Objective Observe the efficacy and safety of glucocorticoid deescalation combined antiviral therapy on severe hepati‐tis B .Methods Forty‐nine patients with early severe hepatitis were selected and divided them into observation group(n=24) and control group (n=25) randomly .Control group were given the conventional synthetic therapy while observation group were given conventional synthetic therapy combined glucocorticoid deescalation therapy .Then compare the differences of TBIL ,ALT ,PTA be‐fore and after treatment ,and the improvement time of syndrome and death rate between two groups .Results The death rate in ob‐servation group and control group were 16 .7% and 32 .0% respectively(P<0 .05) ,the improvement time of syndrome in observa‐tion group were shorten than control group(P<0 .05) ,and there was no statistical difference in bleeding and infection rate between two groups(P>0 .05) .After the treatment ,the TBIL ,ALT ,PT were improved in both two groups(P<0 .05) .But the observation group had better improvement than control group ,the difference had statistical significance(P<0 .05) .After treatment ,the changes of HBV DNA between two groups showed no significant difference(P> 0 .05) .Conclusion Glucocorticoid deescalation therapy could decrease the death rate ,shorten the syndrome lasting time in treating severe hepatitis B .It was an effective and safety meth‐od .

Objective:To evaluate the effect of different options of preoperative biliary drainage (PBD) on perioperative complications of patients undergoing pancreaticoduodenectomy (PD).Methods:The clinical data of patients undergoing PD for periampullary carcinoma from January 2016 to November 2021 at Third Affiliated Hospital of Naval Medical University (Shanghai Eastern Hepatobiliary Surgery Hospital) were retrospectively analyzed. The 303 patients including 199 males and 104 females, aged (64.2±8.8) years. According to PBD, the patients were divided into two groups: percutaneous transhepatic biliary drainage (PTBD) group ( n=228) and endoscopic retrograde cholangiopancreatography (ERCP) group ( n=75). PBD operation-related complications (including bleeding, biliary leakage, etc.), postoperative complications of PD (including pancreatic fistula, biliary leakage, surgical site infection, etc.) and perioperative complications (PBD operation-related complications + postoperative complications of PD) were compared between the two groups. Univariate and multivariate logistic regression analysis were used to analyze factors influencing perioperative complications of PD. Results:The incidence of PBD operation-related complications in PTBD group was 10.1% (23/228), lower than that in ERCP group 25.3%(19/228), and the difference was statistically significant (χ 2=10.99, P=0.001). The incidence of postoperative complications of PD in PTBD group was 38.2%(87/228), lower than that in ERCP group 69.3%(52/75), the difference was statistically significant (χ 2=22.09, P<0.001). The incidence of total perioperative complications in PTBD group was 44.3% (101/228), lower than that in ERCP group 73.3%(55/75), the difference was statistically significant (χ 2=19.05, P<0.001). Multivariate logistic regression analysis showed that patients with periampullary carcinoma undergoing ERCP biliary drainage and PD had increased risk of surgical site infection ( OR=2.86, 95% CI: 1.59-5.16, P<0.001) and pancreatic fistula ( OR=3.06, 95% CI: 1.21-7.74, P=0.018). Conclusion:ERCP biliary drainage is a risk factor for postoperative pancreatic fistula and surgical site infection in patients with periampullary carcinoma undergoing PD. PTBD should be recommended as the first choice for the patients underwent PD.

Objective:To study the risk factors of lymph node metastases in patients with intrahepatic cholangiocarcinoma (ICC) and to establish a risk prediction model of lymph node metastases in ICC.Methods:The clinicopathological data of 587 ICC patients who underwent radical hepatectomy and lymph node dissection at Third Affiliated Hospital of Naval Medical University (Shanghai Eastern Hepatobiliary Surgery Hospital) from January 2007 to December 2011 were retrospectively analyzed. There were 395 males and 192 females with ages which ranged from 20 to 82 (54.7±10.8) years. Independent risk factors of lymph node metastases were studied using univariate and multivariate logistic regression analysis, and a risk prediction model was established. Receiver operating characteristic (ROC) curve was used to evaluate the accuracy of this model.Results:Of 587 patients, 158 (26.9%) had lymph node metastases. Multivariate logistic regression analysis showed that platelet count >300×10 9/L ( OR=1.985, 95% CI: 1.030-3.824, P=0.041), carbohydrate antigen 19-9 >37 U/ml ( OR=2.978, 95% CI: 1.994-4.448, P<0.001), tumor situated in left hemiliver ( OR=1.579, 95% CI: 1.065-2.341, P=0.023), multiple tumors ( OR=1.846, 95% CI: 1.225-2.783, P=0.003), and absence of cirrhosis ( OR=2.125, 95% CI: 1.192-3.783, P=0.011) were independent risk factors for lymph node metastases in ICC. The area under the ROC curve was 0.714, with a cutoff value of 0.215, and the sensitivity and specificity being 75.9% and 58.3%, respectively. Conclusions:The risk prediction model of ICC lymph node metastases was established using readily available clinical data obtained before operation. This model has good predictive values and can provide a reference for treatment decision on patients with ICC.