589 CLINICAL VALIDATION OF A NEW 3D STRAIN METHOLOGY FOR THE ASSESSMENT OF LEFT VENTRICULAR DEFORMATION IMAGING IN HYPERTENSIVE PATIENTS

Abstract

Aim: The background was to test a new 3D strain algorithm in a clinical setting in order to estimate global and regional left ventricular function against reference methods1,2,3,4. Methods: Were evaluated 30 patients– 10 normals (mean age 30y ±9.7) and 20 patients with arterial hypertension with moderate LVH (mean age 58y±14.0). Standard echocardiography were acquired from apical and parasternal short axis view (frame rate- 65±7 frames/s) and Real time three– dimensional echocardiography (3D) from the apical view at 20±4 frames/s and a full volume data set was captured. Commercial speckle tracking tool was used to extract strain curves - εLL, εCC, εRR from the B-mode images. Finally, εLL, εCC, εRR were estimated from 3D volumetric data using the new methodology (splineMIRIT). For all methods, strain curves were extracted in an 18 segments model of the left ventricle and expressed as percentage. Results: For the global 3D strain in the normals we found values: εLL- 16.33±1.65; εCC, -16.62±1.52; εRR 47.02±13.36. There were significant differences for 3D strain between the normals and hypertensive group (εLL -12.73±2.83, p<0.003; εCC -13.91±4.22, p= 0.047; εRR 30.51±8.18, p<0.009). Similar relationships in global and segmental strain were obtained for the reference methods. Bland-Altman analysis confirmed that the 3D strain algorithm gives lower strain values compared with both reference methods. Conclusion: 3D strain is a feasible method for the assessment of myocardial deformations at the global and segmental levels. The results, obtained from 3D strain algorithm are comparable with those from TDI and speckle tracking. As only a single acquisition is required, 3D method may offer advantages over the current 2D techniques.