Three-dimensional echocardiographic assessment of left ventricular volume in different heart diseases using a fully...

HeartModel (HM) is a fully automated adaptive quantification software that can quickly quantify left heart volume and left ventricular function. This study used HM to quantify the left ventricular end-diastolic (LVEDV) and end-systolic volumes (LVESV) of patients with dilated cardiomyopathy (DCM), coronary artery heart disease with segmental wall motion abnormality, and hypertrophic cardiomyopathy (HCM) to determine whether there were differences in the feasibility, accuracy, and repeatability of measuring the LVEDV, LVESV, LV ejection fraction (LVEF) and left atrial end-systolic volume (LAESV) and to compare these measurements with those obtained with traditional two-dimensional (2D) and three-dimensional (3D) methods.

To evaluate the application value of HM in quantifying left heart chamber volume and LVEF in clinical patients.

A total of 150 subjects who underwent 2D and 3D echocardiography were divided into 4 groups: (1) 42 patients with normal heart shape and function (control group, Group A); (2) 35 patients with DCM (Group B); (3) 41 patients with LV remodeling after acute myocardial infarction (Group C); and (4) 32 patients with HCM (Group D). The LVEDV, LVESV, LVEF and LAESV obtained by HM with (HM-RE) and without regional endocardial border editing (HM-NE) were compared with those measured by traditional 2D/3D echocardiographic methods to assess the correlation, consistency, and repeatability of all methods.

(1) The parameters measured by HM were significantly different among the groups (P < 0.05 for all). Compared with Groups A, C, and D, Group B had higher LVEDV and LVESV (P < 0.05 for all) and lower LVEF (P < 0.05 for all); (2) HM-NE overestimated LVEDV, LVESV, and LAESV with wide biases and underestimated LVEF with a small bias; contour adjustment reduced the biases and limits of agreement (bias: LVEDV, 28.17 mL, LVESV, 14.92 mL, LAESV, 8.18 mL, LVEF, -0.04%). The correlations between HM-RE and advanced cardiac 3D quantification (3DQA) (r _s = 0.91-0.95, P < 0.05 for all) were higher than those between HM-NE (r _s = 0.85-0.93, P < 0.05 for all) and the traditional 2D methods. The correlations between HM-RE and 3DQA were good for Groups A, B, and C but remained weak for Group D (LVEDV and LVESV, r _s = 0.48-0.54, P < 0.05 for all); and (3) The intraobserver and interobserver variability for the HM-RE measurements were low.

HM can be used to quantify the LV volume and LVEF in patients with common heart diseases and sufficient image quality. HM with contour editing is highly reproducible and accurate and may be recommended for clinical practice.