Cardiac resynchronization therapy (CRT) has important role in the contemporary treatment of heart failure, systolic dysfunction and mechanical disynchrony. Classical indications for CRT are severe heart failure (NYHA class III or IV), a broad QRS (more than 120 ms) and left ejection fraction less than 35% despite optimal medical therapy. Several have studies demonstrated the important role of echocardiography in patient selection for CRT, follow up and estimation of CRT effects, as well as the optimization of biventricular pacemaker. Basically, there are three types of cardiac asynchrony: interventricular asynchrony, between the right and left ventricle, intraventricular asynchrony, between the myocardial segments within the left ventricle and atrioventricular asynchrony, between the atria and ventricles. Although many echocardiographic techniques are used in patient selection for CRT, no ideal approach has yet been found. There are several techniques and parameters used in the assessment of myocardial asynchrony: two dimensional (2D) echocardiography, one dimensional echocardiography (M-mode), Doppler echocardiography, different modalities of tissue Doppler including Colour Coded Tissue Doppler Imaging - TDI, measurements of local tissue deformation indices (strain and strain rate), speckle tracking, 3D echocardiography, semiquantitative assessment of myocardial border, vector velocity imaging. Each of these techniques has advantages and limitations. A special accent in this revue is on the consensus report from the American Society of Echocardiography Dyssynchrony Writing group. According to this consensus report color coded tissue Doppler is the most appropriate technique for myocardial asynchrony estimation and patients selection for CRT. The same group recommended that definitive decision for CFT implantation should not be based only on echocardiographic analysis, but rather on the whole clinical aspect of the patient.
Can machine learning improve patient selection for cardiac resynchronization therapy?
