Objectives: Left ventricular (LV) remodelling has an important prognostic value in non ischemic dilated cardiomyopathy (NIDC). 2D echocardiography (2DE) is the most commonly used for assessment of LV remodelling. Real time 3D echocardiography (RT3DE) is more accurate for left ventricle volumes than 2DE, because of its independence to any geometrical assumption. The aim of this study is to compare 2DE and RT3DE measurements of LV remodelling in patients with non NIDC.

Materials and methods: 2DE and RT3DE acquisitions were realised in 41 patients with NIDC using iE33. Philips^® Q Lab software prototype was used for 3D data analysis. 2D volumes were derived using the modified Simpson plan rule. Following 2D and 3D parameters were analysed: Left ventricle End Diastolic Volume (EDV), Left ventricle End Systolic Volume (ESV), Left ventricle Ejection Fraction (EF), Left ventricle Mass (LVM), Sphericity Index (2DSI/RT3DSI). RT3DSI is calculated by dividing EDV by a sphere volume whose diameter was derived from the major end-diastolic LV long axis. In RT3DE, we measured several regional LV volumes to determine LVEDV gradient.

Results: There is a good correlation between 2D and 3DEDV (0.94), 2D and 3DESV (0.96) and 2D and 3DEF (0.91). Correlation is suitable for SI (0.63) and LV long axis (0.78). In RT3DE, 3DEDV seems to be the most accurate parameter to detect LV remodelling, with a cutt-off of 131 mL. In 2DE, sphericity index appears to be the stronger parameter of LVR.

We found the same end-diastolic volume gradient for each patient regardless EF, which indicate an homogenous LV remodelling.

Conclusions: Providing a good evaluation of global and regional LV volumes, RT3DE seems to be more accurate than 2DE to determine LV remodelling.