INTRODUCTION: Predicting reverse remodeling after cardiac resynchronization therapy (CRT) remains challenging and different etiologies of heart failure might hamper identification of predictors. OBJECTIVE: Assess the incremental value of mechanical dyssynchrony besides electrical dyssynchrony for predicting CRT response. METHODS: 227 patients (51% ischemic) received CRT. Response was defined as ≥15% left ventricular end systolic volume decrease after six months. Prediction models were developed comprising clinical parameters and electrical dyssynchrony (Model A), subsequently complemented with mechanical dyssynchrony (Model B). Models were compared by area under the receiver-operating curve (AUC), net reclassification index (NRI) and integrated discrimination improvement (IDI) for the complete cohort, ischemic (ICM) and non-ischemic (NICM) subpopulations. RESULTS: Model B performed significantly better than Model A supported by AUC, NRI and IDI. Furthermore, model B significantly better predicted response for NICM than ICM. CONCLUSION: Electrical dyssynchrony and mechanical dyssynchrony are essential to predict CRT response. Nevertheless, response prediction for ICM remains challenging.
INTRODUCTION: Predicting reverse remodeling after cardiac resynchronization therapy (CRT) remains challenging and different etiologies of heart failure might hamper identification of predictors. OBJECTIVE: Assess the incremental value of mechanical dyssynchrony besides electrical dyssynchrony for predicting CRT response. METHODS: 227 patients (51% ischemic) received CRT. Response was defined as ≥15% left ventricular end systolic volume decrease after six months. Prediction models were developed comprising clinical parameters and electrical dyssynchrony (Model A), subsequently complemented with mechanical dyssynchrony (Model B). Models were compared by area under the receiver-operating curve (AUC), net reclassification index (NRI) and integrated discrimination improvement (IDI) for the complete cohort, ischemic (ICM) and non-ischemic (NICM) subpopulations. RESULTS: Model B performed significantly better than Model A supported by AUC, NRI and IDI. Furthermore, model B significantly better predicted response for NICM than ICM. CONCLUSION: Electrical dyssynchrony and mechanical dyssynchrony are essential to predict CRT response. Nevertheless, response prediction for ICM remains challenging.
Authors: Wouter M van Everdingen; Alexander H Maass; Kevin Vernooy; Mathias Meine; Cornelis P Allaart; Frederik J De Lange; Arco J Teske; Bastiaan Geelhoed; Michiel Rienstra; Isabelle C Van Gelder; Marc A Vos; Maarten J Cramer Journal: Cardiovasc Ultrasound Date: 2017-10-18 Impact factor: 2.062
Authors: W M van Everdingen; J C Schipper; J van 't Sant; K Ramdat Misier; M Meine; M J Cramer Journal: Neth Heart J Date: 2016-01 Impact factor: 2.380