Prediction of Response to Cardiac Resynchronization Therapy: The Selection of Candidates for CRT (SCART) Study

Abstract
Background: The aim of this study was to evaluate the ability of baseline clinical and echocardiographic parameters to predict a positive response to CRT. Methods: We analyzed 6‐month data from the first 133 consecutive patients enrolled in a multicenter prospective study. These patients had symptomatic heart failure (HF) refractory to pharmacological therapy (NYHA class II–IV), left ventricular ejection fraction (LVEF) ≤35%, and prespecified electrocardiographic, echocardiographic or tissue Doppler imaging markers of left ventricular (LV) dyssynchrony. Results: After a follow‐up period of 6 months, 1 patient died and 13 were hospitalized for worsening HF. There were significant (P < 0.01) clinical, functional, and echocardiographic improvements that included: New York heart Association Class, Quality‐of‐Life Score, QRS duration, LVEF, LV end‐diastolic and end‐systolic diameter (LVESD), and severity of mitral regurgitation A positive response was documented in 90/133 (68%) patients who presented an improved clinical composite score associated to an increase in LVEF ≥ 5 units. A multivariate analysis identified that a smaller LVESD (OR = 0.957, 95% CI 0.920–0.996; P = 0.030) and longer interventricular mechanical delay (IVMD) (OR = 1.017, 95% CI 1.005–1.029, P = 0.007) as independent predictors of a positive response. Receiver‐operating curve analysis showed that a positive response to CRT may be predicted in patients with IVMD > 44 ms (with a sensitivity of 66% and a specificity of 55%) or with LVESD < 60 mm (with a sensitivity of 66% and a specificity of 61%). Conclusions: Our results confirm the limited value of QRS duration in the selection of patients for CRT. A less‐advanced stage of disease and echocardiographic evidence of interventricular dyssynchrony demonstrated to predict response to CRT, while intraventricular dyssynchrony did not predict response.

This publication has 31 references indexed in Scilit: