Background: Fragmented QRS (fQRS) has been shown to predict cardiac events in select patient populations. Whether fQRS improves patient selection for primary prevention patients eligible for implantable cardioverter-defibrillator (ICD) therapy remains unknown.
Methods and results: In a prospective, multisite cohort of 842 patients with left ventricular dysfunction (ejection fraction < or =35%) representing both ischemic and nonischemic etiology, the presence of fQRS on ECG was assessed using standardized criteria. The association between fQRS and all-cause and arrhythmic mortality was evaluated overall and stratified by ICD status using multivariable Cox regression models, adjusted for demographic, clinical, and treatment variables. Fragmented QRS was present in 274 (32.5%) patients, and there were 191 (22.7%) deaths during a mean follow-up of 40+/-17 months. Rates of all-cause mortality did not differ between the fQRS+ (19.7%) and fQRS- (24.1%) groups; adjusted hazard ratio, 0.88; 95% confidence interval, 0.63-1.22; P=0.43. Additionally, rates of arrhythmic mortality were similar between the fQRS+ (9.9%) and fQRS- (12.7%) groups: adjusted hazard ratio, 0.77; 95% confidence interval, 0.49-1.31; P=0.38. Subgroup analyses found no association between fQRS and mortality when the cohort was further stratified by ICD status, etiology of left ventricular dysfunction, wide (>/=120 ms) versus narrow (<120 ms) QRS duration, or fQRS myocardial territory.
Conclusions: In this prospective, multisite cohort of primary prevention patients with left ventricular dysfunction, the presence of fQRS on ECG was not associated with a higher risk of either all-cause or arrhythmic mortality. These findings do not provide evidence that fQRS would be effective in risk stratifying primary prevention patients eligible for ICD therapy.