ObjectiveTo analyze risk factors of rebleeding after laparoscopic selective pericardial devascularization combined with splenectomy.MethodsThe clinical data of 147 cases of portal hypertension treated by the laparoscopic selective devascularization combined with splenectomy from February 2014 to September 2018 were retrospectively analyzed. The univariate analysis was used to find the difference between the rebleeding group (n=20) and the non-bleeding group (n=127), then the multivariate logistic regression analysis was selected for screening out the most closely related risk factors for the rebleeding.ResultsThe rebleeding occurred in 20 of 147 patients (13.6%). There were differences in the platelet count, prothrombin time, serum albumin, diameter of main portal vein, classification of liver function, ascites, hepatic encephalopathy, and portal hypertensive gastropathy between the rebleeding group and the non-bleeding group (P<0.050). The classification of liver function ［OR=3.444, 95% CI (1.211. 9.793）, P=0.020］, ascites ［OR=2.859, 95% CI (1.069, 7.645), P=0.036］, hepatic encephalopathy ［OR=4.265, 95% CI（1.121, 16.230）, P=0.033］, and portal hypertensive gastropathy ［OR=6.744, 95% CI (1.675, 27.156), P=0.007］, and platelet count ［OR=4.744, 95% CI (1.073, 20.969), P=0.040］ were the independent factors for the postoperative rebleeding by the logistic regression analysis.ConclusionFor patients with risk factors of rebleeding, preoperative treatment should be actively taken and postoperative prevention of rebleeding should be highly vigilant.
ObjectiveTo explore the value of artificial intelligence (AI) diagnostic imaging system and three dimensional computed tomographic bronchoangiography (3D-CTBA) surgical planning system in the management of multiple primary lung cancer (MPLC).MethodsThe clinical data of 53 patients with MPLC treated surgically in our hospital from January 2018 to August 2020 were retrospectively analyzed, including 16 males and 37 females, with a median age of 60 (39-75) years. The patients' preoperative CT was analyzed by AI and manually, and the data of patients who underwent 3D-CTBA were compiled to evaluate the value of AI and 3D-CTBA in the diagnosis and treatment of MPLC, respectively.Results The sensitivity of AI screening for MPLC was 84.91%. The sensitivity (91.90% vs. 83.78%) and accuracy (85.60% vs. 84.00%) of AI diagnosis of high-risk MPLC infiltrative lesions were better than those of manual diagnosis. 3D-CTBA was used for planning the surgery in 12 patients, and the intraoperative situation was generally consistent with the reconstructed results.ConclusionAI is of high value in identifying infiltrative lesions of MPLC. 3D-CTBA reconstruction of anatomical structures is accurate and can guide preoperative planning.