), a net increase of 29 m on average.15 Similarly, 68% of patients in the CRT arm of the MIRACLE study improved their NYHA functional class by at least 1 step compared with 38% of patients in the control group (P<0.001). Importantly, this study randomized patients only after a successful implantation; up to 10% of patients were not able to achieve placement of the CRT device. Although these differences are important, they point out that the changes in exercise capacity and symptomatic status that occur over the entire study population are of relatively small magnitude (particularly when changes in the control population are taken into account). When corrected for changes in the control population, for instance, only 30% of the CRT population experienced improvement in their NYHA functional class. These findings suggest that improved criteria for selection are needed to help define patients who are most likely to derive benefit. A variety of echocardiographic approaches, including determination of interventricular dyssynchrony by assessing the differences in timing between the onset of the QRS to peak aortic versus peak pulmonic flow, determination of intraventricular dyssynchrony by assessing septal to posterior wall motion delay, and the use of tissue Doppler measurements, have been used to measure the amount of dyssynchrony in patients.16 The effects of timing of the performance of these studies (eg, during active treatment for decompensation) and their reproducibility, however, have not been adequately defined. Consequently, despite the fact that they would appear to help define the patients who are most likely to benefit from CRT (particularly when the QRS duration is <150 ms), they are not routinely used in clinical practice.
Disadvantages of CRT
Although CRT is clearly not going to be helpful for all heart failure patients, it is important to review some of the reasons why caution is needed in expanding use of this therapy beyond the clinical populations in which it was studied. The most obvious considerations are cost, resources consumed, and the risks of the procedure. The costs include not only those associated with hospitalization and the device itself but also those that accrue during the follow-up of the patient. Implantation involves hospitalization in a monitored setting, followed by frequent clinic visits to assess the adequacy of wound healing and the appropriate functioning of the device. Although adverse events are relatively uncommon with CRT, serious complications can occur.8 These include bleeding (≈1%), infection (≈1%), hematoma (≈1%), pneumothorax (≈1%), pericardial effusion with/without cardiac tamponade (≈1%), myocardial infarction and/or stroke (≈0.2%), dissection or perforation of the coronary sinus (≈1%), lead dislodgement (≈5%), extracardiac stimulation (≈5%), and risks associated with the use of intravenous contrast. As seen in the patient whose history is described above, provocation of arrhythmias is another potential complication of CRT.17 There also are risks associated with the exposure to radiation during the procedure, particularly to the operator who often participates in multiple implantation procedures over an extended period of time.18
Discussion of the Clinical Case
The clinical history described above illustrates several aspects of the argument. The patient certainly met EF and QRS criteria for CRT. All issues outlined in Table 2 had been addressed, and he was receiving optimal medical management, including target doses of neurohormonal blocking agents. He had a very healthy lifestyle and diet and was compliant with the treatment regimen. Moreover, he demonstrated substantial LV enlargement as a result of adverse remodeling of the chamber, a factor known to be associated with increased risk for future morbidity and mortality. He was, however, only minimally symptomatic. In addition, he had atrial fibrillation, and the QRS duration was 130 ms. The database from clinical trials for the use of CRT in such patients is limited, and the likelihood of success is much less assured than in patients who are in sinus rhythm but have more advanced symptoms of heart failure. Similarly, given the QRS duration, evidence of dyssynchrony would have provided a firmer basis for recommending CRT. So, in this case, the likelihood of gain was uncertain (and almost certainly of lesser magnitude than for more standard indications), but both the costs and risks were the same as in patients more likely to benefit. Initiation of both atrial and ventricular arrhythmias has been associated with CRT, and this patient experienced an increase in life-threatening ventricular arrhythmias immediately after CRT placement. Although the occurrence of ventricular rhythm disturbances cannot be attributed to the placement of the LV pacing wire with absolute certainty, the temporal association is striking and highly suggestive that either alterations in electrical activity caused by initiation of LV pacing or mechanical irritation resulting from the pacing lead was responsible.18 In the end, the patient had a protracted medical course related to the procedure without evidence of tangible benefit from CRT.
It is clear that CRT is an important new modality of therapy that can benefit the clinical course of selected heart failure patients. Available clinical trial results indicate that patients with systolic dysfunction (ie, EF ≤0.35), NYHA functional class III symptoms, or ambulatory class IV symptoms despite optimal medical management and a prolonged QRS are the most likely to benefit. However, a prolonged QRS does not necessarily connote the presence of dyssynergy, and particularly for patients with a QRS between 120 and 150 ms, evidence of mechanical dyssynergy should be sought before CRT is recommended. Similarly, patients with either less or more severe heart failure symptoms would not in most cases be considered candidates for CRT on the basis of the database that is presently available. A point that cannot be overstated is that CRT should be considered only when a patient continues to fulfill EF, symptom, and QRS criteria for candidacy even after optimal medical management has been used. By carefully selecting patients, we can avoid the costs and risks of implantation in patients with a low likelihood of a favorable response. As more information from clinical trials, particularly those using various echo Doppler techniques to quantify dyssynchrony, becomes available, we anticipate that the use of CRT may well be shown to benefit a higher percentage of patients than we have reckoned. This would indeed be a welcome addition to our therapeutic arsenal, provided that physicians who are implanting in the community continue to adhere to the general guidelines for patient selection that have been outlined.
Dr Greenberg has received a research grant from Aventis; has been on the speakers’ bureau for GlaxoSmithKline, Merck, Pfizer, and Medtronic; has received honoraria from GlaxoSmithKline, Merck, Pfizer, and NitroMed; and has served on the advisory boards for GlaxoSmithKline and NitroMed. Dr Mehra has received research grants from Scios, Medtronic, Guidant, GlaxoSmithKline, Roche, and ORQIS; has served on the speakers’ bureau for and received honoraria from Scios, Medtronic, Guidant, GlaxoSmithKline, and AstraZeneca; and has served on the advisory boards for XDX and Scios.
Taylor AL, Ziesche S, Yancy C, Carson P, D’Agostino R Jr, Ferdinand K, Taylor M, Adams K, Sabolinski M, Worcel M, Cohn JN, for the African-American Heart Failure Trial Investigators. Combination of isosorbide dinitrate and hydralazine in blacks with heart failure. N Engl J Med. 2004; 351: 2049–2057.
Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, Tavazzi L, for the Cardiac Resynchronization-Heart Failure (CARE-HF) Study Investigators. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med. 2005; 52: 1539–1549.
Strickberger SA, Conti J, Daoud EG, Havranek E, Mehra MR, Pina IL, Young J, for the Council on Clinical Cardiology Subcommittee on Electrocardiography and Arrhythmias and the Quality of Care and Outcomes Research Interdisciplinary Working Group, Heart Rhythm Society. Patient selection for cardiac resynchronization therapy: from the Council on Clinical Cardiology Subcommittee on Electrocardiography and Arrhythmias and the Quality of Care and Outcomes Research Interdisciplinary Working Group, in collaboration with the Heart Rhythm Society. Circulation. 2005; 111: 2146–2150.
Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T, Carson P, DiCarlo L, DeMets D, White BG, DeVries DW, Feldman AM, for the Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) Investigators. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med. 2004; 350: 2140–2150.
Auricchio A, Yu CM. Beyond the measurement of QRS complex toward mechanical dyssynchrony: cardiac resynchronisation therapy in heart failure patients with a normal QRS duration. Heart. 2004; 90: 479–481.
Abraham WT, Young JB, Leon AR, Adler S, Bank AJ, Hall SA, Lieberman R, Liem LB, O’Connell JB, Schroeder JS, Wheelan KR, for the Multicenter InSync ICD II Study Group. Effects of cardiac resynchronization on disease progression in patients with left ventricular systolic dysfunction, an indication for an implantable cardioverter-defibrillator, and mildly symptomatic chronic heart failure. Circulation. 2004; 110: 2864–2868.
Castro PF, Mc-Nab P, Quintana JC, Bittner A, Greig D, Vergara I, Vukasovic JL, Corbalan R, Copaja M, Diaz-Araya G, Chiong M, Troncoso R, Alcaino H, Lavandero S. Effects of carvedilol upon intra- and interventricular synchrony in patients with chronic heart failure. Am J Cardiol. 2005; 96: 267–269.
Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, Kocovic DZ, Packer M, Clavell AL, Hayes DL, Ellestad M, Trupp RJ, Underwood J, Pickering F, Truex C, McAtee P, Messenger J, for the MIRACLE Study Group. Multicenter InSync Randomized Clinical Evaluation. Cardiac resynchronization in chronic heart failure. N Engl J Med. 2002; 346: 1845–1853.
Response to Greenberg and Mehra
William T. Abraham, MD
Drs Greenberg and Mehra and I agree that not all heart failure patients with ventricular dyssynchrony should receive cardiac resynchronization therapy (CRT). On the other hand, current evidence and contemporary heart failure guidelines support the use of CRT in all eligible patients defined by the inclusion and exclusion criteria of the major clinical trials. Those criteria have been emphasized in both articles. Where we disagree is the application of CRT within this indicated population. To date, no reliable predictors of response to CRT have emerged (including QRS morphology or duration in the range >120 ms). Likewise, no such predictors are known to forecast the treatment effect of heart failure drug therapies, yet we offer evidence-based, guideline-recommended drug therapies to all eligible patients unless contraindicated. In fact, physician adherence to the use of these drug therapies is used as a national metric for physician performance and quality of care in heart failure. In this regard, heart failure device therapies such as CRT should not be viewed as different from drug therapies, particularly when the evidence base is equally strong. Until data prove otherwise, it is unethical to exclude patients from CRT because of right bundle-branch block morphology or a QRS duration between 120 and 150 ms on the basis of underpowered post hoc subgroup analyses of clinical trials. Likewise, the belief that mechanical rather than ECG measures of dyssynchrony provide better patient selection criteria for CRT must be proved in an adequately powered prospective trial before the definition of ventricular dyssynchrony is changed. At present, there exists a moral obligation to offer this life-sustaining therapy to all eligible patients with a reasonable expectation for survival and likelihood of responsiveness to CRT.
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.