Objective:To compare the efficacy of surgical resection and radiofrequency ablation (RFA) treatment for China liver cancer staging (CNLC) Ia hepatocellular carcinoma (HCC) at difficult-to-reach locations.Methods:A retrospective analysis was conducted on the clinical data of 114 patients with CNLC Ia HCC at Ⅶ、Ⅷ、Ⅳb or Ⅰ segments that were difficult-to-reach locations who were admitted to People's Hospital of Zhengzhou University from December 2018 to December 2023. Among the patients, 85 were males and 29 were females, aged (58.1±1.0) years. The patients were divided into two groups: a RFA group with 31 cases and a surgical resection group with 83 cases. Compare the levels of alanine transaminase (ALT) and aspartate transaminase (AST) before and after surgery, the surgical time, intraoperative blood loss, postoperative hospital stay, postoperative complications, recurrence free survival rate, and cumulative survival rate between the two groups.Results:The comparison of age, gender, ALT, and AST between the two groups showed no statistically significant differences (all P>0.05). The differences in ALT and AST levels before and after surgery in the RFA group were (134.8±38.7) U/L and (195.1±53.9) U/L, respectively, which were significantly lower than those in the surgical resection group [(226.8±17.9) U/L and (229.5±16.2) U/L] ( t=-2.45 and -1.12, P=0.016 and 0.041). The RFA group had shorter operation time [(69.2±11.7) min vs. (210.6±8.9) min], less intraoperative blood loss [(8.7±3.8) ml vs. (238.6±20.8) ml], and shorter postoperative hospital stays [(6.4±1.0) d vs. (13.1±0.4) d] compared to the surgical resection group, with all differences statistically significant (all P<0.05). The overall complication rates were 19.4% (6/31) in the RFA group and 22.9% (19/83) in the surgical resection group, showing no significant difference ( χ2=0.16, P=0.685). No statistically significant diffe-rence was found in recurrence-free survival rates between the two groups ( χ2=0.13, P=0.717). Similarly, there was no statistically significant difference in cumulative survival rates between the groups ( χ2<0.01, P=0.978). Conclusion:For HCC at CNLC Ⅰa in challenging locations, RFA demonstrated shorter operation time and postoperative hospital stay, less intraoperative bleeding, and superior liver function recovery compared with surgical resection, while no significant difference was observed in survival outcomes between the two treatment groups.

Objective:To analyze the application value of three-dimensional visualization combined with indocyanine green fluorescence staining in laparoscopic resection of patients with complex liver cancer.Methods:The data of patients with complex liver cancer (liver cancer located in liver segments Ⅶ, Ⅷ, Ⅳa, Ⅴ and caudate lobe) who underwent laparoscopic anatomical liver resection in Zhengzhou University People's Hospital from August 2018 to August 2023 were retrospectively analyzed. A total of 334 patients were enrolled, including 249 males and 85 females, with the age of (57.0±10.6) years. Among the 334 patients, patients who underwent laparoscopic liver resection using three-dimensional visualization combined with indocyanine green fluorescence staining were included in the combined group ( n=128), and the other patients who underwent traditional laparoscopic liver resection were included in the traditional group ( n=206). Propensity score was used to match the preoperative indicators and postoperative pathology between the two groups. The two groups were compared in terms of gender, age, maximum diameter of tumor, alpha-fetoprotein, number of tumors, tumor location, American Society of Anesthesiologists (ASA) grade, operation time, portal occlusion time, intraoperative blood loss, surgical margin and postoperative aspartate transaminase (ALT), alanine transaminase (AST), albumin, total bilirubin, and severe postoperative complications (Clavien-Dindo grade≥grade Ⅲ). The Kaplan-Meier method was used to analyze the prognosis, and the log-rank test was used to compare the survival rate. Results:After propensity score matching, 120 cases were included in the combined group and the traditional group, respectively. There were no significant differences in gender, age, maximum diameter of tumor, alpha-fetoprotein, number of tumors, tumor location, and ASA grade between the two groups (all P>0.05). The intraoperative blood loss in the combined group was 200.0 (150.0, 300.0) ml, and the positive surgical margin rate was 6.7% (8/120), which were lower than 300.0 (150.0, 500.0) ml and 15.8% (19/120) in the traditional group, with statistically significant differences ( Z=-2.43, P=0.015, χ2=5.05, P=0.025). There were no statistically significant differences in the operation time, portal occlusion time, and postoperative ALT, AST, albumin, and total bilirubin between the two groups (all P>0.05). Eight cases (6.7%) in the combined group and 11 cases (9.2%) in the traditional group had severe complications after surgery, and all were discharged after treatment. The 1-, 3-, and 5-year recurrence-free survival rates of the combined group were 83.3%, 61.7%, and 58.3% respectively, while that of the traditional group were 71.7%, 52.5%, and 49.2%, respectively, with significant difference ( χ2=4.57, P=0.031). There was no significant difference in the cumulative survival rate between the combined group and the traditional group ( χ2=0.66, P=0.417). Conclusion:Compared with laparoscopic liver resection, laparoscopic liver resection for patients with complex liver cancer using three-dimensional visualization technology combined with indocyanine green fluorescence staining technology can reduce intraoperative blood loss, positive margin rate, and postoperative recurrence.

Objective:To compare the efficacy of surgical resection and radiofrequency ablation (RFA) treatment for China liver cancer staging (CNLC) Ia hepatocellular carcinoma (HCC) at difficult-to-reach locations.Methods:A retrospective analysis was conducted on the clinical data of 114 patients with CNLC Ia HCC at Ⅶ、Ⅷ、Ⅳb or Ⅰ segments that were difficult-to-reach locations who were admitted to People's Hospital of Zhengzhou University from December 2018 to December 2023. Among the patients, 85 were males and 29 were females, aged (58.1±1.0) years. The patients were divided into two groups: a RFA group with 31 cases and a surgical resection group with 83 cases. Compare the levels of alanine transaminase (ALT) and aspartate transaminase (AST) before and after surgery, the surgical time, intraoperative blood loss, postoperative hospital stay, postoperative complications, recurrence free survival rate, and cumulative survival rate between the two groups.Results:The comparison of age, gender, ALT, and AST between the two groups showed no statistically significant differences (all P>0.05). The differences in ALT and AST levels before and after surgery in the RFA group were (134.8±38.7) U/L and (195.1±53.9) U/L, respectively, which were significantly lower than those in the surgical resection group [(226.8±17.9) U/L and (229.5±16.2) U/L] ( t=-2.45 and -1.12, P=0.016 and 0.041). The RFA group had shorter operation time [(69.2±11.7) min vs. (210.6±8.9) min], less intraoperative blood loss [(8.7±3.8) ml vs. (238.6±20.8) ml], and shorter postoperative hospital stays [(6.4±1.0) d vs. (13.1±0.4) d] compared to the surgical resection group, with all differences statistically significant (all P<0.05). The overall complication rates were 19.4% (6/31) in the RFA group and 22.9% (19/83) in the surgical resection group, showing no significant difference ( χ2=0.16, P=0.685). No statistically significant diffe-rence was found in recurrence-free survival rates between the two groups ( χ2=0.13, P=0.717). Similarly, there was no statistically significant difference in cumulative survival rates between the groups ( χ2<0.01, P=0.978). Conclusion:For HCC at CNLC Ⅰa in challenging locations, RFA demonstrated shorter operation time and postoperative hospital stay, less intraoperative bleeding, and superior liver function recovery compared with surgical resection, while no significant difference was observed in survival outcomes between the two treatment groups.

Objective:To analyze the application value of three-dimensional visualization combined with indocyanine green fluorescence staining in laparoscopic resection of patients with complex liver cancer.Methods:The data of patients with complex liver cancer (liver cancer located in liver segments Ⅶ, Ⅷ, Ⅳa, Ⅴ and caudate lobe) who underwent laparoscopic anatomical liver resection in Zhengzhou University People's Hospital from August 2018 to August 2023 were retrospectively analyzed. A total of 334 patients were enrolled, including 249 males and 85 females, with the age of (57.0±10.6) years. Among the 334 patients, patients who underwent laparoscopic liver resection using three-dimensional visualization combined with indocyanine green fluorescence staining were included in the combined group ( n=128), and the other patients who underwent traditional laparoscopic liver resection were included in the traditional group ( n=206). Propensity score was used to match the preoperative indicators and postoperative pathology between the two groups. The two groups were compared in terms of gender, age, maximum diameter of tumor, alpha-fetoprotein, number of tumors, tumor location, American Society of Anesthesiologists (ASA) grade, operation time, portal occlusion time, intraoperative blood loss, surgical margin and postoperative aspartate transaminase (ALT), alanine transaminase (AST), albumin, total bilirubin, and severe postoperative complications (Clavien-Dindo grade≥grade Ⅲ). The Kaplan-Meier method was used to analyze the prognosis, and the log-rank test was used to compare the survival rate. Results:After propensity score matching, 120 cases were included in the combined group and the traditional group, respectively. There were no significant differences in gender, age, maximum diameter of tumor, alpha-fetoprotein, number of tumors, tumor location, and ASA grade between the two groups (all P>0.05). The intraoperative blood loss in the combined group was 200.0 (150.0, 300.0) ml, and the positive surgical margin rate was 6.7% (8/120), which were lower than 300.0 (150.0, 500.0) ml and 15.8% (19/120) in the traditional group, with statistically significant differences ( Z=-2.43, P=0.015, χ2=5.05, P=0.025). There were no statistically significant differences in the operation time, portal occlusion time, and postoperative ALT, AST, albumin, and total bilirubin between the two groups (all P>0.05). Eight cases (6.7%) in the combined group and 11 cases (9.2%) in the traditional group had severe complications after surgery, and all were discharged after treatment. The 1-, 3-, and 5-year recurrence-free survival rates of the combined group were 83.3%, 61.7%, and 58.3% respectively, while that of the traditional group were 71.7%, 52.5%, and 49.2%, respectively, with significant difference ( χ2=4.57, P=0.031). There was no significant difference in the cumulative survival rate between the combined group and the traditional group ( χ2=0.66, P=0.417). Conclusion:Compared with laparoscopic liver resection, laparoscopic liver resection for patients with complex liver cancer using three-dimensional visualization technology combined with indocyanine green fluorescence staining technology can reduce intraoperative blood loss, positive margin rate, and postoperative recurrence.

Objective:To investigate changes in the density and spatial distribution of CD69 + T cells within hepatocellular carcinoma tissues following immune checkpoint blockade (ICB) therapy, and to explore their correlation with tumor infiltrating immune cell. Methods:Tumor specimens were collected from 12 patients with hepatocellular carcinoma who were admitted to the Department of Hepatobiliary and Pancreatic Surgery of People's Hospital of Zhengzhou University and the First Affiliated Hospital of Zhengzhou University from July 2023 to July 2024. There were 10 males and 2 females, aged (58.5±5.6) years. Of the 12 patients, 6 cases underwent radical surgery directly and 6 underwent radical surgery after immunotherapy. The maximum tumor diameter and tumor volume of the immunotherapy group were measured by imaging. The density and distribution of immune cells such as CD8 + CD69 + T, CD4 + CD69 + T and programmed death-1 (PD-1) were detected by immunohistochemistry and immunofluorescence. The number of immune cells around the target cells was calculated to evaluate the effective score, and the intercellular distance was measured to evaluate the intercellular interaction. Results:The maximum tumor diameter and tumor volume of 6 patients after immunotherapy were lower than before treatment, and the differences were statistically significant (all P<0.05). The density of PD1 + cells in the immunotherapy group was 36.6 (25.9, 55.9) cells/mm 2, which was less than that in the control group 53.9 (38.3, 84.5) cells/mm 2, and the difference was statistically significant ( Z=-2.66, P=0.008). In the control group, the number of CD8 + CD69 + T cells was positively correlated with CD8 + PD1 + T cells and CD8 + PD1 + CD103 + T cells, and the correlation coefficients were 0.42 and 0.40, respectively ( P=0.001, 0.002). The effective scores of CD8 + CD69 + T cells and CD8 + PD1 + T, CD4 + CD103 + T, CD4 + PD1 + CD103 + T and CD8 + PD1 + CD103 + T cells in the above three areas in the immunotherapy group were lower than those in the control group, with statistical significance (all P<0.05). The distance between CD8 + CD69 + T and CD4 + CD69 + CD103 + T cells in the interface area of the control group was closer than that of the immunotherapy group, and the difference was statistically significant ( t=2.67, P=0.009). Conclusion:After immunotherapy in hepatocellular carcinoma patients, PD-1+ cells and immune cells around CD8 + CD69 + T cells decreased, and this change was related to the distance between CD8 + CD103 + T cells.

Objective:To analyze the risk factors for clinically relevant delayed gastric emptying (CR-DGE) following laparoscopic pancreaticoduodenectomy (LPD) and to develop a model to predict the postoperative CR-DGE after LPD using the machine-learning approach with multi-model comparison.Methods:Clinical data of 278 patients with tumors located in the pancreatic head and periampullary region undergoing LPD at People’s Hospital of Zhengzhou University from January 2019 to December 2023 were retrospectively analyzed, including 167 males and 111 females, aged 59 (53, 66) years. According to the occurrence of DGE, patients were divided into the CR-DGE group ( n=94) and the non-CR-DGE group ( n=184). Main clinical characteristics were compared between the groups, including pancreatic duct diameter, intraoperative blood loss and operative time. The perioperative indicators were selected using the least absolute shrinkage and selection operator (LASSO) algorithm. Following variable selection, 278 patients were allocated into a training set ( n=222) and a validation set ( n=56) in an 8∶2 ratio. Eight machine learning models were selected to model the training set: random forest, adaptive boosting, light gradient boosting, multilayer perceptron, support vector machine, K-nearest neighbor algorithm, decision tree and complementary set plain bayes. The area under the curve (AUC) of receiver operating characteristic curve of the validation set was utilized to identify the optimal model. The predictive performance of the optimal model was evaluated using calibration plots and decision curve analysis (DCA). The contribution of each feature to the prediction is assessed using Shapley additive explanation (SHAP). Results:Univariate analysis showed statistically significant differences between the CR-DGE and non-CR-DGE groups in terms of age [66(62, 69) years vs. 56(51, 60), years], diabetes [42.6%(40/94) vs. 11.4%(21/184)], level of fibrinogen [3.43(2.74, 4.18) g/L vs. 3.84(3.19, 4.68) g/L], pancreatic duct diameter [2.00(1.50, 2.70) mm vs. 3.40(1.60, 5.00) mm], intraoperative blood loss [300(200, 600) ml vs. 200(150, 300) ml], operative time [472(430, 502) min vs. 430(365, 475) min], clinically relevant postoperative pancreatic fistula [34.0%(32/94) vs. 3.8%(7/184)], abdominal fluid accumulation [46.8%(44/94) vs. 12.5%(23/184)], postoperative hemorrhage [20.2%(19/94) vs. 3.3%(6/184)], abdominal infection [28.7%(27/94) vs. 11.4% (21/184)] and duration of postoperative gastrointestinal decompression [4.00 (2.00, 6.00) d vs. 3.00 (2.00, 5.00) d] (all P<0.05). The eleven variables selected via LASSO were incorporated into each of the eight machine learning models. Results demonstrated that the random forest model achieved the highest performance in the validation set, with an AUC of 0.894 (95% CI: 0.800-0.985), accuracy of 0.820 and sensitivity of 0.606. Calibration plots and DCA confirmed the robustness of the random forest model. SHAP analysis indicated that age, pancreatic duct diameter and preoperative aspartate aminotransferase were important predictors in the random forest model. Conclusion:The random forest model developed in this study demonstrated a good predictive performance for CR-DGE after LPD and may assist in the early identification of high-risk patients in clinical practice.

Objective:To investigate changes in the density and spatial distribution of CD69 + T cells within hepatocellular carcinoma tissues following immune checkpoint blockade (ICB) therapy, and to explore their correlation with tumor infiltrating immune cell. Methods:Tumor specimens were collected from 12 patients with hepatocellular carcinoma who were admitted to the Department of Hepatobiliary and Pancreatic Surgery of People's Hospital of Zhengzhou University and the First Affiliated Hospital of Zhengzhou University from July 2023 to July 2024. There were 10 males and 2 females, aged (58.5±5.6) years. Of the 12 patients, 6 cases underwent radical surgery directly and 6 underwent radical surgery after immunotherapy. The maximum tumor diameter and tumor volume of the immunotherapy group were measured by imaging. The density and distribution of immune cells such as CD8 + CD69 + T, CD4 + CD69 + T and programmed death-1 (PD-1) were detected by immunohistochemistry and immunofluorescence. The number of immune cells around the target cells was calculated to evaluate the effective score, and the intercellular distance was measured to evaluate the intercellular interaction. Results:The maximum tumor diameter and tumor volume of 6 patients after immunotherapy were lower than before treatment, and the differences were statistically significant (all P<0.05). The density of PD1 + cells in the immunotherapy group was 36.6 (25.9, 55.9) cells/mm 2, which was less than that in the control group 53.9 (38.3, 84.5) cells/mm 2, and the difference was statistically significant ( Z=-2.66, P=0.008). In the control group, the number of CD8 + CD69 + T cells was positively correlated with CD8 + PD1 + T cells and CD8 + PD1 + CD103 + T cells, and the correlation coefficients were 0.42 and 0.40, respectively ( P=0.001, 0.002). The effective scores of CD8 + CD69 + T cells and CD8 + PD1 + T, CD4 + CD103 + T, CD4 + PD1 + CD103 + T and CD8 + PD1 + CD103 + T cells in the above three areas in the immunotherapy group were lower than those in the control group, with statistical significance (all P<0.05). The distance between CD8 + CD69 + T and CD4 + CD69 + CD103 + T cells in the interface area of the control group was closer than that of the immunotherapy group, and the difference was statistically significant ( t=2.67, P=0.009). Conclusion:After immunotherapy in hepatocellular carcinoma patients, PD-1+ cells and immune cells around CD8 + CD69 + T cells decreased, and this change was related to the distance between CD8 + CD103 + T cells.

Objective:To analyze the risk factors for clinically relevant delayed gastric emptying (CR-DGE) following laparoscopic pancreaticoduodenectomy (LPD) and to develop a model to predict the postoperative CR-DGE after LPD using the machine-learning approach with multi-model comparison.Methods:Clinical data of 278 patients with tumors located in the pancreatic head and periampullary region undergoing LPD at People’s Hospital of Zhengzhou University from January 2019 to December 2023 were retrospectively analyzed, including 167 males and 111 females, aged 59 (53, 66) years. According to the occurrence of DGE, patients were divided into the CR-DGE group ( n=94) and the non-CR-DGE group ( n=184). Main clinical characteristics were compared between the groups, including pancreatic duct diameter, intraoperative blood loss and operative time. The perioperative indicators were selected using the least absolute shrinkage and selection operator (LASSO) algorithm. Following variable selection, 278 patients were allocated into a training set ( n=222) and a validation set ( n=56) in an 8∶2 ratio. Eight machine learning models were selected to model the training set: random forest, adaptive boosting, light gradient boosting, multilayer perceptron, support vector machine, K-nearest neighbor algorithm, decision tree and complementary set plain bayes. The area under the curve (AUC) of receiver operating characteristic curve of the validation set was utilized to identify the optimal model. The predictive performance of the optimal model was evaluated using calibration plots and decision curve analysis (DCA). The contribution of each feature to the prediction is assessed using Shapley additive explanation (SHAP). Results:Univariate analysis showed statistically significant differences between the CR-DGE and non-CR-DGE groups in terms of age [66(62, 69) years vs. 56(51, 60), years], diabetes [42.6%(40/94) vs. 11.4%(21/184)], level of fibrinogen [3.43(2.74, 4.18) g/L vs. 3.84(3.19, 4.68) g/L], pancreatic duct diameter [2.00(1.50, 2.70) mm vs. 3.40(1.60, 5.00) mm], intraoperative blood loss [300(200, 600) ml vs. 200(150, 300) ml], operative time [472(430, 502) min vs. 430(365, 475) min], clinically relevant postoperative pancreatic fistula [34.0%(32/94) vs. 3.8%(7/184)], abdominal fluid accumulation [46.8%(44/94) vs. 12.5%(23/184)], postoperative hemorrhage [20.2%(19/94) vs. 3.3%(6/184)], abdominal infection [28.7%(27/94) vs. 11.4% (21/184)] and duration of postoperative gastrointestinal decompression [4.00 (2.00, 6.00) d vs. 3.00 (2.00, 5.00) d] (all P<0.05). The eleven variables selected via LASSO were incorporated into each of the eight machine learning models. Results demonstrated that the random forest model achieved the highest performance in the validation set, with an AUC of 0.894 (95% CI: 0.800-0.985), accuracy of 0.820 and sensitivity of 0.606. Calibration plots and DCA confirmed the robustness of the random forest model. SHAP analysis indicated that age, pancreatic duct diameter and preoperative aspartate aminotransferase were important predictors in the random forest model. Conclusion:The random forest model developed in this study demonstrated a good predictive performance for CR-DGE after LPD and may assist in the early identification of high-risk patients in clinical practice.

Objective:To establish a machine learning model to predict the risk of post hepatectomy liver dysfunction (PHLD) in patients with hilar cholangiocarcinoma (HCCA).Methods:Clinical data of 203 patients with HCCA undergoing open radical hemihepatectomy in Henan University People's Hospital from January 2017 to December 2023 were retrospectively analyzed, including 112 males and 91 females, aged 63 (55, 69) years. According to the diagnostic criteria for PHLD, patients were divided into two groups: PHLD group ( n=45) and non-PHLD group ( n=158). Clinical data such as age, sex, neutrophil count (NEU), systemic immunoinflammatory index (SII), nutritional prognosis index (PNI), neutrophil to lymphocyte ratio (NLR), operative time and complications were compared between the two groups. The variables with statistically significant difference between the two groups were included in seven machine learning models, namely logistic regression, random forest, extreme gradient boosting, light gradient boosting, decision tree, gaussian naive bayes and support vector machine. The area under receiver operating characteristic curve optimization model was adopted, and Shapliga sum-interpretation method (SHAP) was used to analyze and interpret the final optimal model. Results:There were statistically significant differences in age, preoperative data including management of jaundice, albumin, total bilirubin, aspartate aminotransferase, NEU, SII, PNI, and NLR, operative time, postoperative complication of Dindo-Clavien≥Grade Ⅲ, and the ratio of FLR/TLV between in the two groups (all P<0.05). Finally, it was determined that the prediction performance of the extreme gradient boosting model was the best, with an area under curve of 0.888 (95% CI: 0.776-0.985), an accuracy of 0.854, a sensitivity of 0.506, a specificity of 0.965, an F1 value of 0.625, and a Kappa value of 0.519. SHAP analysis of the extreme gradient boosting model showed that total bilirubin on admission, operation time, postoperative complication of Dindo-Clavien≥grade Ⅲ, SII and NEU were five important factors of this model, which were positively correlated with the occurrence of PHLD in HCCA patients. Conclusion:The extreme gradient boosting model established in this study has a good predictive performance and stability for PHLD in HCCA patients.

Objective:To establish a machine learning model to predict the risk of post hepatectomy liver dysfunction (PHLD) in patients with hilar cholangiocarcinoma (HCCA).Methods:Clinical data of 203 patients with HCCA undergoing open radical hemihepatectomy in Henan University People's Hospital from January 2017 to December 2023 were retrospectively analyzed, including 112 males and 91 females, aged 63 (55, 69) years. According to the diagnostic criteria for PHLD, patients were divided into two groups: PHLD group ( n=45) and non-PHLD group ( n=158). Clinical data such as age, sex, neutrophil count (NEU), systemic immunoinflammatory index (SII), nutritional prognosis index (PNI), neutrophil to lymphocyte ratio (NLR), operative time and complications were compared between the two groups. The variables with statistically significant difference between the two groups were included in seven machine learning models, namely logistic regression, random forest, extreme gradient boosting, light gradient boosting, decision tree, gaussian naive bayes and support vector machine. The area under receiver operating characteristic curve optimization model was adopted, and Shapliga sum-interpretation method (SHAP) was used to analyze and interpret the final optimal model. Results:There were statistically significant differences in age, preoperative data including management of jaundice, albumin, total bilirubin, aspartate aminotransferase, NEU, SII, PNI, and NLR, operative time, postoperative complication of Dindo-Clavien≥Grade Ⅲ, and the ratio of FLR/TLV between in the two groups (all P<0.05). Finally, it was determined that the prediction performance of the extreme gradient boosting model was the best, with an area under curve of 0.888 (95% CI: 0.776-0.985), an accuracy of 0.854, a sensitivity of 0.506, a specificity of 0.965, an F1 value of 0.625, and a Kappa value of 0.519. SHAP analysis of the extreme gradient boosting model showed that total bilirubin on admission, operation time, postoperative complication of Dindo-Clavien≥grade Ⅲ, SII and NEU were five important factors of this model, which were positively correlated with the occurrence of PHLD in HCCA patients. Conclusion:The extreme gradient boosting model established in this study has a good predictive performance and stability for PHLD in HCCA patients.