Introduction Perivascular Adipose Tissue Fat Attenuation Index (PVAT‐FAI) is assessed using computed tomography angiography (CTA) and has emerged as a biomarker for the development and progression of coronary inflammation. In this study we aim to evaluate the feasibility of using pericarotid FAI in patients with large vessel occlusion strokes. Methods Retrospective cross‐sectional analysis of a single comprehensive‐stroke center prospective thrombectomy database for consecutive patients treated between Nov‐Dec 2024. FAI values were obtained as Hounsfield unit (HU) of fat from the region of interest (ROI) in CTA and range from ‐190 to ‐30 HU (lipid rich‐less inflammation; more aqueous‐more inflammation). In the axial view of CTA‐DICOM image in 3D‐Slicer, a semi‐automated method using the contrast‐media allowed the lumen of the carotid bulb ipsilateral to the stroke to be segmented (Figure1A/1B) while another segment was manually delineated over 15 slices (0.62 mm) (Figure 1C/1D) . The average diameter of the vessel was obtained using the centerline extraction (Figure 1E) and used that to extend the ROI radially from the vessel wall and lumen. The lumen/vessel are subtracted from segmentations, to generate the perivascular region (Figure 1F). The PVAT region of lumen/vessel is exported to obtain HU values. FAIs were analyzed using intraclass coefficient (ICC) after 2 independent readers analyzed the first 5 cases. A generalized linear regression and ordinal regression models were used to evaluate relation between FAI (via vessel wall segmentation), atherosclerosis and outcomes(mRS), adjusting for age, smoking, sex and NIHSS and ASPECTS, respectively. Results Of the 90 consecutive thrombectomies, 50 met inclusion criteria. ICC for FAI was 0.64; p<0.01 via lumen and 0.99; p<0.01 via vessel wall (used for FAI analyses) segmentation. The overall median FAI was ‐62.01 HU (IQR: 73.46 to ‐55.54). Of the 50 patients, 38(76%) had atherosclerosis at carotid bifurcation by CTA. The median FAI of patients with atherosclerosis was ‐60.87 HU IQR: ‐70.42 to ‐54.41, numerically higher than in non‐atherosclerotic bifurcations (‐66.58 HU IQR: ‐82.04 to ‐58.78, adjusted p‐value 0.3). No clear relationship between degree of stenosis and FAI were observed. Lower FAI was associated with good (mRS0‐2) at discharge (OR 0.37; CI‐ 0.15‐0.92) but was not with 90day (OR 0.56; CI 0.11 ‐ 2.76) outcomes in this preliminary analysis. Conclusion Pericarotid FAI was reliably estimated via CTA. Carotid bulb FAI may constitute a promising non‐invasive biomarker for local inflammation, a key driver of carotid disease activity/progression and overall neurological outcomes. This preliminary experience serves a foundation for further investigation image
Pabaney et al. (Sat,) studied this question.