Abstract Background Dyssynchronous activation of the heart is a prerequisite for cardiac resynchronization (CRT) efficacy. Although the main goal of CRT is to resynchronize ventricular contraction, our abilities to quantify mechanical dyssynchrony are still limited. However, so called “internal stretch fraction” (ISF) might serve as a marker of global discoordination of ventricular contraction. Methods and aim of the study Thirty-six CRT candidates and 7 healthy controls underwent MRI scanning. Feature tracking data (in short axis) were analysed using Segment software and cicrumferential strain (S)/ strainrate (SR) was obtained for each of the 16 left-ventricular (LV) segments. ISF was calculated as an extent of stretch (product of number of stretched regions and their mean amount of stretch, integrated over given time period; ie. area under the positive part of the global SR curve) in relation to the extent of shortening (area under the negative part of the global SR curve) in 6 basal LV segments during the ejection time (ET). The onset and end of ejection for ISF calculation were set from the global S curve (obtained by the averaging of S curves in 6 basal LV segments). For flow-derived ISF (FISF) parameter, ET was defined from the MRI-flow through pulmonic and aortic valve (Fig-1). Higher ISF and FISF marks more ineffective ventricular contraction. Aim of our study was to assess the ability of ISF and FISF to distinguish healthy controls from CRT candidates and to predict CRT responders (↓LVESV>15% after 6 months of CRT). Results ISF and FISF were significantly higher in CRT candidates than in healthy controls, greater difference between the two groups exhibited FISF (Fig-2). There were 37% of nonresponders among CRT recipients. Both parameters (ISF and FISF) were higher in future CRT responders. Cut-off value discriminating CRT responders from CRT nonresponders was >0,045 for ISF and >0,174 for FISF, however FISF had higher sensitivity than ISF (Fig-3; sensitivity 71% (ISF) versus 79% (FISF) and specificity 70% for both ISF and FISF). Conclusion Internal stretch fraction might be a useful parameter for selection of patients with discoordinated ventricular contraction who might profit from CRT. Higher sensitivity of FISF compared to ISF may be explained by the fact that ET derived from global S curve reflects only part of the ejection phase whereas FISF describes presence of ventricular discoordination in the entire ejection phase. Acknowledgement/Funding Supported by Ministry of Health of the Czech Republic, grant nr. 15-31398A. All rights reserved.