S1235 Endoscopic Ultrasound-Guided Shear Wave Elastography of the Liver and Spleen to Predict Complications of Cirrhosis and Correlation With EUS-Guided Portal Pressure Gradient

Abstract

Introduction: Endo-hepatology (EH) is an emerging field which utilizes endoscopic ultrasound (EUS) to evaluate patients with chronic liver disease (CLD). This workup includes EUS-guided liver biopsies, EUS-guided portal pressure gradients (EUS-PPG), EUS-shear wave elastography of the liver (EUS-SWE-L) and spleen (EUS-SWE-S). Our study aimed to determine a correlation between EUS-SWE of the liver , spleen, and EUS-PPG and complications of CLD. Methods: Retrospective analysis of 28 patients from 2021-22. Pearson correlation (PC) coefficients were used to compare EUS-SWE-L and EUS-SWE-S to EUS-PPG. T tests were used to compare EUS-SWE-L and EUS-SWE-S to complications of CLD such as ascites, presence of esophageal varices, and thrombocytopenia. Receiver Operating Characteristics Curve (ROCC) analysis was done to evaluate the utility of shear wave/liver stiffness and shear wave/spleen stiffness to predict the presence of esophageal varices or ascites. Results: 23 patients underwent EUS-SWE-S and 20 had EUS-PPG. Significant correlation noted between platelet count and EUS-SWE-L measurement (PC= -0.496, p=0.01). There was a significant difference in EUS-SWE-L measurement in patients with and without thrombocytopenia (28.90 kPa vs. 17.66 kPa, p=0.002) and in those with and without ascites (29.67 kPa vs. 18.64 kPa, p=0.003). There was also a significant difference in EUS-SWE-S measurements in patients with and without ascites (39.44 kPa vs. 30.06 kPa, p= 0.029). There was a weak correlation between EUS-PPG and EUS-SWE-S (PC = -0.45, p=0.05). ROCC analysis determined the best cut-off for EUS-SWE-L to predict the presence of esophageal varices or ascites was > 22.7 kPa with sensitivity of 78.6% and specificity of 55.6%. Best cut-off for EUS-SWE-S to predict the presence of esophageal varices or ascites was > 31.12 kPa with sensitivity of 85.7% and specificity of 55.6%. (Figure) Conclusion: EUS-SWE of Liver and Spleen showed utility in predicting complications of liver disease such as varices, ascites and thrombocytopenia. There was also a weak, but statistically insignificant correlation between EUS-SWE-S and EUS-PPG. The limitations of our study include small sample size and, therefore, decreased power of the study. Further evaluation with a larger sample size is needed to better elucidate the findings determined in our study regarding EUS-PPG, EUS-SWE-S, and EUS-SWE-L. Our data suggests that these techniques have significant potential in determining those at greatest risk of complications of clinically significant CLD (Table).Figure 1.: Descriptive Statistical Analysis Table 1. - Patient characteristics Total number of patients, n 28 Age, mean in years (std. dev.) 50.25 (+/- 12.02) Male, n (%) 16 (57%) Race- white, n (%) 28 (100%) BMI, mean (std. dev) 31.53 (+/- 9.85) Liver biopsy performed, n (%) 22 (78.6%) Stage > F3 on biopsy, n (%) 16 (72.7%) Esophageal varices, n (%) 9 (32.1%) Ascites, n (%) 12 (42.9%) Plt count < 150, n (%) 16 (57.1%) MELD-Na, mean (std. dev.) 12.67 (+/- 4.97) Child score, mean (std. dev.) 6.81 (+/- 1.82)