Identification of Confounding Factors on 2-D Shear-Wave Elastography of Kager’s Fat Pad: An Ex-Vivo Study

Abstract

Introduction: To identify the confounding factors presented by Keger’s fat pad (KFP) using 2-D shear-wave elastog-raphy (2-D SWE) in phantom. Methods: The study was conducted between January and June 2023. A total of eleven bovine ankles were scanned using various gel layers, scanning planes, ankle positions, and ROI sizes. The stiffness measurement (in m/s) of KFP for each ankle was acquired within a day. To assess interobserver reliability, seven an-kles were rescanned by a second observer. Result: Wilcoxon sign-rank test indicated a significant in shear wave ve-locity (SWV) between longitudinal and transverse plane (p = 0.043). A significant difference was observed in different gel layers (p = 0.002). The SWV was higher when the ROI diameter was 3.0 mm compared to 1.0 mm (p < 0.001). Insignificant difference was also observed in different ankle positions (p = 0.756). The coefficient variance was low-er in longitudinal, one-layer gel, with an ROI of 1.0 mm and in neutral position with (p < 0.05). For interobserver reliability, the intraclass correlation coefficient (ICC) was 0.819 (95% CI, 0.737–0.875) for the proposed protocol. The ICC between the two machines was 0.746 (95 % CI, 0.625 - 0.827) with the Bland Altman test indicating no proportional bias (p = 0.519). SWV in fat pad is affected by the scanning plane, gel layers, and ROI size. Conclusion: Using a standard scanning protocol, 2-D SWE can produce reliable images even with different scanners.