Prediction of response to cardiac resynchronization therapy by speckle tracking echocardiography using different...

Although several echocardiographic approaches exist to quantify mechanical dyssynchrony, the comparative ability of different manufacturers' speckle-tracking software programs to predict response to cardiac resynchronization therapy (CRT) is unknown.

Eighty-four patients with heart failure referred for CRT were studied (mean age, 64 +/- 12 years; mean ejection fraction [EF], 26 +/- 7%; mean QRS duration, 157 +/- 26 ms). Dyssynchrony was assessed using the same midventricular short-axis digital cine loop for each patient with 3 different offline speckle-tracking strain analysis programs: software A, speckle-tracking two-dimensional strain; software B, velocity vector imaging strain; and software C, speckle-tracking strain. Significant dyssynchrony was defined as an anterior septum-to-posterior wall delay > or =130 ms. Follow-up was available for 57 patients (mean, 7 +/- 4 months). Response to CRT was defined as an EF increase > or =15%.

Variability between software results was observed when patients had large degrees of dyssynchrony (> or =200 ms), with limits of agreement from 123 to 214 ms. However, close agreement for identifying patients with significant dyssynchrony was observed: 91% for software A versus B, 96% for software A versus C, and 93% for software B versus C. Importantly, the 3 software approaches' ability to predict EF outcome had similar sensitivities, specificities, and areas under receiver operating characteristic curves: 0.87, 0.86, and 0.86, respectively.

Radial strain dyssynchrony analyses by 3 different speckle-tracking software programs were similarly able to predict EF response to CRT. Although variability in absolute values of dyssynchrony was observed, there was close agreement for determining the presence or absence of significant dyssynchrony. Speckle-tracking echocardiography has potential as a means to quantify dyssynchrony in a multicenter clinical trial or clinical practice.