2025, 41(8): 1579-1588.
DOI: 10.12449/JCH250817
Abstract:
Objective To investigate the independent risk factors for portal vein thrombosis (PVT) in patients with viral hepatitis-related decompensated cirrhosis, and to establish and validate a nomogram risk prediction model. Methods A retrospective analysis was performed for the clinical data of 1 116 patients with decompensated HBV/HCV cirrhosis who attended The Third People’s Hospital of Kunming for the first time from January 2022 to December 2023, and according to the presence or absence of PVT, they were divided into PVT group and control group. The independent samples t-test or the Mann-Whitney U test was used for comparison of continuous data between groups, and the chi-square test was used for comparison of categorical data between groups. Univariate analysis and least absolute shrinkage and selection operator (LASSO) regression analysis were used to identify variables, and a binary logistic regression analysis was used to obtain independent influencing factors and establish a predictive model, which was visualized using a nomogram. The model was validated based on the receiver operating characteristic (ROC) curve, the area under the ROC curve (AUC), the Hosmer-Lemeshow test, Bootstrap sampling (1 000 iterations), the calibration curve, the decision curve analysis (DCA), and the clinical impact curve (CIC). Results There were 178 patients in the PVT group and 938 patients in the control group, and the prevalence rate of PVT was 15.9%(178/1 116). Male patients accounted for 68.5%(764/1 116), and the patients with drinking, Child-Pugh class B liver function, and ascites accounted for 51.0%(569/1 116),78.8%(879/1 116), and 67.1% (749/1 116), respectively. Compared with the control group, the PVT group had significantly higher age (Z=-2.362,P<0.05), prothrombin time (Z=-2.403,P<0.05), international normalized ratio (Z=-2.470,P<0.05), free thyroxine (Z=-5.910,P<0.05), D-dimer (Z=-5.764,P<0.05), interleukin-6 (Z=-6.581,P<0.05), interleukin-10(IL-10) (Z=-3.915,P<0.05), interleukin-8 (Z=-3.705,P<0.05), diameter of the portal vein (Z=-9.690,P<0.05), and spleen thickness (Z=-7.183,P<0.05), as well as significantly lower levels of white blood cell count (Z=-2.115,P<0.05), platelet count (Z=-3.026,P<0.05), fibrinogen (Z=-2.169,P<0.05), alanine aminotransferase (Z=-3.151,P<0.05), prealbumin (Z=-3.509,P<0.05),cholinesterase (Z=-3.415,P<0.05), alpha-fetoprotein (Z=-3.513,P<0.05), triglycerides (Z=-2.679,P<0.05), CD3 cell count (Z=-6.059,P<0.05), CD4 cell count (Z=-7.257,P<0.05), CD8 cell count (Z=-2.340,P<0.05), CD4+/CD8+ cell ratio (Z=-4.479,P<0.05), triiodothyronine (Z=-3.338,P<0.05), free triiodothyronine (FT3) (Z=-9.560,P<0.05), and portal blood flow velocity (Z=-4.568,P<0.05). The multivariate logistic regression analysis was performed for the variables with statistical significance identified by the LASSO regression analysis, and the results showed that age (odds ratio[OR]=1.046,95% confidence interval[CI]:1.026 — 1.066), CD4+/CD8+ cell ratio (OR=0.568,95%CI:0.410 — 0.787),FT3(OR=0.956,95%CI:0.944 — 0.968), IL-10 (OR=1.021,95%CI:1.001 — 1.042), diameter of the portal vein (OR=1.446,95%CI:1.329 — 1.574), and spleen thickness (OR=1.035,95%CI:1.014 — 1.055) were independent influencing factors. A model was established as Logit(P)=-8.784+0.045×age-0.566×CD4+/CD8+-0.046×FT3+0.021×IL-10+0.369×diameter of the portal vein+0.034×spleen thickness, and a nomogram model was established and validated based on this model, with an AUC of 0.859 (95%CI:0.833 — 0.887). The Hosmer-Lemeshow test showed that the model had a high goodness of fit (χ²=11.349,P=0.183). Bootstrap internal validation showed a mean absolute error of 0.006 and a C-index of 0.855. The decision curve analysis showed that the model had a high net clinical benefit within a wide range of thresholds. Conclusion Age, CD4+/CD8+ ratio, FT3, IL-10, diameter of the portal vein, and spleen thickness may be independent influencing factors for PVT in patients with decompensated HBV/HCV cirrhosis. The predictive model established based on these six variables can help to predict the risk of PVT in patients with hepatitis-related decompensated cirrhosis in the early stage in clinical practice.
TIAN RH, WANG YZ, WEI HY, et al. Establishment and validation of a nomogram model for patients with decompensated HBV/HCV cirrhosis comorbid with portal vein thrombosis[J]. J Clin Hepatol, 2025, 41(8): 1579-1588.. doi: 10.12449/JCH250817.