# Ultrasonography Elastography to Predict the Diagnosis, Severity, and Treatment Indication of Esophageal Varices in Patients with Chronic Liver Diseases

## TL;DR

This study explores using ultrasound elastography to non-invasively assess esophageal varices in patients with chronic liver disease, offering a potential alternative to uncomfortable endoscopy.

## Contribution

The study introduces cut-off values for elastography parameters to predict the presence and treatment need for esophageal varices.

## Key findings

- Elastography parameters Vs and F-index were significantly higher in patients with esophageal varices.
- Cut-off values for Vs and F-index showed high sensitivity for detecting varices and determining treatment need.
- Elastography performed similarly to other methods in predicting variceal presence and severity.

## Abstract

Background/Objectives: Esophageal varices (EVs) are a serious complication of liver cirrhosis. Guidelines for cirrhosis/chronic liver diseases (CLDs) do not specify a follow-up period or the need for esophagogastroduodenoscopy (EGD). EGD is a useful but uncomfortable procedure for the assessment of varices. Follow-up with abdominal ultrasonography (AUS) is recommended in patients with CLDs. If EVs are assessed by AUS, more patients eligible for endoscopic screening of EVs can be selected. We aimed to investigate whether AUS elastography [shear wave (Vs) and F-index] can predict the diagnosis, severity, and treatment indication of EVs. Methods: Between April 2018 and October 2022, we retrospectively collected data of 194 patients who underwent elastography and EGD for CLDs. The correlations between Vs/F-index values and presence/severity of EVs were evaluated. Each cut-off value for diagnosis and treatment indication of EVs was investigated. Results: 85 patients without exclusion criteria were enrolled. Vs and F-index values were significantly higher in patients with EVs than in patients without EVs (p = 0.0005 and 0.0021, respectively) and positively correlated with severity of EVs. The cut-off Vs and F-index values for the presence of EVs were 1.63 m/s and 1.88, respectively, with 88.1%/83.3% sensitivity, 48.8%/51.2% specificity, and 0.71/0.70 area under the curve (AUC). The cut-off Vs and F-index values for treatment indication were 1.71 m/s and 2.08, respectively, with 100%/88.2% sensitivity, 45.6%/52.9% specificity, and 0.69/0.70 AUC. There were no significant differences between the two modalities. Conclusions: Elastography may provide objective assessment and thus be a non-invasive screening tool for diagnosis and treatment indication of EVs.