OUP user menu

Effect of cardiac resynchronization therapy and implantable cardioverter defibrillator on quality of life in patients with heart failure: a meta-analysis

Shaojie Chen, Yuehui Yin, Mitchell W. Krucoff
DOI: http://dx.doi.org/10.1093/europace/eus168 1602-1607 First published online: 28 October 2012


Aims To compare the improvement of quality of life (QoL) in cardiac resynchronization therapy plus implantable cardioverter defibrillator (CRT-D) therapy with that in implantable cardioverter defibrillator (ICD) therapy alone for patients with heart failure.

Methods and results Medline, Embase, The Cochrane Library, and US Food and Drug Administration website were searched for published studies up to 31 December 2011. Studies were considered for inclusion if they were randomized controlled trials (RCTs) that compared the efficacy of CRT-D therapy with ICD therapy in patients with heart failure. The primary outcome of this study was improvement in QoL. Four RCTs with 1655 patients were included in this meta-analysis. Overall, the QoL score of patients in CRT-D group significantly improved [weighted mean difference (WMD): −6.02; 95% confidence interval (CI): −10.56 to −1.48] compared with that in the ICD only group. However, the benefit with respect to QoL in CRT-D group was not maintained when subset meta-analysis was performed in patients with New York Heart Association (NYHA) class I–II (WMD: 0.19; 95% CI: −3.89 to 4.72), whereas the patients with NYHA class III–IV in CRT-D group still experienced a significant improvement of QoL score compared with ICD group (WMD: −8.49; 95% CI: −13.39 to −3.59).

Conclusions CRT-D therapy improves the QoL compared with ICD therapy alone, especially in patients with moderate to severe heart failure.

  • Heart failure
  • Cardiac resynchronization therapy
  • Quality of life


Heart failure is a common clinical syndrome affecting ∼ 2.5% adults in developed countries and the cost of heart failure exceeds $33 billion per year.1 It substantially impairs quality of life (QoL), increases the risk of sudden death and hospitalization rate which poses significant economic burden to the society.1 The outcome in patients with heart failure remains unsatisfactory despite optimal pharmacotherapy using medications such as renin–angiotensin–aldosterone system antagonist or β-blockers.2

Implantable cardioverter defibrillator (ICD) therapy has been proved to reduce mortality in patients at high risk of life-threatening arrhythmia and left ventricular systolic dysfunction.37 Cardiac resynchronization therapy (CRT) treatment, which paces both the right and left ventricles mechanically, improves haemodynamic efficiency and reduces mortality from heart failure.811 A previous meta-analysis published in 2007 involving 4420 participants in 14 clinical trials has demonstrated that CRT therapy is associated with a 22% reduction of risk in overall mortality and a 37% reduction in heart failure-related hospitalization when compared with standard medical therapy.12 Nowadays, according to the guidelines, patients with New York Heart Association (NYHA) class III or IV, left ventricular ejection fraction (LVEF) ≤35%, QRS complex duration ≥120 ms, and sinus rhythm are should receive CRT therapy and pharmacological treatment.13,14

However, it is becoming increasingly clear that, for many heart failure patients, QoL is at least as important as traditional endpoints such as mortality or hospitalization. Evidence whether this implantable device significantly improves or worsens the QoL in heart failure patients is scarce. In this meta-analysis, we systematically summarized updated evidence to determine the potential benefit in terms of QoL in CRT plus implantable cardioverter defibrillator (CRT-D) and ICD alone therapy in patients with heart failure.


Search strategy

This systematic meta-analysis included electronic searches in Medline, Embase, The Cochrane Library, and the US Food and Drug Administration website. Articles relevant to ‘CRT’ and ‘clinical trial’ were searched without restriction to language. The literature search was updated on 31 December 2011.

Inclusion and exclusion criteria

Studies were considered eligible if they met the following requirements: (i) randomized control clinical trials; (ii) enrolled patients with heart failure, (LVEF) ≤35%, QRS duration ≥120 ms; (iii) compared the efficacy of CRT-D therapy with ICD alone therapy; (iv) follow-up at least 3 months; and (v) included >30 participants. The exclusion criteria were: (i) equivocal study design or group allocation; (ii) CRT-D group and ICD group have significantly different baseline characteristics (regarding NYHA class, underlying disease, laboatory exam, etc.).

Quality assessment

Two investigators independently reviewed and evaluated the studies for this analysis according to the guideline of the PRISMA statement:15 (i) clear definition of study population? (ii) clear definition of outcomes and outcome assessment? (iii) independent assessment of outcome measurements? (iv) sufficient duration of follow-up? (v) no selective loss of follow-up? (vi) identification of important confounders and prognostic factors? All included studies that complied with these key points were considered as highly methodological quality studies. The discrepancy was resolved by consensus if necessary.

Statistical analysis

The difference between two groups was estimated by weighted mean difference (WMD) with a two-tailed 95% confidence interval (CI) in a fixed-effects model for consistent studies, or a random-effects model for heterogeneous studies. The WMDs and 95% CIs were estimated for continuous endpoints (improvement of QoL scores). Statistic value I2 was used to quantify the degree of inconsistency with a score of 25, 50, and 75% representing low, moderate, and high levels of inconsistency. P < 0.05 was considered statistically significant. The statistical analyses were performed by Review Manager Software (Version5.0, The Cochrane Collaboation, Oxford, UK).


Qualitative results

Study selection and baseline characteristics

A total of 3079 studies were identified by preliminary literature search (Figure 1). Of these studies, 3051 articles met exclusion criteria. Among the 28 remaining studies, 20 studies did not fulfil the inclusion criteria because 9 studies compared the efficacy of conventional pacing with biventricular pacing; 7 studies compared CRT with optimal medical treatment; 3 studies enrolled < 25 participants; 1 study enrolled narrow QRS complex patients. Consequently, 8 studies were considered for inclusion.1623 However, four of these studies were excluded because no data on QoL were available.1619 Therefore, four randomized controlled trials (RCTs) were finally included in this meta-analysis.2023 Table 1 shows the quality assessment of included studies according to the PRISMA statement.

View this table:
Table 1

Methodological quality of four included randomized controlled trials

Clear definition of study population?YesYesYesYes
Clear definition of outcomes and outcome assessment?YesYesYesYes
Independent assessment of outcomes?YesYesYesYes
Sufficient duration of follow-up?YesYesYesYes
No selective of loss during follow-up?YesYesYesYes
Prognostic factors and confounders identified?YesYesYesYes
Figure 1

Flow diagram of studies selection process.

The baseline characteristics of patients in included studies shown in Table 2. Of these studies, three studies were parallel-design RCTs,20,22,23 while one study was a crossover design RCT.21 Two of the four included studies predominantly included NYHA class III or IV patients,20,21 and the other two studies included NYHA class I or II patients.22,23 The overall population included in this study was 1655, of which 936 patients were allocated to CRT-D group and 719 to ICD group. The mean age ranged from 62 to 67 years, and patients were predominantly men. The mean follow-up duration ranged from 6 to 12 months. The mean QRS duration at baseline varied from 153 to 165 ms, and the mean LVEF at baseline ranged from 22 to 27%. None of the studies included patients with atrial fibrillation. Patients in each study received medications according to the current guidelines.

View this table:
Table 2

Baseline characteristics of four included randomized controlled trials

StudyStudy designSample sizeNYHA class (%)Mean follow-up duration (months)Mean age (SD)Male (%)Ischaemic cardio-myopathy (%)Mean ejection fraction (%)QRS width ms(SD)AF %RBBB %ACEI or ARB/β-blocker/ diuretic (%)
MIRACLE-ICD20RCT parallelCRT-D: 187; ICD: 182III, IV: 100667 (10)777024 (6.2)163 (22)01391/60/94
Contak-CD21RCT crossoverCRT-D: 245; ICD: 245II:33; III, IV: 67666 (11)846922 (7)158 (26)03288/47/86
MIRACLE-ICD II22RCT parallelCRT-D: 85; ICD: 101II: 100663 (12)895724.5 (6.7)165 (24)NR1696/64/83
REVERSE23RCT parallelCRT-D: 419; ICD: 191I, II: 1001262 (11)795527 (7)153 (22)0NR97/95/79
  • ACEI, angiotensin converting enzyme inhibitor; AF, atrial fibrillation; ARB, angiotensin receptor blocker; NYHA, New York Heart Association; RBBB, right bundle branch block; RCT, randomized controlled trial; SD, standard deviation.

Quantitative results

Improvement in quality of life

A summary of the pooled effect of CRT-D or ICD alone therapy on improvement in QoL based on Minnesota Living with Heart Failure Questionnaire (MLHFQ) is shown in Figure 2. Two studies that predominantly included NYHA class III or IV patients showed that patients in CRT-D group had a greater improvement in QoL than those in the ICD alone group,20,21 while two studies that included NYHA class I or II patients did not show in any difference between two groups.22,23 Overall, the improvement in QoL was significant in patients in the CRT-D group. The WMD for the QoL score in the CRT-D group was −6.02 points more than that in the ICD alone group (95% CI: −10.56 to −1.48; I2= 99%; the random-effect model was performed) (Figure 2A). However, when we included studies that enrolled patients with NYHA class I–II,2123 the CRT-D therapy did not show any appreciable improvement of QoL score compared with ICD therapy alone (WMD: 0.19; 95% CI: −3.89 to 4.27; I2= 84%; the random-effect model was performed) (Figure 2B). When we included studies that enrolled patients with NYHA class III–IV,20,21 patients in the CRT-D group experienced a significant improvement in the QoL score compared with the ICD group (WMD: −8.49; 95% CI: −13.39 to −3.59; I2= 99%; the random-effect model was performed) (Figure 2C).

Figure 2

Meta-analysis of included randomized controlled studies for quality of life; (A) meta-analysis of four randomized controlled studies enrolling patients with New York Heart Association class I–IV; (B) meta-analysis of three randomized controlled studies in patients with New York Heart Association class I–II; (C) meta-analysis of two randomized controlled studies in patients with New York Heart Association class III–IV.

Analysis of heterogeneity

When all the included studies were analysed using the fixed-effect model, a significant heterogeneity as a result of differences across studies was observed (χ2= 443.28; P < 0.001; I2= 99%). Two sources of significant heterogeneity were identified. The Contak-CD trial was different from other studies because it enrolled patients with NYHA III-IV heart failure in a crossover design. Another source of heterogeneity may be the enrolment of patients with different NYHA classes. Therefore, the removal of MIRACLE-ICD and Contak-CD studies could minimize the across-study heterogeneity (χ2= 0.06; P= 0.81; I2= 0%) (Figure 3).

Figure 3

Subgroup analysis for heterogeneity across four included randomized controlled studies.


Heart failure is characterized by decreased haemodynamic function, exercise capacity, and QoL due to poor left ventricular systolic or diastolic function, and patients with heart failure are at increased risk of sudden cardiac death. Implantable cardioverter defibrillator treatment in heart failure patients results in a reduction of all cause mortality,37 while CRT is a technique that corrects ventricular dyssynchrony, reverses cardiac enlargement, increases haemodynamic efficiency, and improves exercise capacity. For a decade, a number of randomized, controlled, and observational studies have demonstrated that CRT can improve clinical status and overall survival.811 Consequently, it was a reasonable conception to combine CRT with ICD therapy to improve both survival and QoL. Recently, an international, muticentre, prospective, randomized controlled study that enrolled 1798 patients with mild to moderate heart failure and mean follow-up of 40 months concluded that the overall mortality and hospitalization for heart failure were significantly reduced by CRT-D therapy compared with ICD therapy alone.19 Previous meta-analyses showed that CRT with optimal pharmacotherapy resulted in a substantial reduction of mortality in patients with moderate to severe heart failure compared with optimal medical therapy, and this effect was also accompanied by the potential improvement of echocardiographic parameters, QRS duration, and hospitalization.12 Recent large-size meta-analyses also demonstrated that CRT plus ICD and standard pharmacotherapy significantly reduced all cause mortality, or heart failure hospitalization compared with ICD plus standard pharmacotherapy.2426

On the other hand, heart failure is one of the chronic diseases that most affects QoL, the physical condition of patients is compromised by dyspnea, fatigue, and limitation of physical activity, and commonly by symptoms of the underlying causes of their condition, for example, angina. When dietary restrictions, difficulties in performing one's normal work, progressive loss of self-reliance, the side effects of medication, and recurrent hospitalization are also taken into account, it is easy to understand why QoL of these patients can be poor. For this reason, more and more clinical studies have taken QoL on the agenda as an important outcome variable. The MLHFQ contains 21 questions designed to determine how heart failure affects the physical, psychological, and socioeconomic condition of patients. It provides an individualized picture of a number of limiting situations associated with the cardiac syndrome. The questionnaire refers to the signs and symptoms of heart failure, social relationships, physical and sexual activity, work, and emotions. The answer for each question is chosen from a scale of 0 (none) to 5 (very much), the greater the score, the worse the QoL. To date, the MLHFQ has been a specific questionnaire developed for heart failure patients, and it is now the most popular tool commonly used in clinical trials where QoL is a primary or secondary outcome.27

Four randomized controlled studies compared QoL with CRT-D and ICD therapy. Some studies reported a significant improvement of QoL in CRT-D group, while others reported a similar benefit in both groups. Although we could not obtain the individual data of these clinical trials, the possible reasons for that may be the intrinsic differences in study design and patient enrolment. In this meta-analysis, a significant heterogeneity was observed when we pooled data on QoL, which was attributed to the different symptom status at baseline. Patients with NYHA class I or II symptoms included in 2 studies had a better MLHFQ score at baseline and did not show any significant improvement with CRT-D therapy.22,23 The other two studies in patients with NYHA class III or IV symptoms and a poor MLHFQ score at baseline showed a significant improvement in QoL with CRT-D.20,21 Thus, patients with moderate to severe heart failure symptoms could obtain maximal benefit in terms of improvement of QoL from CRT-D therapy, while the improvement of QoL may be small in patients with minimal to mild heart failure symptoms due to better QoL at baseline.

The results of this study was an important complement to the previous studies testing the effects of CRT therapy in patients with heart failure. Although a number of these studies showed a benefit of survival in CRT-D therapy, several questions were left unanswered. Because most patients included in the previous studies had different NYHA class, or the studies included in the meta-analysis had intrinsically different study design, the effect of CRT in patients with less severe heart failure symptoms remained uncertain. More than 60% patients in this meta-analysis were in the diagnosed as having ischaemic cardiomyopathy, and the majority of patients with indications to CRT therapy in these studies were at high risk of sudden death, for whom CRT plus ICD or ICD only would be appropriate. Implantable cardioverter defibrillator has been proven to reduce mortality in patients with high risk of sudden death; however, but few studies specifically demonstrated the practical improvement of QoL in a long-term follow-up. The present meta-analysis which combined the updated data might renew some of these concerns, supporting CRT plus ICD in the context of optimal pharmacotherapy for patients with heart failure, especially for the patients in moderate or severe NYHA category.

Nonetheless, the sample size of each study included in this meta-analysis is relatively small, and the results of this study may also be underpowered due to a small number of studies included in this meta-analysis. Individual patient data from the clinical trials would be needed to further examine this issue. In summary, this meta-analysis suggests that CRT-D therapy improves QoL in patients with heart failure compared with ICD therapy alone in the context of optimal medical management, and patients with NYHA class III–IV may have a maximal benefit.


We would like to acknowledge all the investigators in clinical trials that we included.

Conflict of interest: none declared.


View Abstract