This review assessed whether cardiac resynchronization (a type of pacemaker-based therapy) reduces mortality from progressive heart failure. The authors found that mortality during 3 to 6 months of follow-up was significantly reduced among patients who received cardiac resynchronization compared with controls. This was a well-conducted and clearly reported review.

To determine whether cardiac resynchronisation decreases mortality from progressive heart failure.

MEDLINE (from 1966 to 2002), EMBASE (from 1980 to 2002, week 24), the Cochrane Controlled Trials Register (Issue 2, 2002), the ClinicalTrials.gov database of the National Institutes of Health, and the U.S. Food and Drug Administration website. The keywords used were listed in the paper. The authors also conducted additional searches using authors' names and trial acronyms, and hand and electronic searches of annual scientific sessions of conference proceedings (American College of Cardiology, American Heart Association, European Society of Cardiology, North American Society of Pacing and Electrophysiology) and bibliographies of 43 recent narrative reviews. The authors stated that the searches were limited to 1994 to 2002 since modern cardiac resynchronisation was first reported in 1994.

Study designs of evaluations included in the review

Only randomised trials were included in the review. For crossover trials, only data from the first randomised crossover period were included. Trials in which the randomised follow-up was less than 3 months were excluded.

Specific interventions included in the review

Comparisons of cardiac resynchronisation with a control. Cardiac resynchronisation is a type of pacemaker, which uses a left ventricular lead, usually positioned in a coronary vein to stimulate the left ventricle, at or near the time of right ventricular depolarisation. Only trials using implantable cardioverter defibrillators (ICDs) were included. The control groups in the included studies had ICDs implanted, but the cardiac resynchronisation was turned off.

Participants included in the review

Other than humans with heart failure, the authors did not specify any inclusion or exclusion criteria for the participants. Those included in the review were men and women with a mean age of 63 to 66 years, and with a New York Heart Association (NYHA) functional class of II to IV. All individual studies had excluded patients with conventional pacemaker indication. In addition, they either excluded or did not report the inclusion of patients with a pacemaker contraindication or patients with atrial arrhythmias.

Outcomes assessed in the review

Trials reporting death, hospitalisation for heart failure, or ventricular arrhythmias were included. Results were presented for the following outcomes: progressive heart failure mortality, non-heart failure mortality, all-cause mortality, heart failure hospitalisations and ventricular arrhythmias.

How were decisions on the relevance of primary studies made?

The authors state that two reviewers assessed eligibility in an unblinded standardised manner (see Other Publications Of Related Interest no.2).

Trial methodology was assessed on the basis of the reported use of intention-to-treat analysis and the reported allocation generation, allocation concealment and blinding (see Other Publications Of Related Interest no.1). These details were listed for each included study. The authors state that two reviewers extracted data for the validity assessment in an unblinded standardised manner. Any disagreements between the reviewers were resolved by consensus.

Two reviewers extracted the data in an unblinded standardised manner. The abstracted data included eligibility criteria, baseline characteristics, interventions, outcomes and reported methodological quality. Odds ratios (OR) and 95% confidence intervals (CIs) were calculated for each individual study.

How were the studies combined?

The ORs from the included trials were pooled using a random-effects model, which used weighting based on the inverse variance (DerSimonian and Laird method) to generate a pooled OR and 95% CI.

How were differences between studies investigated?

Chi-squared tests were used to test for heterogeneity.

Eleven reports of 4 randomised controlled trials, with a total of 1,634 patients, were included in the meta-analysis.

A narrative description of the included studies was provided in addition to the quantitative results.

Cardiac resynchronization was associated with a 51% reduction in death from progressive heart failure (4 trials); the OR was 0.49 (95% CI: 0.25, 0.93). There was no evidence of statistical heterogeneity (P=0.85).

Based on data from 3 trials (1,080 randomised patients), cardiac resynchronisation decreased the combined end point of death from progressive heart failure or cardiac transplantation by 59%; the OR was 0.41 (95% CI: 0.19, 0.87). Heterogeneity was not reported.

Cardiac resynchronization was not associated with a statistically- significant effect on non-heart failure mortality (4 trials); the OR was 1.15 (95% CI: 0.65, 2.02). Heterogeneity was not reported. Cardiac resynchronisation was, however, associated with a trend towards reduced all-cause mortality (4 trials); the OR was 0.77 (95% CI: 0.51, 1.18). There was no evidence of statistical heterogeneity between the 4 trials (P=0.83).

Based on data from 3 studies (1,497 patients), cardiac resynchronisation reduced hospitalisation for heart failure by 29%; the was OR 0.71 (95% CI: 0.53, 0.96).

Among patients with ICDs, cardiac resynchronisation was not associated with a statistically-significant reduction in patients experiencing ventricular tachycardia or ventricular fibrillation; the OR was 0.92 (95% CI: 0.67, 1.27).

Sensitivity analyses (for death from progressive heart failure) revealed that fixed- and random-effects models yielded identical results. The exclusion of patients with mild heart failure (NYHA class II) resulted in a change in the OR for death from progressive heart failure from 0.49 (95% CI 0.25, 0.93) to 0.51 (95% CI 0.26, 1.0). The results were similar for patients who received pacemakers and those who received ICDs. The exclusion of single trials one by one had little effect on the ORs.

Cardiac resynchronisation reduces mortality from progressive heart failure in patients with symptomatic left ventricular dysfunction. It also reduces heart failure hospitalisation and shows a trend towards reducing all-cause mortality.

This was a clearly written review and meta-analysis. The authors appear to have been rigorous in their methodology. The search strategy appeared comprehensive, and it is unlikely that published trials would have been missed using this approach. The methods used to assess validity were described, but the authors did not use these results in the sensitivity analyses. However, the authors did note that the exclusion of any single study did not affect the results appreciably. The authors noted that the point estimate of the OR for death from progressive heart failure was very similar following the exclusion of patients with mild heart failure (NYHA class II), but made no comment on the fact that this result only just reached conventional levels of statistical significance (P=0.05).

Overall, the authors' conclusions are consistent with their results.

Practice: The authors did not state any implications for practice.

Research: The authors state that determining whether the costs of cardiac resynchronisation are offset by reduced expenditures for hospitalisation requires further study. They also suggest that future trials will better define those patients who will derive greatest benefit from cardiac resynchronisation.

1. Balk EM, Bonis PA, Moskowitz H, Schmid CH, Ioannidis JP, Wang C, et al. Correlation of quality measures with estimates of treatment effect in meta-analyses of randomized controlled trials. JAMA 2002;287:2973- 82. 2. Berlin JA. Does blinding of readers affect the results of meta- analyses? Lancet 1997;350:185-6.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.