Noninvasive Prediction of Portal Hypertension in Cirrhosis Using Sound Touch Viscoelastography

1. Research subjects: Subjects were included in strict accordance with the preparation procedures (inclusion criteria, informed consent), and with the approval of the Ethics Committee of the First Affiliated Hospital of Wenzhou Medical University, relevant data were collected within 1 week after the patients were enrolled and before the HVPG test.. 2. Clinical baseline information and laboratory data: including name, gender, age, blood pressure, BMI (height, weight), causes of liver cirrhosis and other general information, and relevant laboratory tests, including platelet count, hemoglobin, albumin, prothrombin time, international normalized ratio, total bilirubin, creatinine, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, blood ammonia, etc. 3. Conventional gray-scale ultrasound and color Doppler examination: including hepatic artery inner diameter, hepatic artery peak flow velocity, hepatic artery resistance index, portal vein trunk inner diameter, portal vein average flow velocity, spleen size, etc. 4. STVi detection: Use the color Doppler ultrasound system (including elastic components) of Shenzhen Mindray Biomedical Electronics Co., Ltd., equipped with an abdominal convex array probe with a probe frequency of 1\~6 MHz. The probe was placed in a supine position or slightly tilted to the left, with the right arm raised and fully abducted to increase the width of the intercostal space, and the liver viscoelastic index of the right lobe of the liver was measured between the intercostals. During measurement, the subject should hold his breath for 3 to 5 seconds in a calm state. Do not hold his breath after taking a deep breath. The sampling frame should be placed in a place with uniform echo in the liver parenchyma, avoiding large blood vessels, bile ducts and ribs. The sampling frame should be as parallel to the liver capsule as possible and placed 1 to 2 cm below the liver capsule and no more than 6 cm. The region of interest is preferably placed in the center of the elastogram, and the diameter of the sampling frame is recommended to be ≥1.5 cm. Take the median after 3 valid measurements of the same site. The sampling results require measurement success rate ≥60% and IQR/Median≤0.3. 5. HVPG detection: Use the right jugular vein approach, refer to the Chinese Expert Consensus on Clinical Application of Hepatic Venous Pressure Gradient (2018 Edition) , and select the balloon catheter to the hepatic vein under fluoroscopy, at a distance from the inferior vena cava 2\~4 cm, wait at least 20 s (some patients may take longer to reach a stable reading), and then read the Free Hepatic Venous Pressure (FHVP) after the pressure value is stable. After injecting contrast medium or air to expand the balloon to fully block the hepatic venous blood flow, wait at least 40 s until the pressure value is stable, and then read the Hepatic Venous Wedge Pressure (WHVP). Keep the balloon in the inflated state, instruct the patient to hold his breath, slowly inject 5 ml of contrast agent through the balloon catheter, and perform hepatic venography to confirm that there is no contrast agent reflux or venous-venous collateral shunt. Calculated according to the formula HVPG = WHVP - FHVP, HVPG should be the average of 2 measurements. 6. Group according to the HVPG test results and divide them into 2 groups: CSPH group (HVPG≥10 mmHg) and non-CSPH group (HVPG\<10 mmHg). 7. Build a model: Screen out independent variable information related to the occurrence of CSPH through model variable screening and correlation analysis, and use it to build a prediction model. And establish a Nomogram model to realize the visualization of the model. 8. Evaluate the model: Discrimination of the model: ROC curve analysis, C-index; Calibration of the model: Hosmer-Lemeshow goodness of fit test, consistency curve (test); Clinical net benefit assessment: Decision curve analysis ( DCA).