ABC | Volume 114, Nº2, February 2020

Original Article Ribeiro et al. Simplified method to assess diastolic function by CMR Arq Bras Cardiol. 2020; 114(2):284-292 Table 2 – Cardiac magnetic resonance parameters of patients with aortic valve disease and controls Aortic regurgitation Aortic stenosis Controls p n (%) 11(42.3) 15 (57.7) 19 LVEDV, ml 299.6 ± 68.5 179.99 ± 42.1 129 ± 24.7 < 0.001 LVESV, ml 148.9 ± 60.4 82.0 ± 28.7 45.5 ± 9.4 < 0.001 LVEF, % 51.7 ± 11.4 55.1 ± 9.1 64.7 ± 5.3 < 0.001 LV mass, g 264.2 ± 42.4 272.8 ± 45.5 118.1 ± 40.5 < 0.001 Eccentric Hypertrophy, n (%) 10(90.9) 1(6.7) - Concentric Hypertrophy, n (%) 1(9.1) 14(93.3) - < 0.001 LVEDV: left ventricular end-diastolic volume; LVESV: left ventricular end-systolic volume; LVEF: left ventricular ejection fraction; LV: left ventricular. Definition criteria of concentric hypertrophy is LV mass to LVEDV ratio > 1.16 g/ml. Table 3 – Comparison of atrioventricular motion variables between patients with aortic valve disease (AVD) and healthy volunteers Control AVD p MD (cm) –0.169 ± 0.034 –0.115 ± 0.035 < 0.0001 MVED (s -1 ) 1.439 ± 0.388 0.65 ± 0.413 < 0.0001 VDS (s -1 ) 0.029 ± 0.069 0.224 ± 0.232 < 0.0001 VDS/MVED 0.021 ± 0.051 0.352 ± 0.292 < 0.0001 MD: maximum displacement; MVED: maximum velocity early diastole; VDS: velocity in diastasis. Discussion Novel correlates of diastolic LV function measured by CMR originally investigated in this study were markedly abnormal in patients with AVD. Measured at the AVJ, patients with AVD had significantly lower maximum displacement, slower velocity during early diastolic filling, and higher velocity during diastasis compared to normal control subjects. Saba et al. 9 reported diastolic LV function alterations evaluated through the AVJ motion by CMR in patients with hypertrophic cardiomyopathy compared to normal control patients. Results from our control group were very similar to those reported by these authors, although with slightly greater values mainly because we used a more lateral anatomical landmark. Also, we used only one measurement instead of two of AVJ displacement, hence adopting one well-defined reference point of the AVJ lateral wall, in a more simplified method. LV hypertrophy and diastolic function LV hypertrophy is a recognized risk factor for cardiac morbidity and mortality 15 and is associated with systolic and/or diastolic function disturbances. 16-18 In patients with AVD, diastolic and systolic function disturbances have important implications for morbidity and mortality, before and after aortic valve replacement. 16-21 In the study by Lamb et al., 22 the ejection fractionwas largely unaffected in the group of patients with severe AVD, suggesting that a deterioration of the ejection fraction should be considered as a sign of severe and advanced disease, 22 which was corroborated by other authors. 20,22 After aortic valve replacement, LV diastolic function improves, as indicated by parameters of transmitral flow. 22 In our results, we not only detect diastolic dysfunction in AVDpatients compared to normal controls, but also demonstrated a worse diastolic dysfunction in patients with aortic regurgitation. LV mass was significantly and independently correlatedwith all linearmeasurements of diastolic function evaluated by CMR. Diastolic dysfunction evaluation Phase contrast magnetic resonance imaging allows measurement of flow velocity as well as flow volumes across the mitral valve orifice, providing a new means of diastolic function assessment that may be even more sensitive than Doppler echocardiography. Although it is a well-established tool to assess systolic dysfunction, it is rarely used clinically to assess LV diastolic function, which may require additional dedicated sequences and extensive post-processing. 7 In this sense, in a recently published study by Dusch at al., 7 similar to our study, the authors used a horizontal-long axis SSFP sequence, which they called midwall longitudinal fractional shortening. They verified the percentage of shortening of the distance from the anterior leaflet mitral valve basis to apical endocardium in diastole in relation to systole, comparing these measures to the echocardiogram of 80 patients with varied cardiomyopathies and different degrees of diastolic function. 23,24 Using a simpler method than the one used in the present study, Dusch at al. 7 were able to detect that the midwall longitudinal fractional shortening of grade II/III was significantly lower than that of grade 0/I. 288