A NOVEL 3-DIMENSIONAL TECHNIQUE IN MEASURING PERICORONARY EPICARDIAL ADIPOSE TISSUE

Abstract

Pericoronary epicardial adipose tissue radiodensity (PCATrd) is of emerging interest as a risk factor for coronary artery disease (CAD), but can be difficult to measure. This study aims to investigate a novel semi automated three-dimensional quantification of PCATrd. Fifty-five patients with suspected CAD were retrospectively identified. PCATrd was measured in contrast CT data-set using image analysis software (ITK-SNAP, Version 3.6.0). We applied a thresholding map, using in-house MatLab code with range of (-190,-3) Hounsfield unit derived from previous work, to the contrast CT data-set. The cursor was probed in the center of the short axis of the coronary arteries in the coronal, sagittal and axial images simultaneously. Then the 3D round brush segmented the epicardial adipose tissue with the orthogonal distance from the coronary vessel wall equal to the coronary radius and, the segmentation followed the path of 4 major coronary arteries from proximal to the distal terminal (Figure 1 A & B). The PCATrd of each coronary artery was automatically computed. We compared this 3D segmenting method (3D_PCATrd) with a 7-segment segmentation method (7S_PCATrd) published by the Society of Cardiovascular Computed Tomography. We further compared both PCATrd estimates with coronary artery disease severity (non-obstructive, obstructive and extensive obstructive) by using ordered logistic regression. The 3D_PCATrd was substantially different from the 7S_PCATrd in the general_PCATrd, LM, LAD and RCA, but similar in LCX (Table 1 & Figure 1C). Standard deviations around 3D_PCATrd estimates were significantly smaller than estimates from 7S_PCATrd. The 3D_PCATrd, but not 7S_PCATrd, was independently associated with severity of coronary lesions (multivariable odds ratio 0.86 for per 1 HU increase, p=0.017). Our novel 3-dimension quantified PCATrd technique produced estimates that were significantly different than estimates derived from the SCCT published technique, was more precise than the 7-segment quantified PCATrd and was better associated with coronary artery disease severity.View Large Image Figure ViewerDownload Hi-res image Download (PPT)