ABC | Volume 114, Nº2, February 2020

Original Article Longitudinal Shortening of the Left Ventricle by Cine-CMR for Assessment of Diastolic Function in Patients with Aortic Valve Disease Sergio Marrone Ribeiro, 1 C lerio Francisco de Azevedo Filho, 2 Roney Sampaio, 3 Flávio Tarasoutchi, 3 Max Grinberg, 3 R oberto Kalil-Filho, 3 Carlos Eduardo Rochitte 3 Universidade Estadual Paulista (UNESP), 1 Botucatu, SP – Brazil Universidade do Estado do Rio de Janeiro (UERJ), 2 Rio de Janeiro, RJ – Brazil Instituto do Coração (InCor) – Universidade de São Paulo (USP), 3 São Paulo, SP – Brazil Mailing Address: Sergio Marrone Ribeiro • Universidade Estadual Paulista (UNESP) - Doenças Tropicais e Diagnóstico por Imagem - Distrito de Rubião Júnior s/n. Postal Code 18618-970, Botucatu, SP – Brazil E-mail: marrone@fmb.unesp.br Manuscript received September 17, 2018, revised manuscript April 08, 2019, accepted May 15, 2019 Invited editor for this paper: Dr. João Cavalcante. DOI: 10.5935/abc.20190193 Abstract Background: Diastolic dysfunction, commonly evaluated by echocardiography, is an important early finding in many cardiomyopathies. Cardiac magnetic resonance (CMR) often requires specialized sequences that extends the test time. Recently, feature-tracking imaging has been made available, but still requires expensive software and lacks clinical validation. Objective: To assess diastolic function in patients with aortic valve disease (AVD) and compare it with normal controls by evaluating left ventricular (LV) longitudinal displacement by CMR. Methods: We compared 26 AVD patients with 19 normal controls. Diastolic function was evaluated as LV longitudinal displacement in 4-chamber view cine-CMR images using steady state free precession (SSFP) sequence during the entire cardiac cycle with temporal resolution < 50 ms. The resulting plot of atrioventricular junction (AVJ) position versus time generated variables of AVJ motion. Significance level of p < 0.05 was used. Results: Maximum longitudinal displacement (0.12 vs. 0.17 cm), maximum velocity during early diastole (MVED, 0.6 vs. 1.4s -1 ), slope of the best-fit line of displacement in diastasis (VDS, 0.22 vs. 0.03s -1 ), and VDS/MVED ratio (0.35 vs. 0.02) were significantly reduced in AVD patients compared with controls, respectively. Aortic regurgitation showed significantly worse longitudinal LV shortening compared with aortic stenosis. Higher LV mass indicated worse diastolic dysfunction. Conclusions: A simple linear measurement detected significant differences on LV diastolic function between AVD patients and controls. LV mass was the only independent predictor of diastolic dysfunction in these patients. This method can help in the evaluation of diastolic dysfunction, improving cardiomyopathy detection by CMR, without prolonging exam time or depending on expensive software. (Arq Bras Cardiol. 2020; 114(2):284-292) Keywords: Cardiovascular Diseases/mortality; Cardiomyopathy, Hypertrophic/complications; Diagnostic Imaging; Echocardiography; Magnetic Resonance Spectroscopy; Heart Failure; Aortic Valve Insufficiency. Introduction Diastolic dysfunction is an early marker of cardiac disease and precedes systolic dysfunction. It can occur in the presence or absence of symptoms and with normal or abnormal systolic function. 1,2 There is a high morbidity and mortality associated with this condition due to the potential transition to diastolic heart failure, but it may be underdiagnosed because of the diagnostic criteria. Diastolic dysfunction has an increasing incidence with age and is associated with diabetes mellitus, atrial fibrillation, coronary artery disease, pulmonary hypertension, 3-6 and congenital heart diseases. Left ventricular (LV) hypertrophy has been associated with impaired diastolic function, which is commonly described in systemic hypertension, aortic valve diseases and hypertrophic cardiomyopathy. 7-9 Echocardiography is the most used technique for diastolic dysfunction evaluation in daily clinical routine. Cardiovascular magnetic resonance (CMR) has been widely used for the evaluation of LV morphology and systolic function due to its excellent image quality and lack of geometric assumptions. 9 However, CMR is less used for evaluating diastolic function despite the development of several relevant techniques, 10 including the use of volumetric filling curves, 11 phase-contrast imaging, 12 myocardial tissue tagging, 13 and strain-encoded imaging. 14 The reasons for the limited utilization of these techniques in clinical practice are the time-consuming processes for additional image acquisition and post-processing. For instance, obtaining LV volume curves over the entire cardiac cycle, with the mandatory tracking of endocardial and epicardial contours for all cardiac phases in a cine-CMR series takes a long time and requires a specialized software with automated contour detection. Additionally, other specialized techniques of diastolic dysfunction evaluation 284