Objective To explore the feasibility and safety of liver transplantation （LT） in treatment of upper gastrointestinal hemorrhage in patients with portal hypertension， and to compare the therapeutic effects with conventional operation （CO）. Methods The clinical data of 303 patients with bleeding portal hypertension from Feb. 2009 to Feb. 2012 in the department of hepatobiliary and pancreatic surgery of First Affiliated Hospital of Zhejiang University were retrospectively analyzed. One hundred and one patients received LT procedure （LT group）， whereas the other 202 patients received CO procedure （CO group）. Postoperative follow-up period was 8-44 months （average 26 months）. Results Liver function before operation in CO group was significantly better than that in LT group（P＜0.01）. The mortality of CO group and LT group were 7.4%（14/189） and 3.0%（3/101， P=1.00）， respectively. The rebleeding rate of patients underwent LT was 2.0%（2/101）， significantly lower than that of CO group 〔9.5%（18/189）， P＜0.05〕. The vanish rate of esophagogastric varice in patients underwent LT was 86.1%（87/101）， significantly lower than that of CO group 〔54.5%（86/189）， P＜0.01〕. Conclusions LT treatment for bleeding portal hypertension is feasible and safe. Patients with good liver function despite hemorrhage history may be managed satisfactorily with conventional surgery. LT is the only curative treatment for patients with portal hypertension in end-stage liver disease.
Objective To apply the method of evidence-based medicine to identify the best therapy option for an emergency patient with upper gastrointestinal hemorrhage. Methods According to time and logical sequence of clinical events, a complete decision tree was built after the following steps to find the best treatment: clear decision-making, drawing decision tree graphics, listing the outcome probability, giving appropriate values to the final outcome, calculating and determining the best strategies. Results The performance of endoscopic therapy for the patient with upper gastrointestinal hemorrhage within the first six hours had little effect on the prognosis. Interventional therapy after the failure of endoscopic therapy had less mortality than direct surgical exploration. Conclusion Making clinical decision analyses via drawing the decision tree can help doctors clarify their ideas, get comprehensive views of clinical problems, and ultimately choose the best treatment strategy for patients.