Defining the optimal technique for a novel EUS platform combining two-dimensional shear wave elastography and attenuation imaging: A preclinical study

Background and Objectives: Noninvasive diagnostics have become an integral part of the evaluation of patients with chronic liver diseases. Ultrasound-based elastography estimates the degree of fibrosis within the liver by measuring the speed through which shear waves, generated from a push-pulse, propagate through the liver. Ultrasound can also be used to estimate the degree of hepatic steatosis by measuring the attenuation of sound waves through the liver. Ultrasound-based elastography and attenuation typically use a transabdominal probe, though recent developments allow for both two-dimensional shear wave elastography (2D-SWE) and attenuation imaging (ATI) through a curvilinear echoendoscope. The aim of this study was to define the optimal technique for 2D EUS-SWE ATI and to compare performance with currently available modalities. Methods: A curvilinear echoendoscope was used to test the accuracy of measurements compared with a hydrogel reference model for different regions of interest depth, size, placement method, and number of measurements. These measurements were then repeated by 5 expert endosonographers using surgically implanted reference models in porcine liver. Statistical analysis was performed using R (version 4.4.1, R Core Team 2021, Vienna, Austria). Results: Optimal settings for region of interest size, depth, placement, and number were identified. Accuracy (defined by % deviation from reference) of SWE was highest for 2D-SWE compared with conventional modalities, while accuracy of the novel platform’s ATI was similar to the controlled attenuation parameter. There was excellent interobserver agreement using the novel platform with optimal technique (intraclass correlation coefficient of 0.999 [ P < 0.001]). Conclusions: Our study serves as the basis for obtaining measurements with optimal accuracy using a novel EUS-based platform combining 2D-SWE and ATI. In this preclinical model, the novel platform demonstrated excellent accuracy across a broad range of reference models, with excellent interobserver agreement, and was more accurate than conventional modalities using point SWE.