Work

基于高尔基体蛋白73及前白蛋白的血清学标志物模型无创评估临床显著性门静脉高压

Year: 2026

Type: article

Abstract: Objective To investigate the diagnostic value of a serological marker model based on Golgi protein 73 (GP73) and prealbumin for clinically significant portal hypertension (CSPH). Methods Clinical data of 237 patients with cirrhotic portal hypertension admitted to Taiyuan Third People's Hospital from January 2021 to December 2022 were collected retrospectively. All patients underwent serum GP73 and prealbumin (PA) testing and hepatic venous pressure gradient (HVPG) measurement. Routine blood tests, liver function, coagulation function and other related indicators were also collected. According to whether they were in the compensated stage, all patients were divided into a compensated group (n=97) and a decompensated group (n=140). Patients with compensated cirrhosis(n=97) were further randomly divided into a training set and a valiation set at a ratio of 7:3 to construct and validate the diagnostic model. A diagnostic model for clinically significant portal hypertension (CSPH) was constructed using logistic regression analysis based on serological markers. The area under the receiver operating characteristic (ROC) curve (AUC) was employed to evaluate the model's discriminative ability for diagnosing CSPH. Internal validation was performed using the bootstrap method (bootstrap sampling) to assess the model's stability and consistency, with comparisons to that of the aspartate aminotransferase-to-platelet ratio index (APRI) and the fibrosis-4 (FIB-4) index. Intergroup comparisons of measurement data were performed using t-tests or rank-sum tests, while intergroup comparisons of categorical data were conducted using the χ(2) test. Results Among the 237 patients with cirrhotic portal hypertension, 97 (40.93%) had compensated cirrhosis and 140 (59.07%) had decompensated cirrhosis. There were 144 males (60.76%) and 93 females (39.24%), with a median age of 54 years. The predominant etiologies were hepatitis B in 150 cases (63.29%), hepatitis C in 30 cases (12.66%), alcoholic hepatitis in 11 cases (4.64%), combined hepatitis B and C in 2 cases (0.84%), and other etiologies in 44 cases (18.57%). GP73 levels were positively correlated with HVPG (r=0.34, P<0.001) in patients with compensated cirrhosis, whereas prealbumin levels were negatively correlated with HVPG (r=-0.36, P<0.001). Multivariate logistic regression analysis revealed that GP73, PA, and platelet count were independent predictors of clinically significant portal hypertension (CSPH), and a diagnostic model for CSPH was established accordingly, designated the GPP model. The model had an area under the receiver operating characteristic curve (AUC) of 0.88 (95% confidence interval: 0.81-0.93), a sensitivity of 0.89, a specificity of 0.77, a positive predictive value of 0.65, and a negative predictive value of 0.93, with the selection of the optimal cutoff value of 0.74 for diagnosing CSPH in the training set. The model had an AUC of 0.82, a sensitivity of 0.83, a specificity of 0.65, a positive predictive value of 0.62, and a negative predictive value of 0.85, with the selection of the optimal cutoff value of 0.74 for diagnosing CSPH in the validation set. The GPP score following bootstrap resampling had a mean AUC of 0.87, which was higher than the mean AUCs for the APRI (0.79) and FIB-4 (0.76) scores when compared with the GPP score with APRI and FIB-4. Conclusion The GPP score, which incorporates GP73, prealbumin, and platelet count, provides a novel and non-invasive approach for assessing CSPH in patients with compensated cirrhosis.

Source: PubMed Central

DOI: 10.3760/cma.j.cn501113-20250824-00345