Portal hypertension (PH) is a group of syndromes characterized by abnormal changes in the portal blood flow system, mostly caused by cirrhosis. It is an important factor affecting the clinical prognosis of cirrhotic patients, and its severity determines the occurrence and development of cirrhotic complications.

Clinically, measurement of portal venous pressure directly is highly invasive, and factors such as intra-abdominal pressure changes can interfere with the results, limiting its clinical application. Hepatic venous pressure gradient (HVPG) is the gold standard for assessing PH in cirrhosis. The normal range of HVPG is 3~5 mmHg, and HVPG ≥5 mmHg indicates the presence of PH. AASLD stated that HVPG ≥10 mmHg is defined as clinically significant portal hypertension (CSPH), and the risk of decompensation events is significantly increased at this stage. However, HVPG is an invasive test, which is unacceptable to some patients, such as being expensive, difficult to repeat, and poor patient compliance.

Non-invasive tests for PH include serological tests, anatomical imaging and combination models. Imaging evidence of portal collateral circulation or hepatic blood flow in the portal venous system based on ultrasound Doppler, CT or magnetic resonance imaging techniques can assist to diagnose PH. In addition, elastography techniques such as transient elastography, point shear wave elastography, two-dimensional shear wave elastography and magnetic resonance elastography can be used to measure liver stiffness and spleen stiffness to assess PH. Some biochemical markers are also considered as non-invasive tests for PH. However, the available biomarkers are not yet a substitute for the HVPG accurately, and therefore, there is an urgent need for the development of biomarkers associated with HVPG in clinical practice.

Metabolomics is a method to analyze the concentrated changes of endogenous small molecule metabolites under the combined effect of genetic, biological and environmental factors with the help of various high-throughput technologies. Metabolites are at the end of the biological information flow, and their changes are the ultimate expression of the information from the coordinated action of each group, objectively reflecting the overall changes of the organism. Currently, metabolomics techniques have been widely used in screening biomarkers of liver diseases. Wang et al. applied GC-TOF/MS and UPLC-QTOF/MS to study the urinary metabolomics of patients with hepatitis B cirrhosis and showed that α-hydroxymaurolate, tyrosine-betaine, 3-hydroxyisovaleric acid, knife-serine succinate, estrone and GUDCA were significantly altered in different Child-Pugh grades of cirrhosis, suggesting that these metabolites are potential biomarkers to identify different pathological stages of cirrhosis. Therefore, metabolomics is a reliable and valid tool for biomarker discovery.

In conclusion, this study analyzed significantly altered metabolites in patients with hepatitis B cirrhosis using metabolomics to explore potential differential metabolites that are highly correlated with HVPG. Further, serological biomarkers were identified as an alternative to HVPG testing through model construction and validation.

Portal hypertension (PH) is a group of syndromes characterized by abnormal changes in the portal blood flow system, mostly caused by cirrhosis. It is an important factor affecting the clinical prognosis of cirrhotic patients, and its severity determines the occurrence and development of cirrhotic complications.

Clinically, measurement of portal venous pressure directly is highly invasive, and factors such as intra-abdominal pressure changes can interfere with the results, limiting its clinical application. Hepatic venous pressure gradient (HVPG) is the gold standard for assessing PH in cirrhosis. The normal range of HVPG is 3~5 mmHg, and HVPG ≥5 mmHg indicates the presence of PH. AASLD stated that HVPG ≥10 mmHg is defined as clinically significant portal hypertension (CSPH), and the risk of decompensation events is significantly increased at this stage. However, HVPG is an invasive test, which is unacceptable to some patients, such as being expensive, difficult to repeat, and poor patient compliance.

Non-invasive tests for PH include serological tests, anatomical imaging and combination models. Imaging evidence of portal collateral circulation or hepatic blood flow in the portal venous system based on ultrasound Doppler, CT or magnetic resonance imaging techniques can assist to diagnose PH. In addition, elastography techniques such as transient elastography, point shear wave elastography, two-dimensional shear wave elastography and magnetic resonance elastography can be used to measure liver stiffness and spleen stiffness to assess PH. Some biochemical markers are also considered as non-invasive tests for PH. A study by Buck's team showed that serum inflammatory markers in cirrhotic patients were highly correlated with HVPG. Liver fibrosis index has been proposed for the grading and identification of fibrosis with moderate accuracy. Bone bridging protein levels were also correlated with the degree of liver fibrosis and CSPH. Overall, non-invasive serological markers have great potential in assessing PH. However, the available biomarkers are not yet a substitute for the HVPG accurately, and therefore, there is an urgent need for the development of biomarkers associated with HVPG in clinical practice.

Metabolomics is a method to analyze the concentrated changes of endogenous small molecule metabolites under the combined effect of genetic, biological and environmental factors with the help of various high-throughput technologies. Metabolites are at the end of the biological information flow, and their changes are the ultimate expression of the information from the coordinated action of each group, objectively reflecting the overall changes of the organism. Currently, metabolomics techniques have been widely used in screening biomarkers of liver diseases. Wang et al. applied GC-TOF/MS and UPLC-QTOF/MS to study the urinary metabolomics of patients with hepatitis B cirrhosis and showed that α-hydroxymaurolate, tyrosine-betaine, 3-hydroxyisovaleric acid, knife-serine succinate, estrone and GUDCA were significantly altered in different Child-Pugh grades of cirrhosis, suggesting that these metabolites are potential biomarkers to identify different pathological stages of cirrhosis. Therefore, metabolomics is a reliable and valid tool for biomarker discovery.

In conclusion, this study analyzed significantly altered metabolites in patients with hepatitis B cirrhosis using metabolomics to explore potential differential metabolites that are highly correlated with HVPG. Further, serological biomarkers were identified as an alternative to HVPG testing through model construction and validation.

Inclusion Criteria:

1. Case group: age 18-75 years old; patients with hepatitis B cirrhosis diagnosed by liver biopsy, or definite clinical data, or imaging findings; undergo HVPG examination; with written informed consent.
2. Control group: no liver-related diseases, matched with the case group in terms of race, age, sex, and BMI.

Exclusion Criteria:

1. Other etiologies caused cirrhosis, such as HCV, autoimmune hepatitis, alcoholic, non-alcoholic liver disease, etc;
2. Patients with severe liver failure, hepatocellular carcinoma, portal vein thrombosis.
3. Patients with recent blood transfusion due to bleeding.
4. Patients who are pregnant or lactating.
5. Patients with endocrine and metabolic diseases such as diabetes mellitus.
6. Patients treated with anticoagulants, or using drugs that may affect visceral hemodynamics or portal pressure within the last 2 weeks.
7. Severe coagulation dysfunction, international normalized ratio > 5.
8. Those who are unable to lie flat or cannot tolerate the procedure.

The study intended to include patients with hepatitis B cirrhosis who underwent HVPG examination in the hepatology outpatient/inpatient department as the case group, and healthy controls from the physical examination department, including 50 patients with HVPG <10 mmHg and 50 patients with HVPG ≥10 mmHg in the case group, and 50 healthy controls in the control group.