Objectives: To test the hypothesis that echocardiographic dyssynchrony may assist in the selection of patients with borderline QRS duration for cardiac resynchronization therapy (CRT).
Background: Although echocardiographic dyssynchrony is currently not recommended to select patients with QRS duration widening for CRT, its utility in patients with borderline QRS widening is unclear.
Methods: Of 221 consecutive heart failure patients with an ejection fraction (EF) < or =35% referred for CRT, 86 had a borderline QRS duration of 100 to 130 ms (115 +/- 8 ms) and 135 patients had wide QRS >130 ms (168 +/- 26 ms). Dyssynchrony was assessed using interventricular mechanical delay, tissue Doppler imaging longitudinal velocity opposing wall delay, and speckle tracking radial strain for septal to posterior wall delay. Response to CRT was defined as > or =15% increase in EF, and reverse remodeling as > or =10% decrease in end-systolic volume.
Results: There were 201 patients with baseline quantitative echocardiographic data available, and 187 with follow-up data available 8 +/- 5 months after CRT. A smaller proportion of borderline QRS duration patients (53%) were EF responders compared with 75% with widened QRS (p < 0.05). Interventricular mechanical delay > or =40 ms and opposing wall delay > or =65 ms were predictive of EF response in the wide QRS duration group, but not the borderline QRS duration group. Speckle tracking radial dyssynchrony > or =130 ms, however, was predictive of EF response in both wide QRS interval patients (88% sensitivity, 74% specificity) and borderline QRS interval patients (79% sensitivity, 82% specificity) and associated reverse remodeling with reduction in end-systolic volume (p < 0.0005).
Conclusions: Radial dyssynchrony by speckle tracking strain was associated with EF and reverse remodeling response to CRT in patients with borderline QRS duration and has the potential to assist with patient selection.
2010 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.