BackgroundQRS duration is an important risk factor for both sudden cardiac death and all-cause mortality in patients with moderate to severe left ventricular systolic dysfunction. However, it is unknown whether baseline QRS duration can help predict the mortality benefit from implantable cardioverter-defibrillators (ICDs) when used for primary prevention purposes.ObjectiveOur objective was to examine the relationship between baseline QRS duration and mortality reduction with ICDs implanted for primary prevention by performing a meta-analysis of all available randomized clinical trials.MethodsA systematic literature search of SCOPUS (covering MEDLINE and EMBASE) was conducted for randomized controlled clinical trials involving ICDs. Primary prevention trials of systolic dysfunction that reported all-cause mortality according to subgroups on the basis of baseline QRS duration were included.ResultsA total of four randomized controlled clinical trials enrolling 3561 patients were included. Patients with QRS durations less than 120ms [n=2002, hazard ratio 0.78, 95% confidence interval (CI) 0.63-0.96, P=0.018] and at least 120ms (n=1559, hazard ratio 0.59, 95% CI 0.48-0.73, P=0.0000009) had significant decreases in all-cause mortality. On heterogeneity analysis, a strong trend for greater magnitude of benefit in patients with QRS at least 120ms was observed (22% risk reduction vs. 41% risk reduction, P=0.066).ConclusionFor primary prevention purposes, ICDs are effective in reducing all-cause mortality in moderate to severe systolic dysfunction regardless of QRS duration. However, patients with a baseline QRS duration of at least 120ms seem to derive a greater mortality benefit from ICDs.