Cardiac Resynchronization Therapy for Mild Congestive Heart Failure
Executive Summary
Background
Congestive heart failure (CHF) is common, and rapidly increasing in incidence. CHF carries a poor prognosis, with an estimated 30–50% 1-year mortality for patients with advanced disease. It is also associated with a high burden of illness, high resource utilization, and frequent hospitalizations. The current treatment for CHF involves addressing the underlying cause(s), lifestyle modifications, and pharmacologic interventions. In the majority of cases, treatment is not curative, but intended to ameliorate symptoms and improve function.
Approximately 20–30% of patients with CHF exhibit dyssynchronous contractions of the left and right ventricles due to conduction system disease. Dyssynchrony further depresses the already impaired pumping ability of the heart. Cardiac resynchronization therapy (CRT) is intended to correct dyssynchronous ventricular contractions. CRT uses biventricular pacing to simultaneously stimulate both ventricles in order to achieve coordinated contractions.
CRT therapy has demonstrated benefit in class III and class IV CHF. A systematic review of 9 randomized, controlled trials of CRT for class III/IV CHF concluded that CRT reduced mortality, improved quality of life, and improved functional status. Much of the focus of new research in CRT is to evaluate whether the benefits of CRT extend to patients with less severe heart failure.
Objective
To determine whether cardiac resynchronization therapy improves health outcomes for patients with mild congestive heart failure, defined as New York Heart Association (NYHA) class I or II CHF.
Search Strategy
Electronic search of MEDLINE® (via PubMed) was performed using the keywords "CRT," "resynchronization," and "biventricular pacing." These terms were cross-referenced with "CHF," "congestive heart failure," and "cardiomyopathy." Search was performed from January 1995 through December 2009. Electronic search was supplemented with a hand search of relevant bibliographies and use of the "related articles" function in MEDLINE®.
Selection Criteria
Studies were selected for inclusion that had the following characteristics: 1) randomized, controlled trial; 2) included patients with NYHA class I or II CHF, or included a broader population of CHF patients and reported outcomes separately for the group with class I/II CHF; 3) enrolled at least 25 patients per treatment group; and 4) reported on at least one relevant health outcome.
Main Results
A total of 3 randomized, controlled trials enrolling 2,616 patients met the inclusion criteria, with follow-up ranging from 6 months to 2.4 years. The largest trial published to date was the MADIT-CRT trial, a single-blind trial that randomized 1,820 patients with class I/II CHF to an ICD alone or an ICD-CRT device. The MADIT-CRT trial reported a reduction for the ICD-CRT group on the primary outcome, i.e., death or acute heart failure exacerbation. The primary endpoint was reached by 17.2% of patients in the ICD-CRT group compared to 25.3% of patients in the ICD-alone group. This composite outcome is suboptimal for several reasons. First, death and hospitalizations represent fundamentally different outcome measures and therefore do not lend themselves to combination. Second, the outcomes occur at different rates, with hospitalizations much more frequent. This makes it likely that the results on hospitalizations will drive the overall results. Finally, the relative risks for these outcomes are not similar, with a large reduction in relative risk for hospitalizations, and no reduction for death.
As a result, it is preferable to examine the results on the individual outcome measures rather than rely on the composite outcome. The first component of the composite outcome, acute heart failure events, occurred in 22.8% of patients in the ICD-alone group compared with 13.9% of patients in the ICD-CRT group and (relative risk reduction [RRR] 39%, absolute risk reduction [ARR] 8.9%, number needed to treat [NNT] =11.2). This difference in acute heart failure events accounted entirely for the difference on the primary composite outcome. The death rate was similar between groups.
The REVERSE trial enrolled a total of 610 patients, all of whom received a CRT device. Patients were randomized to CRT-ON or CRT-OFF for a period of 12 months in double-blind fashion. The primary outcome was a composite measure that classified patients as improved, unchanged, or worse. There were no significant differences reported on this primary outcome. There was a decrease in hospitalizations for heart failure in the CRT-ON group (4.1%, 17/419) compared with the CRT-OFF group (7.9%, 15/191). Changes in functional status, as measured by the 6-minute walk, were similar between groups. Quality of life (QoL), as measured by the Minnesota Living with Heart Failure Questionnaire, was also similar between groups.
The MIRACLE ICD study was the smallest of the 3 studies, enrolling 186 patients with class II CHF and an indication for an ICD in an unblinded fashion. Patients were randomized to ICD/CRT-ON versus ICD/CRT-OFF and followed for 6 months. There was no difference in the primary outcome of peak oxygen uptake between groups. There were also no differences reported between groups on the secondary outcomes of functional status as measured by the 6-minute walk, QoL, as measured by the Minnesota Living with Heart Failure Questionnaire, and NYHA CHF class.
All 3 randomized, controlled trials reported significant improvements in echocardiographic measures of left-ventricular (LV) pump function. LV ejection fraction improved more in the CRT group in each trial, with a range of improvement of 3.0–11.0%, compared with the control group. There were also substantial improvements in LV end-systolic and end-diastolic volumes (LVESV, LVEDV) in all 3 trials. All 3 trials reported relatively large improvements in the LVESV and the LVEDV in favor of the CRT group.
Complications in these trials were not uniformly reported; however, each trial contained some information on short- and long-term complications. Short-term complication rates ranged from 4–22%, with lead dislodgement and hematoma at the access site most common. Long-term complications were reported by 2 of the trials, with rates of 16% and 35%. The majority of these long-term complications were lead dislodgement.
The MADIT-CRT trial provides data on a limited set of complications of a combined device versus an ICD alone. There were more complications reported for the combined device compared to ICD alone for pneumothorax (1.7% vs. 0.8%), infection (1.1% vs. 0.8%), hematoma requiring evacuation (3.3% vs. 2.5%), coronary venous dissection (0.5% vs. 0.0%), and LV lead dislodgement (4.0% vs. 0%).
Author's Conclusions and Comments
The available evidence reports benefits on some outcomes, but not on others. As a result, the most challenging analytic aspect of evaluating these data is considering the clinical importance of the different outcomes, and determining whether differences in the subset of outcomes that report benefit represent adequate evidence for improvement in health outcomes when weighed against the risks of the procedure.
The most important outcomes for this treatment are mortality from CHF, progression to more advanced disease, functional status, and quality of life. None of these outcomes showed differences in any of the 3 available trials. In the 2 trials reporting mortality outcomes, one showed a slightly lower rate for the CRT group, while the other showed a slightly lower rate for the control group. For the outcomes of functional status and quality of life, the 2 trials including these outcomes did not report any group differences. Therefore, it can be concluded with a moderately high degree of certainty that CRT in patients with mild CHF does not lead to improvements in mortality, quality of life, or functional status over the short to medium term.
The outcome measures that did show improvement were hospitalizations (or acute "CHF events" in the MADIT-CRT trial) and echocardiographic measures of cardiac morphology and function. Hospitalizations for CHF are an important outcome measure, as a reduction in hospitalizations would be of benefit for the individual patient. Reducing hospitalizations will also prevent the iatrogenic complications associated with hospitalization.
However, for several reasons, this evidence is not definitive in determining whether CRT leads to a health outcome benefit. Hospitalizations, or acute heart failure events, are the most subjective of the outcomes reported in these trials. Hospitalization involves a decision by a treating clinician that involves a substantial degree of judgment. These decisions can be influenced by a number of factors and may not be solely the result of exacerbation of disease. Thresholds for admission to the hospital may vary substantially by individual clinicians and/or geographic regions. As a result, the lack of blinding of clinicians in 2 of the 3 trials represents a potential bias in this outcome measure. This leaves only 1 trial, the REVERSE trial, which reports a difference in hospitalizations that is not prone to bias.
Even if the reported difference in hospitalizations is real, this may not represent a large effect, and the benefit may not outweigh the risks. Using the results reported in REVERSE, there is a relative risk reduction of 48% and an absolute risk reduction of 3.8% for CHF hospitalizations. This translates to a number needed to treat of 26 patients over a period of 1 year to prevent 1 hospitalization. This relatively small benefit in hospitalizations needs to be weighed against the risks of the procedure and the adverse effects of having a CRT device implanted long-term. While the risks of the procedure are uncommon, some may be serious and exceed the benefit of reduced hospitalizations. Minor adverse events, such as lead dislodgement, are more common and may involve some degree of morbidity and repeat procedures.
In the 2 trials that report rates of lead dislodgement, the MIRACLE trial reported a rate of 5.8% over a 6 month period and the REVERSE trial reported a rate of 10.6% over a 1-year period. This would translate roughly to 1 in 10 patients experiencing lead dislodgement over a 1-year period, which is equivalent to a number needed to harm of approximately 10. Thus it appears more likely that a patient will develop lead dislodgement, or another long-term complication, than would prevent a hospitalization.
For patients with indications for an ICD, a combined ICD/CRT device is often used. In this situation, the additional risk of CRT implantation compared to ICD alone is the proper comparison to determine the risks of CRT, and the risk/benefit ratio is shifted more favorably toward CRT use. However, the evidence is not sufficient to estimate the precise rates of incremental complications of a combined device compared with an ICD alone.
The echocardiographic outcomes reported in these trials show consistent, large improvements associated with CRT therapy. However, the clinical importance of these intermediate outcomes is uncertain. While LVEF and other echocardiographic parameters do correlate with mortality in CHF, this correlation has not been shown for patients with a CRT device. It is possible that CRT induces changes in these parameters when measured on echocardiography, but that they do not translate to physiologic improvements.
Finally, if the CRT device is actually leading to better pump function of the heart, this should be evident in other measures of quality of life and functional status. Since none of the available studies report any differences in functional status or quality of life, there is further concern that the improvements in the echocardiographic measures may not be translating into real improvements in health outcomes.
Based on the available evidence, the Blue Cross and Blue Shield Association Medical Advisory Panel made the following judgments about whether the use of cardiac resynchronization therapy for class I/II congestive heart failure meets the Blue Cross and Blue Shield Association's Technology Evaluation Center (TEC) criteria.
1. The technology must have final approval from the appropriate government regulatory bodies.
U.S. Food and Drug Administration (FDA) indications are limited to patients with class III/IV failure, none of the approved devices currently available have indications for treatment of patients with class I and/or II CHF. Use in mild heart failure, therefore, meets this criterion as an off-label use of an approved device.
One stand-alone biventricular pacemaker (InSync® Biventricular Pacing System, Medtronic) has received approval by the FDA for the treatment of patients with New York Heart Association (NYHA) class III or IV heart failure, on a stable pharmacologic regimen, who also have a QRS duration of ≥130 msec and a left ventricular ejection fraction of ≤35%. Biventricular pacemakers have also been combined with implantable cardiac defibrillators (ICDs). Both Guidant (CONTAK CD® CRT-D System) and Medtronic (InSync® ICD Model 7272) have received FDA approval for combined cardiac resynchronization therapy defibrillators for patients at high risk of sudden cardiac death due to ventricular arrhythmias and who have NYHA Class III or IV heart failure with left ventricular ejection fraction of 35% or less, QRS duration ≥130 msec (≥120 msec for the Guidant device) and remain symptomatic despite a stable, optimal heart failure drug therapy.
At the time this Assessment went to press, the FDA Circulatory System Devices Advisory Panel voted unanimously to recommend approval of CRT devices for use in mild heart failure. The indications proposed by the FDA Advisory Panel include patients in NYHA functional class II or in patients with class I ischemic heart failure with an LVEF <30% and a QRS duration >130 ms. Also added was a requirement that eligible patients also have left-bundle-branch block (LBBB). Note that recommendation of approval does not constitute final approval.
2. The scientific evidence must permit conclusions concerning the effect of the technology on health outcomes.
The evidence is sufficient to permit conclusions concerning the effect of CRT on mortality, functional status and quality of life. For each of these 3 outcome measures, at least 2 of the 3 randomized, controlled trials reported on this outcome. For each outcome, there were no group differences, and there was no apparent trend toward improvement in the CRT group. Therefore, conclusions on these outcome measures can be made over the period of time covered by the study.
The evidence is also sufficient to determine the effect of CRT therapy on echocardiographic parameters while the device is on. The evidence from the included studies is consistent in reporting an improvement in LVEF and LV volumes over the first year of therapy with the CRT device continuously on. The evidence is not sufficient to determine whether these changes represent structural changes in the heart that would persist in absence of the CRT device turned on.
The evidence is not sufficient to permit conclusions on the effect of CRT on hospitalizations. Although this outcome was reported by 2 trials, it is a more subjective outcome that can be influenced by knowledge of group assignment. The MADIT-CRT trial was the largest trial and was single blinded. As a result, there is potential for bias on the outcome of hospitalizations, leaving only one trial that was double blinded and thus avoided this potential bias.
3. The technology must improve the net health outcome.
For the outcomes of mortality, functional status, and quality of life, the evidence does not support the conclusion that the net health outcome is improved. For these outcomes, there were no improvements associated with CRT therapy. Therefore, it can be concluded with a moderately high degree of certainty that there is not improvement in these outcomes over the 1- to 2-year time period covered by these studies.
For the outcome of hospitalizations, the evidence is not sufficient to permit conclusions. For the echocardiographic outcomes, the evidence is not sufficient to conclude that the net health outcome is improved. This is because it is not certain that these changes in cardiac morphology and function translate to physiologic benefits that can be experienced by the patient.
4. The technology must be as beneficial as any established alternatives.
The evidence is not sufficient to determine whether the net health outcome is improved, therefore it cannot be determined whether the technology is as beneficial as alternatives.
5. The improvement must be attainable outside the investigational settings.
Whether CRT for mild heart failure improves health outcomes has not been demonstrated in the investigational setting.
For the above reasons, the use of cardiac resynchronization therapy for class I/II congestive heart failure does not meet the TEC criteria.
Full Study
Cardiac Resynchronization Therapy for Mid Congestive Heart Failure
Full studies are in PDF format. You will need Adobe Acrobat Reader to view all studies.
Download Adobe Acrobat Reader here.
arrhythmia; biventricular; blinding; cardiomyopathy; CARE-HF; CHF; chronic heart failure; COMPANION; congestive heart failure; CONTAK; CRT; CRT-OFF; CRT-ON; diastolic; dysfunction; dyssynchronous; dyssynchrony; echocardiographic; echocardiography; ejection fraction; end-diastolic; end-systolic; hospitalization; ICD; ICD-CRT; implantable cardioverter-defibrillator; InSync; intraventricular; lead dislodgement; left-ventricular; LV; MADIT-CRT; malignant; Minnesota Living with Heart Failure; MIRACLE; mortality; New York Heart Association; NNH; NNT; number needed to harm; number needed to treat; NYHA; pacing; QoL; QRS; quality of life; resync; resynch; REVERSE; systolic; ventricle; volume;
