OUP user menu

Cardiac resynchronization therapy in mild heart failure

Cecilia Linde
DOI: http://dx.doi.org/10.1093/europace/eup307 v72-v76 First published online: 27 October 2009

Abstract

It has been firmly established that cardiac resynchronization therapy (CRT) reduces symptoms and improves mortality in patients with moderate-to-severe chronic heart failure [New York Heart Association (NYHA) class III–IV], despite optimal heart failure medication and with wide QRS complex on the surface electrocardiogram as evidence of ventricular dyssynchrony, but not whether such treatment is efficacious in mildly symptomatic heart failure patients. In such patients, the treatment goal is to prevent disease progression rather than to improve symptoms. The REVERSE trial was the first randomized, controlled study of CRT in NYHA I–II patients. Cardiac resynchronization therapy in this study induced substantial reverse remodelling over 12–18–24 months of follow-up and was linked to a significant delay in the time to first heart failure hospitalization and eventually in the time to the combined endpoint of time to first heart failure hospitalization or death. The MADIT CRT designed as a morbidity–mortality study corroborated these findings with a significant reduction in heart failure events and significant reverse remodelling. These findings most likely will translate into a wider use of CRT in mildly symptomatic patients to prevent disease progression.

  • Cardiac resynchronization therapy
  • Heart failure
  • Randomized controlled trial
  • Biventricular stimulation
  • Reverse cardiac remodelling

Introduction

It has been firmly established that cardiac resynchronization therapy (CRT) reduces symptoms and improves mortality in patients with moderate to severe chronic heart failure (HF) [New York Heart Association (NYHA) class III–IV] despite optimal HF medication and with wide QRS complex on the surface electrocardiogram as evidence of ventricular dyssynchrony. It has also been consistently shown that CRT induces progressive reverse left ventricular (LV) remodelling and halts disease progression in such patients.

It is reasonable to assume that CRT may be beneficial in less severe HF conditions. This assumption is currently being investigated in large randomized studies such as REVERSE, MADIT-CRT, and RAFT.

Effects of cardiac resynchronization therapy in patients with mild heart failure or asymptomatic left ventricular dysfunction

There has been little evidence for cardiac resynchronization therapy (CRT) in NYHA class I–II patients. Smaller randomized, open studies have shown that CRT reversed LV remodelling in patients with NYHA class II HF over a 6-month period compared with control therapy.1,2 Until recently, only one CRT study focused solely on patients with NYHA class II HF—the MIRACLE ICD II trial.2 In this study, all patients had a secondary preventive indication for an implantable cardioverter defibrillator (ICD). A significant improvement in the clinical HF composite endpoint3 was achieved in patients assigned to 6 months of CRT and CRT induced reverse LV remodelling compared with control patients. In contrast, no improvements in quality of life and NYHA class were observed. Likewise, in the CONTAK CD study,1 no significant symptomatic improvement in the CRT compared with the control group was achieved in a subgroup of NYHA class I–II patients. Still, CRT was linked to a significant reduction in LV volumes. In a smaller, open study, LV function significantly improved following CRT whereas NYHA class remained unchanged in most patients.4 This finding illustrates that the main goal in mild HF and asymptomatic HF with previous symptoms cannot be to improve symptoms but rather to prevent or delay disease progression or even to revert it. Reverse remodelling has only recently been recognized as a goal of treatment.5 This is the reason why reverse LV remodelling is an important secondary endpoint in the REVERSE trial and in the ongoing MADIT-CRT trial.

The REVERSE study

The REVERSE study was designed to evaluate the hypothesis that CRT in combination with optimal medical treatment can prevent disease progression and reverse LV remodelling in patients with asymptomatic LV dysfunction with prior HF symptoms (American College of Cardiology/American Heart Association stage C, NYHA class I)6 or in patients with mild systolic HF (NYHA class II).7 Further inclusion criteria were LV end-diastolic diameter >55 mm, LV ejection fraction (LVEF) <40%, sinus rhythm, and QRS duration >120 ms. The primary endpoint was the clinical composite endpoint developed by Packer3 to assess HF patients in clinical trials.

The secondary and powered endpoint was the LV end-systolic volume index (LVESVi, cm/m2 body surface area). Endpoints were evaluated after 12 months of treatment assignment. Left ventricular end-systolic volume index was selected as the major measure of LV remodelling because a clear relationship between improvement in LVESVi and clinical outcome has been demonstrated.810 It was expected that optimal medical therapy plus CRT would result in a greater reduction of LVESVi than optimal medical therapy alone. The study was a double-blind international parallel study comparing CRT-ON with or without an ICD to CRT-OFF with or without an ICD. The trial was conducted between 2004 and 2006 in Europe, the USA, and Canada and included 610 patients at 73 centres. All patients had optimal medical therapy according to the current HF guidelines at study entry.7,11 Over the randomized follow-up period, all evaluations were performed in a double-blind fashion meaning that the electrophysiologist took care of the CRT follow-up, whereas the HF specialist and the echocardiographist arranged the clinical and echocardiographic assessments. The blinded follow-up was 12 months in the USA and Canada. In Europe (partly due to the European Health Economics study), the blinded phase was 24 months.7 Hospitalization for worsening HF was evaluated in a prospective secondary analysis of health-care utilization.

A total of 610 patients with NYHA class I–II HF were randomly assigned to active CRT (CRT-ON; n = 419) or control (CRT-OFF; n = 191) (Table 1). The average age of the patients was ∼60 years, and 83% (CRT-OFF) and 82% (CRT-ON) were in NYHA class II. In spite of excellent HF medication, LV function was indeed compromised with a mean LVEF of 26% and an average LV end-diastolic diameter of 7 and 6.9 cm in the CRT-OFF and CRT-ON groups, respectively. Mean QRS width was on average 154 and 153 ms in the CRT-OFF and CRT-ON groups, respectively, which is less than in previously reported trials. Ischaemic heart disease was present in 51% (CRT-OFF) and 56% (CRT-ON) of the patients. A concomitant ICD was provided in 85% (CRT-OFF) and 82% (CRT-ON) of the patients. There were no statistically significant differences in baseline characteristics between the two randomization arms.

View this table:
Table 1

Patient characteristics in the REVERSE trial (reprinted with permission from Linde et al.12)

ParameterCRT-OFF (n = 191)CRT-ON (n = 419)P-value
Age (years)61.8 ± 11.662.9 ± 10.60.26
Gender (male)152 (80%)327 (78%)0.75
NYHA classification (class II)159 (83%)344 (82%)0.82
Ischaemic97 (51%)236 (56%)0.22
Diabetic46 (24%)91 (22%)0.53
ACE-inhibitors151 (79%)330 (79%)1.00
ARBs39 (20%)88 (21%)0.91
ACE-inhibitors or ARBs186 (97%)404 (96%)0.63
Beta-blockers179 (94%)401 (96%)0.32
At least 50% of target dose of beta-blockers114 (60%)255 (61%)0.79
100% of target dose of beta-blockers62 (32%)153 (37%)0.36
Diuretics148 (77%)339 (81%)0.33
Intrinsic QRS width (ms)154 ± 24153 ± 210.41
LV ejection fraction (%)26.4 ± 7.126.8 ± 7.00.50
LV end-diastolic dimension (cm)7.0 ± 0.96.9 ± 0.90.65
LV end-systolic dimension (cm)5.8 ± 1.15.7 ± 1.00.63
Left ventricular end-systolic volume (cm3)203 ± 91197 ± 740.42
Left ventricular end-diastolic volume (cm3)272 ± 102268 ± 890.60
LV mass (g)274 ± 78269 ± 760.57
Interventricular mechanical delay (ms)33.6 ± 36.433.9 ± 40.10.94
Glomerular filtration rate (mL/min)89.6 ± 36.484.2 ± 31.30.08
Heart rate (bpm)68.2 ± 10.966.9 ± 10.30.17
Supine systolic blood pressure (mmHg)123.4 ± 19.0125.2 ± 18.60.25
Supine diastolic blood pressure (mmHg)72.0 ± 12.072.2 ± 10.80.82
Weight (kg)87.5 ± 18.885.1 ± 17.80.12
Minnesota Living With Heart Failure Score28.8 ± 21.727.0 ± 20.10.32
Kansas City Cardiomyopathy Questionnaire Overall Summary Score70.7 ± 22.173.7 ± 19.00.14
Six-minute hall walk (m)388 ± 132399 ± 1250.34
CRT-ICD implanted163 (85%)345 (82%)0.41
  • P-values are two-sided and were calculated using Fisher's exact test or the two-sample t-test.

The HF clinical composite response endpoint showed that 16% of the patients worsened in the CRT-ON group when compared with 21% in the CRT-OFF group (P = 0.10)12 (Figure 1), indicating that the REVERSE failed to meet its primary endpoint. However, the powered secondary endpoint was highly significant in favour of the CRT-ON group. Patients assigned to CRT-ON showed a larger reduction in LVESVi (−18.4 ± 29.5 mL/m2 in the CRT-ON group vs. −1.3 ± 23.4 mL/m2 in the CRT-OFF group, P < 0.0001); in addition, significant changes in other measures of LV reverse remodelling were shown in the CRT-ON group, including a reduction in LV end-diastolic volume index (LVEDVi) and an increase in LVEF (Figure 2).

Figure 1

The HF clinical composite response. The primary endpoint, comparing the proportion of worsened subjects at 12 months of CRT, was not different between CRT-ON and CRT-OFF groups (P = 0.10). Reprinted with permission from Linde et al.12

Figure 2

Mean LVESVi, LVEDVi, and LVEF at baseline and 12-month follow-up in the CRT-OFF and CRT-ON groups. A significant reduction in volume indexes and LVEF was observed in the CRT-ON group when compared with the CRT-OFF group. Error bars represent 95% confidence intervals. LVESVi, left ventricular end-systolic volume index; LVEDVi, left ventricular end-diastolic volume index; LVEF, left ventricular ejection fraction. Reprinted with permission from Linde et al.12

The change in the 6 min walk distance at 12 months was 12.7 ± 102.4 m in the CRT-ON group and 18.7 ± 105.2 m in the CRT-OFF group (P = 0.52). Also, the quality of life score dropped to a similar degree in both groups (−8.4 ± 17.1 in the CRT-ON group vs. −6.7 ± 15.9 in the CRT-OFF group, P = 0.26). These observations underscore that in patients with mild HF, not much clinical improvement can be expected, since the symptoms before CRT are already minimal. Moreover, the fact that some change in symptoms occurred in the CRT-OFF group is also not surprising, since a placebo effect of CRT (mainly affecting symptoms but not affecting LV size and function) has been reported previously.

Importantly, the effect on LV function and size is highly significant, and only in the CRT-ON group. Furthermore, the significant reverse remodelling was linked to improved HF morbidity.

Time to first HF hospitalization was significantly delayed in the CRT-ON group [hazard ratio (HR) 0.47, P = 0.03] (Figure 3). This translated into a 53% relative risk reduction for first HF-related hospitalization in patients undergoing CRT.

Figure 3

Time to first HF hospitalization in the first 12 months in the CRT-OFF and CRT-ON groups. Heart failure relatedness of hospitalizations was adjudicated by the adverse events advisory committee. The number at risk at 12-month follow-up represents the number of patients who had a randomized follow-up after 365 days and did not experience HF hospitalization. The number of patients drops significantly at 12-month follow-up because the patient cohort in the USA had a maximum follow-up of 12 months only. Reprinted with permission from Linde et al.12

The 18-month results of the REVERSE trial indicate that the clinical benefits from CRT are sustained over time.13 There was a significant reduction in time to first HF-related hospitalization between the CRT-ON group (5.5%) and the CRT-OFF group (3.5%, HR 0.42, P = 0.0005). Similarly, there was a significant difference in the time to first HF hospitalization or death between the CRT-ON group (7.5%) and the CRT-OFF group (15.7%, HR 0.50, P = 0.01)13 (Figure 4). The 24-month results of the European cohort of REVERSE indicate sustained benefit over time with a significant delay in time to first HF hospitalization (HR 0.39, P = 0.01) and to first HF hospitalization or any death (HR 0.38, P = 0.003) in the CRT-ON group.14

Figure 4

Time to first HF hospitalization or death in the 18-month follow-up period in the CRT-OFF and CRT-ON groups. Reprinted with permission from Abraham et al.13

Even though the REVERSE trial was not proposed to demonstrate differences in morbidity and mortality, the significant reverse remodelling linked to reduced HF morbidity indicates that CRT may potentially modify disease progression in mild HF patients.

The MADIT-CRT and RAFT trials

The MADIT-CRT trial studied the effects of CRT in patients with NYHA class I–II HF.15 The primary objective was to determine whether CRT with an ICD in high-risk, relatively asymptomatic subjects with ischaemic and non-ischaemic cardiomyopathies can significantly reduce all-cause mortality or HF events compared with ICD therapy alone over a follow-up period of at least 24 months. The secondary endpoint was the change in LVESV from baseline to 1-year follow-up. The inclusion criteria in MADIT-CRT study are similar as in the REVERSE trial. A difference however is that patients in MADIT-CRT needed to have an LVEF < 30%. Patients with ischaemic cardiomyopathy had to be in NYHA class I and II and those with non-ischaemic cardiomyopathy in NYHA class II.

The results were reported at the Hot-Line session at ESC in Barcelona on 1 September and were simultaneously available online at the New England Journal of Medicine. With a mean follow-up of 2.4 years, the primary endpoint occurred in 187/1089 patients in the CRT-ICD group (17.2%) and 185/731 patients in the ICD only group (25.3%) (HR 0.66, P = 0.001). The benefit was driven by a 41% reduction in the risk for HF events with an equal benefit in patients with ischaemic or non-ischaemic aetiology but with the benefit predominantly in patients with a baseline QRS duration of at least 150 ms. Analysing the Kaplan–Meier estimate of HF survival probability, there was an early diverging of curves in favour of CRT-ICD already after 2 months (Figure 5). The annual mortality rate (3%) was equally low in both randomization arms. Moreover, significant reverse LV remodelling was found at 1 year with a mean increment in LVEF of 11%, a drop of mean LVEDV of 52 mL and in LVESV of 57 mL.

Figure 5

Kaplan–Meier estimates of the probability of survival free of HF in the MADIT-CRT trial with an average follow-up time of 2.4 years. Reprinted with permission from Moss et al.16 Copyright © 2009 Massachusetts Medical Society. All rights reserved.

Finally, the RAFT trial will assess whether the addition of CRT to ICD therapy and optimal medical therapy may reduce the combined endpoint of all-cause mortality and HF-related hospitalizations in patients with NYHA class II–III HF, LVEF < 30%, QRS duration >120 ms, normal sinus rhythm, or atrial fibrillation with spontaneous rhythm <60 bpm or in those with faster intrinsic rhythm combined with atrio-ventricular node ablation.17 The study will enrol 1800 patients and is expected to be completed in 2010.

Conclusions

The goal in treatment of patients with mild HF (NYHA class II) or in asymptomatic HF patients with previous symptoms (NYHA class I) is to prevent disease progression. The REVERSE trial was the first randomized, controlled study and indicated that substantial reverse remodelling is achieved by CRT over 12–18–24-month follow-up which in turn was linked to a significant delay in time to first HF hospitalization and eventually in the time to the combined endpoint of time to first HF hospitalization or death. The MADIT-CRT designed as a morbidity mortality study corroborated these findings with a significant reduction in HF events and significant reverse remodelling. These findings most likely will translate into a wider use of CRT in mildly symptomatic patients to prevent disease progression in such patients.

Conflict of interest: C.L. receives research grants, speaker's honoraria and consulting fees from Medtronic Inc., and speaker's honoraria and consulting fees from St Jude Medical.

References

View Abstract