CLINICAL STUDY. remain incomplete and novel non-invasive strategies are needed in order to effectively identify high-risk HCM patients.

Transcription

1 CLINICAL STUDY Quantitative Differentiation of LV Myocardium with and without Layer-Specific Fibrosis Using MRI in Hypertrophic Cardiomyopathy and Layer-Specific Strain TTE Analysis Nobusada Funabashi, 1 MD, Hiroyuki Takaoka, 1 MD, Koya Ozawa, 1 MD, Tomoko Kamata, 1 MD, Masae Uehara, 2 MD, Issei Komuro, 2 MD and Yoshio Kobayashi, 1 MD Summary To achieve further risk stratification in hypertrophic cardiomyopathy (HCM) patients, we localized and quantified layer-specific LVM fibrosis on MRI in HCM patients using regional layer-specific peak longitudinal strain (PLS) and peak circumferential strain (PCS) in LV myocardium (LVM) on speckle tracking transthoracic echocardiography (TTE). A total of 18 HCM patients (14 males; 58 ± 17 years) underwent 1.5T-MRI and TTE. PLS and PCS in each layer of the LVM (endocardium, epicardium, and whole-layer myocardium) were calculated for 17 AHA-defined lesions. MRI assessment showed that fibrosis was classified as endocardial, epicardial, or whole-layer (= either or both of these). Regional PLS was smaller in fibrotic endocardial lesions than in non-fibrotic endocardial lesions (P = 0.004). To detect LV endocardial lesions with fibrosis, ROC curves of regional PLS revealed an area under the curve (AUC) of and a best cut-off point of 13.5%, with sensitivity of 65.3% and specificity of 54.3%. Regional PLS was also smaller in fibrotic epicardial lesions than in nonfibrotic epicardial lesions (P < 0.001). To detect LV epicardial lesions with fibrosis, ROC curves of PLS revealed an AUC of and a best cut-off point of 9.5%, with sensitivity of 73.5% and specificity of 55.5%. Using whole-layer myocardium analysis, PLS was smaller in fibrotic lesions than in non-fibrotic lesions (P < 0.001). To detect whole-layer LV lesions with fibrosis, ROC curves of regional PLS revealed an AUC of and a best cut-off point of 12.5%, with sensitivity of 79.0% and specificity of 50.7%. There were no significant differences in PCS of LV myocardium (endocardium, epicardium, and whole-layer) between fibrotic and nonfibrotic lesions. Quantitative regional PLS but not PCS in LV endocardium, epicardium, and whole-layer myocardium provides useful non-invasive information for layer-specific localization of fibrosis in HCM patients. (Int Heart J 2018; 59: ) Key words: Left ventricular myocardium, Layer-specific peak longitudinal strain, Speckle tracking transthoracic echocardiography, Magnetic resonance imaging R isk stratification for major adverse cardiac events (MACE) including sudden death is important for the management of patients with hypertrophic cardiomyopathy (HCM) because it has been established that implantable cardioverter defibrillators for preventing sudden cardiac death are effective. 1) Occurrence of non-sustained ventricular tachycardia (VT) 2) and detection of late enhancement (LE) in left ventricular (LV) myocardium suggesting the presence of myocardial fibrosis, by T1 weighted cardiac magnetic resonance imaging (MRI), 3,4) are very useful predictors for MACE in HCM patients as well as non-dilated LV size and mitral regurgitation 5) and dilated phase HCM. 6) Furthermore, the genetic variation in the SCN10A gene might be associated with cardiac conduction abnormalities in HCM patients. 7) However, current stratification algorithms remain incomplete and novel non-invasive strategies are needed in order to effectively identify high-risk HCM patients. We previously compared the clinical significance of occurrence of non-sustained VT and that of the presence of LE in LV myocardium, which suggested the presence of myocardial fibrosis, as detected by computed tomography (CT) during risk stratification for MACE in HCM patients without obstructed coronary arteries. We concluded that the presence of myocardial fibrosis in LV myocardium on CT is a stronger predictor of MACE compared with the presence of non-sustained VT in such patients. 8) We also performed 2-dimensional (2D) speckle tracking transthoracic echocardiography (TTE) to compare the ability to predict the occurrence of MACE between global longitudinal strain (GLS) and global circumferential strain From the 1 Department of Cardiovascular Medicine, Chiba University Graduate School of Medicine, Chiba, Japan and 2 Department of Cardiovascular Medicine, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. This work is partially supported by a Grant from Japan Heart Foundation Research Grant (no grant numbers). Address for correspondence: Nobusada Funabashi, MD, Department of Cardiovascular Medicine, Chiba University Graduate School of Medicine, Inohana, Chuo-ku, Chiba City, Chiba , Japan. nobusada@w8.dion.ne.jp Received for publication April 10, Revised and accepted August 20, Released in advance online on J-STAGE May 9, doi: /ihj All rights reserved by the International Heart Journal Association. 523

2 524 Funabashi, ET AL IntHeartJ May 2018 Figure 1. Actual images of left ventricular (LV) regional peak longitudinal strain (PLS) analysis (upper image) and its strain chart (lower image) on 2 dimensional speckle tracking transthoracic echocardiogram (TTE). On TTE, PLS for each LV segment among 17 lesions (as defined by the American Heart Association) was calculated using Echo PAC, version 113 (GE Healthcare). In the strain chart, the dotted line indicates global strain (GS) which was the average of 6 solid lines. Each colored line indicates strain change during the cardiac cycle at the corresponding same colored site in LV myocardium in the upper TTE image. AVC indicates aortic valve closure. (GCS) in LV myocardium in HCM patients without obstructed coronary arteries. We concluded that both 2D LV GLS and GCS (2D LV GLS>GCS) on TTE can predict whether such patients will have a poor prognosis. 9) Furthermore, a recently developed multi-layer speckle tracking TTE technique has enabled the analysis of LV endocardium, epicardium, and whole layer myocardium strain separately ) We speculated that using speckle tracking TTE for regional layer-specific peak longitudinal (PLS) and peak circumferential (PCS) strain in LV myocardium, we could localize and quantify layer-specific LV myocardial fibrosis confirmed by T1 weighted cardiac MRI in HCM patients. Therefore, in this study, to achieve further risk stratification in HCM patients, we sought to ascertain the relationship between regional layer-specific peak strain (PLS and PCS) in LV myocardium and the presence of layerspecific LV myocardial fibrosis confirmed by T1 weighted cardiac MRI in HCM patients. Methods Patient population: A total of 18 HCM patients (14 males; mean age, 58 ± 17 years) underwent 1.5 Tesla (T) MRI (Achieva, Philips) and TTE (Vivid E9, GE Healthcare). Speckle tracking TTE: Using TTE, PLS and PCS for each layer of the LV myocardium (endocardium, epicardium, and whole layer) among 17 lesions (PLS) 16) and 12 lesions (mid and basal portions, PCS), as defined by the American Heart Association, 17) were calculated using Echo PAC, version 113 (GE Healthcare) (Figure 1). MRI protocol: Using MRI, the patients were placed supine in a clinical 1.5 T MRI with 5-channel cardiac coils around the chest. As previously reported, all MRIs were obtained using electrocardiogram (ECG) gating and during repeated episodes of breath-holding. 18) Surface-coil intensity correction was performed for both cine MRI and MRI for LE. Cine MRIs were acquired using a steadystate free precession sequence. After acquiring cine MRIs on the 2- and 4-chamber long-axis projections, we obtained short axis cine MRIs encompassing the LV from base to apex. MRIs for LE were acquired 10 to 15 minutes after intravenous administration of 0.15 mmol/kg of gadopentetate dimeglumine (Magnevist; Schering AG). An inversion-recovery prepared, T1-weighted, three dimen-

3 IntHeartJ May 2018 LAYER-SPECIFIC FIBROSIS AND STRAIN TTE IN HCM 525 sional gradient-echo sequence was used to obtain MRIs for LE in the same planes as the cine images. The inversion time was adjusted to minimize the signal from the normal myocardium in each patient by using a look-locker sequence to locate the null point of the normal myocardium. A typical inversion time for LE MRI ranged from 230 to 300 ms. Imaging data were analyzed on workstations (Ziostation 2, Ziosoft). In the T1 weighted MRI acquired 10 minutes after contrast injection, we obtained images of the LV short axis view, the long axis 4 chamber views, and the long axis 2 chamber views. If there was an obvious high intensity area in the LV myocardium, we regarded this lesion as being LE positive. The location of LE, suggesting the presence of myocardial fibrosis, was classified as endocardium, epicardium, or whole layer (= either or both the endocardium and epicardium) using T1-weighted MRI (Figure 2). Statistical analysis: For all analyses, P < 0.05 was considered to be statistically significant. Comparison of PLS and PCS (absolute values) on speckle tracking TTE between LV lesions (endocardium, epicardium, or whole layer) with and without fibrosis on MRI were performed using the t test (SPSS software, version 17.0, SPSS, Inc.). Results Patient background and TTE findings are presented in Table I and the MRI LV findings are presented in Table II. Of a total of 306 lesions, fibrosis was detected in 72, 68, and 81 lesions in the LV endocardium, epicardium, and whole layer myocardium using T1 weighted MRI, respectively. Endocardial analysis demonstrated that regional PLS was significantly smaller in fibrotic endocardial lesions (11.7 ± 6.8%) than in non-fibrotic endocardial lesions (15.0 ± 8.6%) on MRI (P = 0.004) (Figure 3). To detect LV endocardial lesions with fibrosis, receiver operating characteristic (ROC) curve analysis of regional PLS revealed an area under the curve (AUC) of (P = 0.008) and a best cut-off point of 13.5%, with sensitivity of 63.9% and specificity of 53.8% (Figure 4). Epicardial analysis demonstrated that regional PLS was significantly smaller in fibrotic epicardial lesions (6.8 ± 4.5%) than in non-fibrotic epicardial lesions (10.6 ± 6.3%) on MRI (P < 0.001) (Figure 5). To detect LV Table II. Magnetic Resonance Imaging Left Ventricular (LV) Findings Figure 2. Typical late gadolinium enhancement image on T1 weighted cardiac magnetic resonance imaging suggesting the presence of myocardial fibrosis in a patient with hypertrophic cardiomyopathy. Myocardial fibrosis was observed in the whole-layer (mid layer dominant), in the whole range portion of the interventricular septum, and the LV anterior and inferior wall (arrows). LV indicates left ventricle; and RV, right ventricle. Variable LV mass (g) 196 ± 93 LV end-diastolic volume (ml) 136 ± 34 LV end-systolic volume (ml) 60 ± 21 LV ejection fraction (%) 56 ± 9 Maximum LV wall thickness (mm) 20 ± 6 At the examination, one patient had headache and cine magnetic resonance imaging could not be performed. These data were acquired from the remaining 17 patients. Table I. Patient Backgrounds and TTE Findings Variable n = 18 Age (years) 58 ± 17 Male 14 (78%) Time interval between TTE and MRI (months) 4 ± 5 Hypertrophic obstructive cardiomyopathy 4 (22%) Dilated phase hypertrophic cardiomyopathy 0 (0%) Left atrial dimension (mm) 44.1 ± 7.6 Left atrial volume index (ml/m 2 ) 29.8 ± 11.0 Inter-ventricular septum thickness diameter on end diastole (mm) 16.9 ± 5.0 LV posterior wall thickness on end diastole (mm) 10.1 ± 1.9 LV end-diastolic dimension (mm) 48.1 ± 5.9 LV end-systolic dimension (mm) 29.8 ± 4.2 LV ejection fraction (%) 64.4 ± 6.5 TTE indicates transthoracic echocardiogram; MRI, magnetic resonance imaging; and LV, left ventricle.

4 526 Funabashi, ET AL IntHeartJ May 2018 epicardial lesions with fibrosis, ROC curve analysis of PLS revealed an AUC of (P < 0.001) and a best cut-off point of 9.5%, with sensitivity of 73.5% and specificity of 55.5% (Figure 6). For whole layer myocardium analysis, PLS was significantly smaller in fibrotic lesions (8.7 ± 4.9%) than in non-fibrotic lesions (12.5 ± 6.7%) on MRI (P < 0.001) (Figure 7). To detect whole layer (endocardial and/or epicardial) LV lesions with fibrosis, ROC curve analysis of regional myocardial PLS revealed an AUC of (P < 0.001) and a best cut-off point of 12.5%, with sensitivity of 79.0% and specificity of 50.7% (Figure 8). There were no significant differences in PCS of LV myocardium (endocardium, epicardium, and whole layer) between fibrotic lesions (30.6 ± 7.6%, 10.8 ± 5.5%, and 17.7 ± 6.4%, respectively) and non-fibrotic lesions (33.4 ± 9.5%, 10.8 ± 6.6%, and 19.3 ± 7.8%, respectively) using MRI (Figures 9-11). Discussion We sought to ascertain the relationship between regional layer-specific peak strain (PLS and PCS) in LV myocardium and the presence of layer-specific LV myo- Figure 3. Comparison of peak longitudinal strain (PLS) (absolute left ventricular endocardial lesions with and without fibrosis on cardiac magnetic resonance imaging in patients with hypertrophic cardiomyopathy. On endocardial analysis, regional PLS was significantly smaller in fibrotic endocardial lesions (n = 72) (11.7 ± 6.8%) than in non-fibrotic endocardial lesions (n = 234) (15.0 ± 8.6%) (P = 0.004). Figure 5. Comparison of peak longitudinal strain (PLS) (absolute left ventricular epicardial lesions with and without fibrosis on cardiac magnetic resonance imaging in patients with hypertrophic cardiomyopathy. On epicardial analysis, regional PLS was significantly smaller in fibrotic epicardial lesions (n = 68) (6.8 ± 4.5%) than in non-fibrotic epicardial lesions (10.6 ± 6.3%) (n = 238) (P < 0.001). Figure 4. Receiver operating characteristic (ROC) curve of regional peak longitudinal strain (PLS) (absolute values) to detect left ventricular endocardial lesions with fibrosis on cardiac magnetic resonance imaging in patients with hypertrophic cardiomyopathy. ROC curve revealed an area under the curve (AUC) of (P = 0.005) and a best cut-off point of 13.5%, with sensitivity of 65.3% and specificity of 54.3%.

5 IntHeartJ May 2018 LAYER-SPECIFIC FIBROSIS AND STRAIN TTE IN HCM 527 Figure 6. Receiver operating characteristic (ROC) curve of regional peak longitudinal strain (PLS) (absolute values) to detect left ventricular epicardial lesions with fibrosis on cardiac magnetic resonance imaging in patients with hypertrophic cardiomyopathy. ROC curve of PLS to detect LV epicardial lesions with fibrosis revealed an area under the curve (AUC) of (P < 0.001) and a best cut-off point of 9.5%, with sensitivity of 73.5% and specificity of 55.5%. Figure 7. Comparison of peak longitudinal strain (PLS) (absolute left ventricular whole layer lesions with and without fibrosis on cardiac magnetic resonance imaging in patients with hypertrophic cardiomyopathy. On whole layer myocardium analysis, PLS was significantly smaller in fibrotic lesions (n = 81) (8.7 ± 4.9%) than in non-fibrotic lesions (n = 225) (12.5 ± 6.7%) (P < 0.001). cardial fibrosis confirmed by T1 weighted cardiac MRI in HCM patients. In this study, we conclude that quantitative regional PLS but not PCS in LV endocardium, epicardium, and whole layer myocardium provides useful non-invasive information for layer-specific localization of fibrosis confirmed on MRI in HCM patients. Clinical significance, implications, and importance of this study: Until now, myocardial fibrosis on MRI on its own, 3,4) or sole use of whole-layer myocardial strain on TTE, 9,19) have been reported as candidates for predictors of HCM in patients. Furthermore, fibrotic lesions on MRI or CT have been reported to have impaired (whole layer) strain values on TTE, 3,20) and the relationship between the presence of LE on MRI and decrement of strain values in each myocardial lesion is estimated to strongly interact with each other. Currently, due to developments in TTE technology, we can measure myocardial strain in endocardial and epicardial layers separately, while also being able to measure whole layer as well. We have already reported the utilities of these novel layer-specific strain measurements ) The spatial resolution of TTE is superior to that of MRI, therefore, if accurate diagnosis of the presence of layer-specific LV myocardial fibrosis is confirmed by T1 weighted cardiac MRI, we speculate that the regional layer-specific peak strain values measured on TTE would be reduced in the corresponding layer-specific lesions with myocardial fibrosis using MRI. Therefore, in this study, we determined the detailed relationship between regional layer-specific peak strain (both PLS and PCS) in LV myocardium and the presence of layer-specific LV myocardial fibrosis confirmed by T1 weighted cardiac MRI in HCM patients. We believe both technologies would be novel and robust risk stratified predictors for HCM patients. Secondly, we aimed to differentiate HCM from other myocardial diseases such as cardiac amyloidosis. We aimed to do this by identifying layer-specific strain on TTE and/or the presence of layer-specific LV myocardial fibrosis on MRI. Cummings and colleagues used a pattern-based approach to assess delayed enhancement in non-ischemic cardiomyopathy on MRI. 21) Spatial resolution on MRI is superior to that of single photon emission CT and positron emission tomography. However, these techniques are inferior to that of CT and TTE, therefore, a partial volume effect may have influenced their results. To confirm the reliability of their hypothesis, we have to confirm that the regional layer-specific peak strain measure-

6 528 Funabashi, ET AL IntHeartJ May 2018 Figure 8. Longitudinal strain (PLS) (absolute values) to detect left ventricular whole layer lesions with fibrosis on cardiac magnetic resonance imaging in patients with hypertrophic cardiomyopathy. ROC curve of regional myocardial PLS to detect whole layer (endocardial and/or epicardial) LV lesions with fibrosis revealed an area under the curve (AUC) of (P < 0.001) and a best cut-off point of 12.5%, with sensitivity of 79.0% and specificity of 49.3%. Figure 9. Comparison of peak circumferential strain (PCS) (absolute left ventricular (LV) endocardial lesions with and without fibrosis on cardiac magnetic resonance imaging (MRI) in patients with hypertrophic cardiomyopathy. There were no significant differences in the PCS of LV endocardium between fibrotic lesions (n = 39) (30.6 ± 7.6%) and non-fibrotic lesions (n = 177) (33.4 ± 9.5%) on MRI. Figure 10. Comparison of peak circumferential strain (PCS) (absolute left ventricular (LV) epicardial lesions with and without fibrosis on cardiac magnetic resonance imaging (MRI) in patients with hypertrophic cardiomyopathy. There were no significant differences in the PCS of LV epicardium between fibrotic lesions (n = 34) (10.8 ± 5.5%) and non-fibrotic lesions (n = 182) (10.8 ± 6.6%) on MRI. ment on TTE in the corresponding layer-specific lesions with myocardial fibrosis on MRI would be reduced. Thus, we believe that quantitative regional PLS but not PCS in endocardium, epicardium, and whole myocardium provides useful non-invasive information for layer-specific localization of fibrosis on MRI in HCM patients. To detect LV layer-specific lesions with fibrosis, we determined the best cut-off point for each layer-specific regional PLS by assessing the area under the curve using receiver operated characteristic analysis. This would be very important and useful to show that the Cummings hypothesis is correct. The third aim was to understand the mechanism of how myocardial fibrosis influences each strain value. To achieve this and confirm the aforementioned hypothesis, it would have been better to use a homogenous population and only include HCM patients to be evaluated in this study. Utility of a multi-layer strain technique on 2D speckle tracking TTE: Using a multi-layer strain technique on 2D speckle tracking TTE, we previously evaluated compensatory mechanisms in HCM patients with preserved LV ejection fraction. 14) In that study, we advocated a new myocardial characteristic indicator; this was the myocardial percentage endocardial strain dependency, as represented by 2D LV GLS and GCS. In such patients, we

7 IntHeartJ May 2018 LAYER-SPECIFIC FIBROSIS AND STRAIN TTE IN HCM 529 Figure 11. Comparison of peak circumferential strain (PCS) (absolute left ventricular (LV) whole layer lesions with and without fibrosis on cardiac magnetic resonance imaging (MRI) in patients with hypertrophic cardiomyopathy. There were no significant differences in the PCS of LV whole layer myocardium between fibrotic lesions (n = 43) (17.7 ± 6.4%) and non-fibrotic lesions (n = 173) (19.3 ± 7.8%) on MRI. concluded that 2D LV GLS was lower than in controls but endocardial GCS was maintained in compensation for reduction in endocardial GLS. Consequently, the percentage endocardial GCS dependency may increase, and with a larger LV size, the compensatory effect will be smaller. Study limitations: This was a non-randomized study in a single center with a small number of patients. We used a novel multilayer TTE technology, and given that the number of patients was small and there were only a few observation periods, we did not detect anyone with events during the observation periods. Therefore, we could not directly evaluate whether regional layer-specific peak strain in LV myocardium and/or the presence of layerspecific LV myocardial fibrosis confirmed by T1 weighted cardiac MRI were robust prognostic predictors in HCM patients. In this study, we did not evaluate the inter-observer and intra-observer consistency of either layer-specific localization of fibrosis on MRI or layer-specific PLS and PCS in LV myocardium on speckle tracking TTE. However, we have previously evaluated inter- and intraobserver consistency in LV myocardial strain measurements using a novel multi-layer technique in patients with severe aortic stenosis and preserved LV ejection fraction. 15) Furthermore, we also evaluated inter- and intra-observer correlation coefficients for a proportion of endocardial GLS and GCS dependency estimates which were defined as the ratio of endocardial strain to epicardial strain in HCM patients with preserved LV ejection fraction 14) and we generated reliable results. The population in this report was comprised of patients with severe aortic stenosis but the measurement of layer-specific strain on TTE was automatically performed using TTE machines and if sufficient image data are acquired, inter- and intra-observer consistency would be less dependent on the different types of disease, and furthermore, both severe aortic stenosis and HCM have thickened LV walls. Therefore, we believe we can adapt the inter- and intra-observer consistency in the measurement of LV myocardial strain using a novel multilayer technique in patients with severe aortic stenosis to that compatible with HCM. When using layer-specific analysis incorporating both TTE and MRI, both modalities are required for excellent spatial resolution and reduction of partial volume effects. Generally, the spatial resolution of MRI is inferior to that of TTE. Therefore, the accuracy of layer-specific localization of fibrosis on MRI may be poorer than that of layerspecific measurements of PLS and PCS in LV myocardium on speckle tracking TTE, which may have influenced the results presented in this study. For analysis of MRI, it is generally preferable that it be performed not only by cardiologists but also by radiologists, or jointly by a radiologist and a cardiologist. Conclusion: Quantitative regional PLS but not PCS in endocardium, epicardium, and whole myocardium provides useful non-invasive information for layer-specific localization of fibrosis on MRI in HCM patients. Disclosures Conflicts of interest: There is no conflict of interest to declare. References 1. Maron BJ, Casey SA, Chan RH, Garberich RF, Rowin EJ, Maron MS. Independent Assessment of the European Society of Cardiology Sudden Death Risk Model for Hypertrophic Cardiomyopathy. Am J Cardiol 2015; 116: Monserrat L, Elliott PM, Gimeno JR, Sharma S, Penas-Lado M, McKenna WJ. Non-sustained ventricular tachycardia in hypertrophic cardiomyopathy: an independent marker of sudden death risk in young patients. J Am Coll Cardiol 2003; 42: Ismail TF, Jabbour A, Gulati A, et al. Role of late gadolinium enhancement cardiovascular magnetic resonance in the risk stratification of hypertrophic cardiomyopathy. Heart 2014; 100: Sakuma H. Late gadolinium enhancement and prognosis of hypertrophic cardiomyopathy. Circ J 2014; 78: Kubo T, Baba Y, Hirota T, et al. Prognostic significance of nondilated left ventricular size and mitral regurgitation in patients with dilated phase of hypertrophic cardiomyopathy. Int Heart J 2017; 58: Yoshinaga M, Yoshikawa D, Ishii H, et al. Clinical characteristics and long-term outcomes of hypertrophic cardiomyopathy. Int Heart J 2015; 56: Iio C, Ogimoto A, Nagai T, et al. Association between genetic variation in the SCN10A gene and cardiac conduction abnormalities in patients with hypertrophic cardiomyopathy. Int Heart J 2015; 56: Takaoka H, Funabashi N, Uehara M, Ozawa K, Kobayashi Y. Successful prediction of MACE by myocardial fibrosis on CT in hypertrophic cardiomyopathy patients without obstructed coronary arteries. Int J Cardiol 2015; 199: Ozawa K, Funabashi N, Takaoka H, Kobayashi Y. Successful MACE risk stratification in hypertrophic cardiomyopathy patients using different 2D speckle-tracking TTE approaches. Int J Cardiol 2017; 228: Ozawa K, Funabashi N, Sugiura A, Kobayashi Y. Layer specific strain measurement and its relationship to heart failure indica-

8 530 Funabashi, ET AL IntHeartJ May 2018 tors in systemic autoimmune disorder patients: A multi-layer transthoracic echocardiography study. Int J Cardiol 2016; 220: Nishi T, Funabashi N, Ozawa K, et al. Resting multilayer 2D speckle-tracking transthoracic echocardiography for the detection of clinically stable myocardial ischemic segments confirmed by invasive fractional flow reserve. Part 1: Vessel-byvessel analysis. Int J Cardiol 2016; 218: Ono R, Funabashi N, Ozawa K, Takaoka H, Kamata T, Kobayashi Y. An educational intervention to help medical students achieve accurate and consistent measurement of longitudinal myocardial strain on transthoracic echocardiogram. Int J Cardiol 2015; 201: Ozawa K, Funabashi N, Kobayashi Y. Left ventricular myocardial strain gradient using a novel multi-layer transthoracic echocardiography technique positively correlates with severity of aortic stenosis. Int J Cardiol 2016; 221: Ozawa K, Funabashi N, Takaoka H, et al. Characteristic myocardial strain identified in hypertrophic cardiomyopathy subjects with preserved left ventricular ejection fraction using a novel multi-layer transthoracic echocardiography technique. Int J Cardiol 2015; 184: Ozawa K, Funabashi N, Kamata T, Kobayashi Y. Inter- and intraobserver consistency in LV myocardial strain measurement using a novel multi-layer technique in patients with severe aortic stenosis and preserved LV ejection fraction. Int J Cardiol 2017; 228: Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2015; 28: Cerqueira MD, Weissman NJ, Dilsizian V, et al. A statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation 2002; 105: Takaoka H, Funabashi N, Uehara M, Iida Y, Kobayashi Y. Diagnostic accuracy of CT for the detection of left ventricular myocardial fibrosis in various myocardial diseases. Int J Cardiol 2017; 228: Funabashi N, Takaoka H, Horie S, et al. Risk stratification using myocardial peak longitudinal-strain on speckle-tracking transthoracic-echocardiogram to predict major adverse cardiac events in non ischemic hypertrophic-cardiomyopathy subjects confirmed by MDCT. Int J Cardiol 2013; 168: Funabashi N, Kataoka A, Horie S, et al. Distinguishing 320 slice CT-detected focal fibrotic lesions and non-fibrotic lesions in hypertrophic cardiomyopathy by assessment of regional myocardial strain using two dimensional speckle tracking echocardiography. Int J Cardiol 2013; 169: e Cummings KW, Bhalla S, Javidan-Nejad C, Bierhals AJ, Gutierrez FR, Woodard PK. A pattern-based approach to assessment of delayed enhancement in nonischemic cardiomyopathy at MR imaging. Radiographics 2009; 29: