1.Distribution characteristics of pathogens and influencing factors analysis of infections within 90 days after liver transplantation
Huabin PENG ; Haofeng XIONG ; Fei HOU ; Shuang ZHAO ; Yizhi ZHANG ; Tingting CUI ; Zhiying HE ; Jingyi LIU ; Liying SUN
Organ Transplantation 2026;17(2):212-226
Objective To investigate the distribution characteristics of pathogens causing infections within 90 days after liver transplantation and the influencing factors of infection. Methods Clinical data of 176 recipients who underwent liver transplantation at the Liver Transplant Center of Beijing Friendship Hospital Affiliated to Capital Medical University from September 2021 to August 2024 were retrospectively analyzed. Patients were divided into the infection group (n=124) and the non-infection group (n=52) based on whether they developed infection within 90 days after transplantation. The distribution characteristics of pathogens in infected patients were analyzed. Univariate and multivariate logistic regression analyses were used to explore the influencing factors of infection. Results Among the 176 liver transplant recipients, 124 cases developed 243 episodes of 518 bacterial, fungal, viral or mycoplasma infections within 90 days after transplantation, with an overall infection rate of 70.5% (124/176). The composition of pathogens was mainly Gram-negative bacteria (38.6%, 200/518), followed by Gram-positive bacteria (32.2%, 167/518) and viruses (15.4%, 80/518), and fungi accounted for 13.1% (68/518). Among Gram-negative bacteria, the main pathogen was Klebsiella pneumoniae (6.8%, 35/518), and among Gram-positive bacteria, the main pathogen was Enterococcus faecalis (8.5%, 44/518). Viruses included Epstein-Barr virus (3.7%, 19/518) and cytomegalovirus (3.7%, 19/518), and fungi were mainly Candida albicans (6.8%, 35/518). The most common infection site among the 243 episodes was pulmonary infection (42.0%, 102/243), followed by abdominal infection (22.6%, 55/243) and bloodstream infection (18.1%, 44/243). The infections mainly occurred within 2 weeks after transplantation (60.9%, 148/243). Multivariate logistic regression analysis indicated that preoperative infection within 2 weeks, a high preoperative model for end-stage liver disease (MELD) score, and preoperative sarcopenia were independent risk factors for infection within 90 days after liver transplantation (all odds ratio>1, P<0.05). After multivariate correction, the levels of CD4+T cells and CD8+T cells within 90 days after surgery were independently associated with the occurrence of infection. Low levels of CD4+T cells and CD8+T cells might be related to an increased risk of infection. Conclusions The infection rate after liver transplantation is high, and the pathogens are mainly Gram-negative bacteria. The lungs are the most common infection site. Preoperative MELD score, preoperative sarcopenia and preoperative infection within 2 weeks are independent risk factors for infection within 90 days after liver transplantation. Regular monitoring of immune indicators CD4+T cells and CD8+T cells levels after transplantation is helpful to reduce the occurrence of post-transplantation infection.
2.Huazhuo Jiedu Prescription Treats Ulcerative Colitis by Inhibiting Excessive Mitophagy via PINK1/Parkin Signaling Pathway
Haofeng ZHANG ; Jinye ZHOU ; Ziwei LIU ; Yican WANG ; Yirui CHENG ; Zheng ZHI ; Qian YANG ; Bolin LI
Chinese Journal of Experimental Traditional Medical Formulae 2025;31(11):182-189
ObjectiveTo investigate the mechanism of Huazhuo Jiedu prescription in treating ulcerative colitis (UC) by regulating mitophagy. MethodsThe genes related to mitophagy and UC were retrieved from GeneCards, and then the common genes of mitophagy and UC were analyzed by metascape to identify the genes related to mitophagy in UC. Animal experiments were carried out to decipher the mechanism by which Huazhuo Jiedu prescription treated UC by regulating mitophagy. Sixty C57BL/6 male mice were randomized into normal, model, high-, medium-, and low-dose (50, 25, 12.5 g·kg-1, respectively) Huazhuo Jiedu prescription, and mesalazine (0.52 g·kg-1·d-1) groups, with 10 mice in each group. After successful modeling by the dextran sulfate sodium-free drinking method, the colonic mucosal damage was observed by hematoxylin-eosin staining, and the ultracellular structure of colon mucosa was observed by transmission electron microscopy. The expression levels of mitophagy-related proteins PTEN-induced putative kinase 1 (PINK1) and Parkin protein were determined by Western blot. The expression of prohibitin 2 (PHB2), ubiquitin-specific protease 15 (USP15), ubiquitin-specific protease 30 (USP30) in the colon tissue was detected by immunofluorescence (IF). ResultsAll the drug intervention groups showed ameliorated pathological manifestations of the colonic mucosa and improved mitochondrial structures in UC mice. Compared with the normal group, the model group demonstrated up-regulated protein levels of PINK1 and Parkin (P<0.05), enhanced average fluorescence intensity of PHB2 (P<0.05), and weakened average fluorescence intensity of USP15 and USP30 (P<0.05). Compared with the model group, the mesalazine group and the high- and medium-dose Huazhuo Jiedu prescription groups showcased down-regulated protein levels of PINK1 and Parkin (P<0.05), decreased average fluorescence intensity of PHB2 (P<0.05), and enhanced average fluorescence intensity of USP15 and USP30 (P<0.05). The low-dose Huazhuo Jiedu prescription group showed down-regulated protein levels of PINK1 and Parkin (P<0.05), weakened average fluorescence intensity of PHB2 (P<0.05), and enhanced average fluorescence intensity of USP15 and USP30 (P<0.05). ConclusionHuazhuo Jiedu prescription can attenuate the intestinal mucosal injury and improve the mitochondrial cell ultrastructure in UC mice by regulating the expression of PINK1-Parkin pathway and inhibiting excessive mitophagy.
3.Huazhuo Jiedu Prescription Treats Ulcerative Colitis by Inhibiting Excessive Mitophagy via PINK1/Parkin Signaling Pathway
Haofeng ZHANG ; Jinye ZHOU ; Ziwei LIU ; Yican WANG ; Yirui CHENG ; Zheng ZHI ; Qian YANG ; Bolin LI
Chinese Journal of Experimental Traditional Medical Formulae 2025;31(11):182-189
ObjectiveTo investigate the mechanism of Huazhuo Jiedu prescription in treating ulcerative colitis (UC) by regulating mitophagy. MethodsThe genes related to mitophagy and UC were retrieved from GeneCards, and then the common genes of mitophagy and UC were analyzed by metascape to identify the genes related to mitophagy in UC. Animal experiments were carried out to decipher the mechanism by which Huazhuo Jiedu prescription treated UC by regulating mitophagy. Sixty C57BL/6 male mice were randomized into normal, model, high-, medium-, and low-dose (50, 25, 12.5 g·kg-1, respectively) Huazhuo Jiedu prescription, and mesalazine (0.52 g·kg-1·d-1) groups, with 10 mice in each group. After successful modeling by the dextran sulfate sodium-free drinking method, the colonic mucosal damage was observed by hematoxylin-eosin staining, and the ultracellular structure of colon mucosa was observed by transmission electron microscopy. The expression levels of mitophagy-related proteins PTEN-induced putative kinase 1 (PINK1) and Parkin protein were determined by Western blot. The expression of prohibitin 2 (PHB2), ubiquitin-specific protease 15 (USP15), ubiquitin-specific protease 30 (USP30) in the colon tissue was detected by immunofluorescence (IF). ResultsAll the drug intervention groups showed ameliorated pathological manifestations of the colonic mucosa and improved mitochondrial structures in UC mice. Compared with the normal group, the model group demonstrated up-regulated protein levels of PINK1 and Parkin (P<0.05), enhanced average fluorescence intensity of PHB2 (P<0.05), and weakened average fluorescence intensity of USP15 and USP30 (P<0.05). Compared with the model group, the mesalazine group and the high- and medium-dose Huazhuo Jiedu prescription groups showcased down-regulated protein levels of PINK1 and Parkin (P<0.05), decreased average fluorescence intensity of PHB2 (P<0.05), and enhanced average fluorescence intensity of USP15 and USP30 (P<0.05). The low-dose Huazhuo Jiedu prescription group showed down-regulated protein levels of PINK1 and Parkin (P<0.05), weakened average fluorescence intensity of PHB2 (P<0.05), and enhanced average fluorescence intensity of USP15 and USP30 (P<0.05). ConclusionHuazhuo Jiedu prescription can attenuate the intestinal mucosal injury and improve the mitochondrial cell ultrastructure in UC mice by regulating the expression of PINK1-Parkin pathway and inhibiting excessive mitophagy.
4.Effect of Xianglian Huazhuo Prescription on Hedgehog Signaling Pathway in Rats with Chronic Atrophic Gastritis
Jinye ZHOU ; Haofeng ZHANG ; Ziwei LIU ; Yican WANG ; Yanru CAI ; Yuxi GUO ; Jie WANG ; Zheng ZHI ; Qian YANG ; Bolin LI
Chinese Journal of Experimental Traditional Medical Formulae 2025;31(18):41-47
ObjectiveTo explore the therapeutic effect and mechanism of Xianglian Huazhuo prescription on chronic atrophic gastritis (CAG) in rats based on the Hedgehog signaling pathway. MethodsThe CAG rat model was established by sodium salicylate, N-methyl-N′-nitro-N-nitroguanidine (MNNG), and irregular feeding. The successfully modeled rats were randomly divided into a model group (180 mg·L-1), a moradan group (1.4 g·kg-1), and Xianglian Huazhuo Prescription groups with high, medium, and low doses (36, 9, 18 g·kg-1), followed by drug intervention. Hematoxylin-eosin (HE) staining was used to observe morphological changes in the gastric mucosa. Transmission electron microscopy was used to observe the ultrastructure of gastric mucosa cells. Real-time quantitative polymerase chain reaction (Real-time PCR) was used to detect the mRNA expression of Sonic Hedgehog (Shh), Patched 1 (Ptch1), and Glioma-associated oncogene homolog 1 (Gli1). Western blot was used to detect the protein expression levels of Shh, Ptch1, and Gli1 in the gastric mucosa. Immunohistochemistry was used to observe the protein expression of the epithelial marker E-cadherin. ResultsCompared with the normal group, the CAG model group showed a reduction in gastric mucosal intrinsic glands and infiltration of inflammatory cells. The ultrastructure of gastric mucosal cells showed nuclear pyknosis, fewer mitochondria, and abnormal mitochondrial structure. The mRNA and protein expression of Shh, Ptch1, and Gli1 in the gastric mucosa were significantly decreased (P<0.05), and E-cadherin protein expression was decreased. Compared with the model group, the intervention groups showed varying degrees of improvement in histopathological morphology and cellular ultrastructure. The mRNA and protein expression of Shh, Ptch1, Gli1, and E-cadherin increased to varying degrees. Xianglian Huazhuo Prescription upregulated the expression of key Hedgehog pathway factors and E-cadherin at both the mRNA and protein levels (P<0.05). ConclusionXianglian Huazhuo prescription has a therapeutic effect on CAG in rats, and its mechanism may be related to activation of the Hedgehog signaling pathway and inhibition of epithelial-mesenchymal transition (EMT).
5.Predictive value of preoperative combined detection of NLR and PTAR for early abdominal infection after liver transplantation
Huabin PENG ; Ying LIU ; Fei HOU ; Shuang ZHAO ; Yizhi ZHANG ; Tingting CUI ; Zhiying HE ; Jingyi LIU ; Haofeng XIONG ; Liying SUN
Organ Transplantation 2025;16(6):931-943
Objective To investigate the predictive value of preoperative combined detection of neutrophil-to-lymphocyte ratio (NLR) and prothrombin time-international normalized ratio to albumin ratio (PTAR) for early abdominal infection after liver transplantation. Methods Clinical data of 287 recipients who underwent liver transplantation at the Liver Transplant Center of Beijing Friendship Hospital, Affiliated to Capital Medical University, from January 2020 to April 2024 were retrospectively analyzed. The patients were divided into infection group (n=60) and non-infection group (n=227) based on whether abdominal infection occurred within 30 days after surgery. The distribution characteristics of pathogens and infection time in infected patients were analyzed. Spearman correlation analysis was used to assess the correlation between NLR, PTAR, Child-Pugh score and preoperative model for end-stage liver disease (MELD) score. Univariate and multivariate logistic regression analyses were performed to identify risk factors for abdominal infection. Receiver operating characteristic (ROC) curves were plotted for NLR, PTAR, and the combined prediction model to evaluate their predictive efficacy for abdominal infection after liver transplantation. Based on the cutoff value of the combined model, recipients were divided into low-risk and high-risk groups, and Kaplan-Meier analysis was used to compare the cumulative incidence of abdominal infection within 30 days after surgery between the two groups. Results Among the 287 recipients who underwent liver transplantation, 60 developed bacterial or fungal abdominal infections postoperatively. A total of 86 strains were isolated from infected patients, with Gram-negative bacteria accounting for 58%, Gram-positive bacteria for 36%, and fungi for 5%. Preoperative NLR and PTAR were positively correlated with Child-Pugh and MELD scores (all 1 > r > 0, P < 0.05). Logistic regression analysis showed that preoperative NLR, preoperative PTAR, postoperative ICU stay duration and postoperative biliary leakage were risk factors for abdominal infection within 30 days after surgery. The area under the curve (AUC) for NLR, PTAR, Child-Pugh score and MELD score were 0.771, 0.735, 0.650 and 0.741, respectively. The AUC for the combined NLR and PTAR prediction model was 0.824 (95% confidence interval: 0.763-0.885, P < 0.001), with a cutoff value of 0.168. Kaplan-Meier analysis showed that the cumulative incidence of abdominal infection within 30 days after surgery was lower in the low-risk group than in the high-risk group, with statistically significant difference (P < 0.001). Conclusions Preoperative NLR and PTAR are independent risk factors for abdominal infection within 30 days after liver transplantation. The combined prediction model of NLR and PTAR may effectively identify high-risk recipients for early abdominal infection after liver transplantation, providing basis for early intervention.
6.Regulation of histidine metabolism by Lactobacillus Reuteri mediates the pathogenesis and treatment of ischemic stroke.
Kangrui HU ; Zhihao ZHOU ; Haofeng LI ; Jijun XIAO ; Yun SHEN ; Ke DING ; Tingting ZHANG ; Guangji WANG ; Haiping HAO ; Yan LIANG
Acta Pharmaceutica Sinica B 2025;15(1):239-255
Increasing evidence has underscored the significance of post-stroke alterations along gut-brain axis, while its role in pathogenesis and treatment of ischemic stroke (IS) remains largely unexplored. This study aimed to elucidate the therapeutic effects and action targets of Panax notoginseng saponins (PNS) on IS and explore a novel pathogenesis and treatment strategy of IS via profiling the microbial community and metabolic characteristics along gut-brain axis. Our findings revealed for the first time that the therapeutic effect of PNS on IS was microbiota-dependent. Ischemia/reperfusion (I/R) modeling significantly down-regulated Lactobacilli in rats, and PNS markedly recovered Lactobacilli, particularly Lactobacillus reuteri (L.Reu). Metabolomics showed a significant reduction in serum histidine (HIS) in clinical obsolete IS patients and rehabilitation period I/R rats. Meanwhile, the L.Reu colonization in I/R rats exhibited significant neuroprotective activity and greatly increased HIS in serum, gut microbiota, and brain. Moreover, exogenous HIS demonstrated indirect neuroprotective effects through metabolizing to histamine. Notably, vagus nerve severance in I/R rats was performed to investigate HIS's neuroprotective mechanism. The results innovatively revealed that PNS could promote HIS synthesis in gut by enhancing L.Reu proportion, thereby increasing intracerebral HIS through peripheral pathway. Consequently, our data provided novel insights into HIS metabolism mediated by L.Reu in the pathogenesis and treatment of IS.
7.Impact of atmospheric particulate matter on daily hospital admissions of patients with chronic kidney disease:a time series analysis
Ruikai WU ; Ying ZHANG ; Long MA ; Haofeng YANG ; Deqi SU
Academic Journal of Naval Medical University 2025;46(6):775-783
Objective To investigate the correlation and lag effect between atmospheric particulate matter and the risk of hospitalization for chronic kidney disease(CKD).Methods The daily hospitalization data for CKD in 9 hospitals in Urumqi from Jan.1,2019,to Dec.31,2020,and the air pollution and meteorological data during the same period were collected.The relationship between PM2.5 and PM10 concentrations and CKD incidence was analyzed after controlling for long-term trends,meteorological factors,and potential confounding factors such as the"day of the week effect"by using the generalized additive model(GAM).The effects of single-day lag of 0-7 d(lag0-lag7)and cumulative lag of 0-7 d(lag01-lag07)were analyzed,and subgroup analyses were conducted for gender,age,and season.On the basis of the single pollutant model,other pollutants were included(at most 2 pollutants were included at a time),and a double pollutant model was constructed to evaluate the stability of the model.Results For every 10 μg/m3 increase in PM2.5 concentration,the highest risk of CKD hospitalization occured when lagged alone at lag2(relative risk[RR]=1.014,95%confidence interval[CI]1.006-1.023)and lagged cumulatively at lag04(RR=1.018,95%CI 1.007-1.029).For every 10 μg/m3 increase in PM10 concentration,the risk of CKD hospitalization was highest when lagged alone at lag0 and lagged cumulatively at lag07(RR=1.012,95%CI 1.007-1.017;RR=1.024,95%CI 1.016-1.032).In gender stratification,for every 10 μg/m3 increase in PM2.5 concentration,the cumulative lag at lag04 indicated that males had the highest risk of CKD hospitalization(RR=1.023,95%CI 1.008-1.038);for every 10 μg/m3 increase in PM10 concentration,the highest risk of CKD hospitalization was observed in males when lagged alone at lag0(RR=1.013,95%CI 1.006-1.020),and in females when lagged alone at lag1(RR=1.013,95%CI 1.006-1.020).In age stratification,for every 10 μg/m3 increase in PM2.5 concentration,the risk of CKD hospitalization was highest in people 65 years old with single-day lag at lag3 and cumulative lag at lag04(RR=1.016,95%CI 1.007-1.026;RR=1.022,95%CI 1.010-1.035);for every 10 μg/m3 increase in PM10 concentration,the cumulative lag at lag07 indicated that individuals aged<65 years old and ≥65 years old had the highest risk of CKD hospitalization(RR=1.027,95%CI 1.017-1.037;RR=1.015,95%CI 1.001-1.028).In seasonal stratification,for every 10 μg/m3 increase in PM2.5 concentration during the cold season,the risk of CKD hospitalization was highest when lagged alone at lag3 and lagged cumulatively at lag07(RR=1.020,95%CI 1.011-1.029;RR=1.025,95%CI 1.011-1.038).For every 10 μg/m3 increase in PM10 concentration during the cold season,the risk of CKD hospitalization was highest when lagged alone at lag2(RR=1.013,95%CI 1.007-1.019).For every 10 μg/m3 increase in PM10 concentration during the warm season,the risk of CKD hospitalization was highest when lagged alone at lag7(RR=1.015,95%CI 1.006-1.024).In the dual pollutant model,the effects of PM2.5 adjusting PM10,SO2,O3 and CO,and PM10 adjusting NO2,SO2,O3,and CO on the risk of CKD hospitalization were still significant(P<0.05).Conclusion The increase in atmospheric particulate matter concentrations of PM2.5 and PM10 can lead to an increased risk of CKD,and there is a lag effect.Men,people under the age of 65 years old,and those in cold seasons(heating periods)are more sensitive to exposure to PM2.5 and PM10.
8.Transplantation of menstrual blood-derived endometrial stem cells allevi-ates chemotherapy-induced intestinal injury and flora imbalance in mice through immunomodulation
Qin LIU ; Mengyuan CHANG ; Haofeng SONG ; Chenxu DU ; Yonghai LI ; Shenghui ZHANG ; Yanli LIU ; Juntang LIN
Chinese Journal of Pathophysiology 2024;40(2):301-308
AIM:To investigate the therapeutic effect of menstrual blood-derived endometrial stem cells(MenSCs)on chemotherapy-induced intestinal mucositis and flora disorders in mice,and to explore the potential mecha-nism.METHODS:The mice were randomly divided into 3 groups including normal treatment,cisplatin(Cis)treatment and Cis+MenSC treatment,with 10 mice in each group.To induce intestinal mucositis,the mice were treated with Cis(2 mg·kg-1·d-1)by intraperitoneal injection for 5 consecutive days.Control mice for normal group were received equal vol-umes of normal saline.For Cis+MenSC treatment,MenSCs(1×106)was transplanted into the mice of Cis treated mice through tail vein.The performances and weight changes of mice were examined during the experiment.After the treat-ment,the small intestine and colon were isolated for subsequent HE staining,the ratio of F4/80 and IL-6 positive cells in small intestine were detected by immunohistochemical staining,and the expression of tight junction,inflammation and apoptosis related proteins was detected by Western blot.16S rDNA amplicon sequencing was performed to detect the diver-sity and richness of intestinal flora in mice.RESULTS:Compared to the Cis group,the MenSCs-treated mice showed sig-nificantly increased body weight,relieved intestinal lymphocytes infiltration,alleviated intestinal villous edema,and or-derly arranged glands in intestinal tissues.Further analysis indicated that MenSCs transplantation significantly up-regulat-ed the expression of intestinal tight junction related proteins ZO-1 and occludin in Cis-treated mice(P<0.05).Subse-quently,MenSCs transplantation significantly inhibited the macrophages infiltration in intestinal tissues(P<0.01),down-regulated the expression of pro-inflammatory factors IL-1 and IL-6 and pro-apoptotic protein Bax(P<0.01),while up-regu-lated anti-inflammatory factor IL-10 and anti-apoptotic protein Bcl-2(P<0.01).Additionally,further microflora sequenc-ing indicated that MenSCs transplantation prevented mice from Cis-induced intestinal flora disorder,and significantly re-duced the abundance of harmful bacteria such as isenbergiella tayi and Anaerotruncus colihominis(P<0.01).At the same time,the abundance of beneficial bacteria Lactobacillus apodemi was increased(P<0.05),thereby restoring the composi-tion and function of healthy intestinal flora.CONCLUSION:MenSCs transplantation alleviates the chemotherapy-in-duced damage of intestinal structure,relieves the symptoms of chemotherapy-induced mucositis and restores the homeosta-sis of intestinal flora in mice.
9.Risk factors and prognosis of recurrence within 6 months after radical resection of intrahepatic cholangiocarcinoma
Zhenwei YANG ; Pengyu CHEN ; Hao YUAN ; Zuochao QI ; Guan HUANG ; Haofeng ZHANG ; Bo MENG ; Xianzhou ZHANG ; Haibo YU
Chinese Journal of General Surgery 2024;39(2):99-104
Objective:To explore the relevant risk factors and prognosis of patients with intrahepatic cholangiocarcinoma (ICC) who experienced recurrence within 6 months after surgeryMethods:This retrospective study included a total of 259 patients with ICC a treated at He'nan Provincial People's Hospital and He'nan Cancer Hospital from Jan 2018 to Jan 2020. The clinical and pathological data ,differences between the group with recurrence within 6 months and the group without recurrence within 6 months were compared using the chi-square test. Logistic regression analysis was used to determine the relevant risk factors for recurrence within 6 months. Kaplan-Meier method was used to construct survival and recurrence curves, and survival rates were calculated.Results:The overall survival and recurrence-free survival of patients in the group with recurrence within 6 months were significantly shorter. CA19-9, tumor longitudinal diameter, microvascular invasion, and neural invasion were identified as independent risk factors for recurrence within 6 months after ICC surgery ( P<0.001). Conclusion:The patient population experiencing recurrence within 6 months after ICC surgery has an extremely poor prognosis and possesses a specific tumor microenvironment. CA19-9, tumor longitudinal diameter, microvascular invasion, and neural invasion were identified as independent risk factors for recurrence within 6 months after ICC surgery.
10.Construction and validation of a machine learning model for preoperative prediction of perineural invasion status in intrahepatic cholangiocarcinoma
Zuochao QI ; Zhenwei YANG ; Qingshan LI ; Hao YUAN ; Pengyu CHEN ; Haofeng ZHANG ; Yanbo WANG ; Dongxiao LI ; Bo MENG ; Haibo YU ; Deyu LI
Chinese Journal of Hepatobiliary Surgery 2024;30(6):424-430
Objective:To construct and validate a machine learning model for preoperative prediction of perineural invasion (PNI) status in intrahepatic cholangiocarcinoma (ICC).Methods:Clincial data of 329 patients, including 245 admitted to Zhengzhou University People's Hospital from January 2018 to June 2023 and 84 admitted to the Affiliated Cancer Hospital of Zhengzhou University from January 2013 to January 2020 were retrospectively analyzed. Patients were divided into a training set ( n=231) and a validation set ( n=98). Clinicopathological data including age, gender, hepatitis B virus (HBV) infection status were collected. Predictive variables were determined using least absolute shrinkage and selection operator (LASSO) regression analysis. Six machine learning algorithms including random forest (RF), logistic regression, and linear kernel-based support vector machine were selected to construct the preoperative prediction model for PNI in ICC. Performance metrics of the model were calculated using a confusion matrix, and the final model was selected. The model performance was evaluated in the validation set. Calibration curves were plotted to evaluate the final model, and a Pareto chart was used to visualize the importance of predictive variables. Results:LASSO regression identified nine predictive variables included in the prediction model, including carbohydrate antigen 19-9 (CA19-9), HBV infection status, alkaline phosphatase, alanine aminotransferase, prothrombin time, total bilirubin, albumin, neutrophil times gamma-glutamyl transferase to lymphocyte ratio, and tumor burden score. Among the trained six models, the area under the curve (AUC) of the RF model was 0.909, with a sensitivity of 0.842 and an accuracy of 0.870. Compared with the AUC of the RF model, the AUCs of the other 5 models were lower (all P<0.05). The AUC of the RF model for predicting PNI in ICC in validation set was 0.736. Calibration curves showed good fit of the RF model's prediction of PNI in ICC in both training and validation sets. The Pareto chart showed that CA19-9 was the most important predictive variable in the model, followed by HBV infection status. Conclusion:The machine learning model based on the RF algorithm has a high accuracy in preoperative prediction of PNI status in ICC.

Result Analysis
Print
Save
E-mail