Incremental changes in QRS duration predict mortality in patients with atrial fibrillation

The purpose of this study was to evaluate serial changes in QRS duration in the 12-lead electrocardiogram (ECG) and their prognostic value in patients with atrial fibrillation (AF).

Retrospective evaluation of the 12-lead ECG was performed in 822 patients with AF. Inclusion criteria included AF in two sequential ECGs and an echocardiogram obtained within 7 days of the second ECG. The mean age of the 228 patients enrolled in the study was 71.2+/-12.9 and 45.6% were females. Nearly half of the patients had hypertension (49.5%) and a history of heart failure (44.7%). The patients were followed for 21+/-19 months, and the end point was all-cause mortality.

Cox proportional hazards analysis demonstrated that the independent predictors of mortality in AF patients were age, renal insufficiency, a history of stroke, heart rate, incremental QRS prolongation (P=0.0418), and the absence of warfarin use. Patients with QRS prolongation>15 ms had a worse prognosis than those with QRS prolongation

A progressive increase in QRS duration predicted a poor prognosis in patients with AF in our study. A prospective study is needed to substantiate the prognostic value of QRS prolongation in AF patients.