View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Elsevier - Publisher Connector Journal of the American College of Cardiology Vol. 56, No. 22, 2010 © 2010 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00 Published by Elsevier Inc. doi:10.1016/j.jacc.2010.06.044 Cardiac Imaging Longitudinal and Circumferential Strain Rate, Left Ventricular Remodeling, and Prognosis After Myocardial Infarction Chung-Lieh Hung, MD,*ʈ¶# Anil Verma, MD,* Hajime Uno, PHD,** Sung-Hee Shin, MD,* Mikhail Bourgoun, MD,* Amira H. Hassanein, MD,* John J. McMurray, MD,† Eric J. Velazquez, MD,‡ Lars Kober, MD,§ Marc A. Pfeffer, MD, PHD,* Scott D. Solomon, MD,* for the VALIANT Investigators Boston, Massachusetts; Durham, North Carolina; Copenhagen, Denmark; and Taipei, Taiwan Objectives We sought to investigate the clinical prognostic value of longitudinal and circumferential strain (S) and strain rate (SR) in patients after high-risk myocardial infarction (MI). Background Left ventricular (LV) contractile performance after MI is an important predictor of long-term outcome. Tissue deforma- tion imaging might more closely reflect myocardial contractility than traditional measures of systolic functions. Methods The VALIANT (Valsartan in Acute Myocardial Infarction Trial) Echo study enrolled 603 patients with LV dysfunc- tion, heart failure, or both 5 days after MI. We measured global peak longitudinal S and systolic SR (SRs) from apical 4- and 2-chamber views and global circumferential S and SRs from parasternal short-axis view with speckle tracking software (Velocity Vector Imaging, Siemens, Inc., Mountain View, California). We related global S and SRs to LV remodeling at 20-month follow-up and to clinical outcomes. Results Both longitudinal (mean: Ϫ5.1 Ϯ 1.6 100/ms) and circumferential SRs (mean: Ϫ8.0 Ϯ 2.8 100/ms) were pre- dictive of death or hospital stay for heart failure (hazard ratio: 2.4, 95% confidence interval [CI]: 2.0 to 3.1, p Ͻ 0.001; hazard ratio: 1.3, 95% CI: 1.2 to 1.4, p Ͻ 0.001, respectively) after adjustment for clinical covariates by Cox proportional hazards, and longitudinal SRs further improved in predicting 18-month survivor on a model based on clinical and standard echocardiographic measures (increase in area under the receiver-operator char- acteristic curve: 0.13, p ϭ 0.009). With multivariable logistic regression, circumferential SRs, but not longitudi- nal SRs, was strongly predictive of remodeling (odds ratio: 1.3, 95% CI: 1.1 to 1.4, p Ͻ 0.001). Conclusions Both longitudinal and circumferential SRs were independent predictors of outcomes after MI, whereas only cir- cumferential SRs was predictive of remodeling, suggesting that preserved circumferential function might serve to restrain ventricular enlargement after MI. (J Am Coll Cardiol 2010;56:1812–22) © 2010 by the American College of Cardiology Foundation Quantification of left ventricular ejection fraction (LVEF) function (6). Moreover, accurate assessment of cardiac on the basis of ventricular volumes has been the primary mechanics has proven elusive to traditional imaging modal- method for assessing myocardial systolic function and myo- ities, due in part to the complex spatial orientation and cardial damage after myocardial infarction (MI) (1–5). This distribution of muscle fibers in the longitudinal and circum- measure, however, is load-dependent and neglects regional ferential direction (7,8). Emerging echocardiographic tech- *From the Cardiovascular Division, Brigham and Women’s Hospital, Boston, Massa- for presentations at meetings. Dr. Pfeffer has received grant/research support from chusetts; †Western Infirmary, Glasgow, United Kingdom; ‡Duke University Medical Amgen, Baxter, Bristol-Myers Squibb, Celladon, Novartis, and Sanofi-Aventis; and Center, Durham, North Carolina; §Rigshospitalet, Copenhagen, Denmark; ʈDivision of has served as a consultant to Affectis Anthera, AstraZeneca, Biogen, Boehringer, Cardiology, Department of Internal Medicine, Mackay Memorial Hospital, Taipei, Boston Scientific, Bristol-Myers Squibb, Celladon, Cytokinetics, Daiichi Sankyo, Gen- Taiwan; ¶Mackay Medical College, Taipei, Taiwan; #Mackay Medicine, Nursing and zyme, Gilead, GlaxoSmithKline, Medtronic, Mirabila, Nicox, Novartis, Roche, Salutria, Management College, Taipei, Taiwan; and the **Department of Biostatistics, Dana- Sanofi-Aventis, and Via Pharmaceuticals. Dr. Solomon has received research support Farber Cancer Institute, School of Public Health, Harvard University, Boston, from Novartis. All other authors have reported that they have no relationships to disclose. Massachusetts. Dr. Velazquez has served as an advisory board member and received Manuscript received January 26, 2010; revised manuscript received June 11, 2010, research grants from Novartis and Boehringer Ingelheim. Dr. Kober has received fees accepted June 14, 2010. JACC Vol. 56, No. 22, 2010 Hung et al. 1813 November 23, 2010:1812–22 Strain Rate and Remodeling and Prognosis After MI niques, including speckle-tracking imaging derived strain investigate the clinical and prog- Abbreviations (S) and strain rate (SR)—which might more accurately nostic importance of both mea- and Acronyms reflect intrinsic measures of myocardial contractility and sures on clinical outcomes and LV area under curve ؍ AUC remodeling in patients with high- confidence interval ؍ CI See page 1823 risk MI. heart failure ؍ HF ejection fraction ؍ enable quantification of regional myocardial deformation— Methods EF end-systolic volume ؍ have allowed for analysis of longitudinal and circumferential ESV left atrial ؍ motion and have contributed substantially to the under- Patient population. The LA left ventricle/ventricular ؍ standing of cardiac mechanics (9–11). VALIANT (Valsartan in Acute LV myocardial infarction ؍ Recent studies comparing longitudinal and circumferen- Myocardial Infarction Trial) en- MI tial function have shown that longitudinal function deteri- rolled 14,703 patients with LV wall motion score ؍ WMSI orates early under cardiac pathologic conditions before the dysfunction, heart failure, or index strain ؍ onset of clinical symptoms and reduction in global ventric- both between 12 h and 10 days S ular function (12), whereas circumferential function might after acute MI and randomized strain rate ؍ SR remain relatively preserved to compensate for cardiac function them to receive valsartan, capto- ؍ when longitudinal function starts to become dysfunctional (13) pril, or a combination (15). The SRs systolic strain rate dimensional-2 ؍ and in patients with early stage heart failure (14). We sought to median follow-up period was 2D Figure 1 Peak Longitudinal and Circumferential S and Systolic SR Curves In a normal subject (A), strain (S) is negative with downward tracing, whereas strain rate (SR) represents the rate of S and is downward with myocardial shortening dur- ing systolic phase and upward during diastolic phase. In a subject with myocardial infarction (B), longitudinal S and SR from infarction segments (red arrows) were lower compared with other segments together with a reduced global (blue arrow) S and SR compared with the normal subject. The representative color codes correspond to each segment curve in S and SR. 1814 Hung et al. JACC Vol. 56, No. 22, 2010 Strain Rate and Remodeling and Prognosis After MI November 23, 2010:1812–22 24.7 months. The VALIANT Echo study (3) enrolled 610 nal and circumferential peak S and systolic strain rate (SRs) patients from the VALIANT population who underwent analysis on the basis of the availability of optimal endocar- 2-dimensional (2D) echocardiography at a mean time of dial border definition. Both longitudinal and circumferential 5.0 Ϯ 2.5 days, and 603 patients had acceptable baseline S and SRs analyses were available in a total of 311 cases. images. Of the 603 patients enrolled for echo substudy after Echocardiographic analysis. All conventional echocardio- exclusion of those with unqualified image quality or missing graphic parameters including M-mode, 2D, and Doppler views, 20% had LV dysfunction—defined as LVEF Յ35% images were analyzed by the core laboratory at the Brigham to 50% of patients had heart failure diagnosed, and 30% had and Women’s hospital from 3 separate cardiac cycles. The both. A scale system ranging from 1 (optimal) to 3 (unac- LV volumes and EF were traced manually at end-diastole ceptable) for each segment was adopted for the selection of and -systole at apical 4- and 2-chamber views and derived image quality suitable for Velocity Vector Imaging analysis from modified biplane Simpson’s method. Other measures, by 2 clinical physicians. Echo images with at least 1 segment including LV mass index, left atrial (LA) volume index, scaled 3 were excluded from future analysis. After detailed mitral regurgitation severity, wall motion score index review of the baseline echo images by 2 physicians, 380 (WMSI), and pulsed wave Doppler assessed from mitral (63%) of the initial 603 VALIANT apical 2- and leaflets tip were calculated as previously described (3,4,16). 4-chamber images and 420 (75%) of the initial 558 VAL- Midwall fractional shortening was calculated from M- IANT short-axis images were chosen for further longitudi- mode–guided LV linear dimensions by using a spherical BaselineTable 1 ClinicalBaseline Characteristics Clinical Characteristics of Patients Stratified of Patients by Stratified Longitudinal by Longitudinal SRs Group SRs Group Longitudinal SRs (100/ms) ؊6.1 ؊6.1 to ؊4.7 ؊4.7 to ؊4.0 >؊4.0 Trend>