Evaluation Of Endoscopic Treatments In Patients With Esophagogastric Variceal Bleeding

Objective:1. To investigate the prophylactic effects of different endoscopic treatments (ligation+cyanoacrylate vs ligation plus sclerotherapy+cyanoacrylate) for esophagogastric varices in patients with liver cirrhosis. A randomized controlled trial was established to further confirm the results.2. To evaluate the efficacy of endoscopic therapy for esophagogastric varices in patients with hepatic carcinoma, long term survival in such patients and to identify prognostic factors for rebleeding and survival.Methods:1. A retrospective cohort was conducted with76patients with liver cirrhosis who were admitted to Zhongshan Hospital during1st April,2004and27th March,2012because of variceal bleeding. Each patient was diagnosed through gastroscopy both esophageal and gastric varices, who received either ligation+cyanoacrylate injection or combined ligation and sclerotherapy+cyanoacrylate injection.2. Cirrhotic patients who received treatment of ligation for esophageal yarices and cyanoacrylate injection for gastric varices from28th,May,2012to26th,Oct,2012were included in the randomized controlled trail.3. A retrospective study of117patients who underwent endoscopic therapy for esophagogastric varices complicated by hepatic carcinoma during1st April,2004and31th October,2012.4. Clinical information and laboratory or imaging results were investigated and recorded as baseline characteristics. All patients were followed up until endpoints. Primary and secondary outcomes were recurrence rate of bleeding, mortality rate, incidence rate of complications, etc. All analysis were performed using the SPSS16.0. Kaplan-Meier survival analysis was conducted and log-rank test was used to compare the curves. Prognostic factors were compared by Cox’s proportional hazard model.Results:1. The median follow-up period of76patients was13.1months. There was no significant difference of rebleeding rate (P=0.867), mortality rate (P=0.756) and eradication rate(P=0.803) between the two groups. The incidence of retrosternal pain (OR=4.00,95%CI1.10-15.84, P=0.038) was significantly higher in combination treatment group. No significant differences were found in the incidence rate of portal hypertensive gastropathy (OR=1.67,95%CI0.57-4.84, P=0.345) between the two groups.2. In the randomized controlled trial,50patients were followed up for a median duration of6.0months.13patients(26.0%) have recurrence of upper gastrointestinal bleeding. The rebleeding rate between the two groups were not significantly different16.0%vs36.0%, P=0.102). In stratified analysis, compared with combination treatment group, the rebleeding rate was significantly lower in ligation group(0.0%vs50.0%, P=0.039) among the patients who received propranolol concomitantly. Combination treatment group was also associated with significantly higher rebleeding risk(11.1%vs45.0%, P=0.023) among the patients without diabetes. However, different endoscopic treatments were not related with significantly different mortality rate(0.0%vs8.0%,P=0.153) or eradication rate between the two groups. The incidence rate of complications or portal hypertensive gastropathy was not significantly different.3. Median follow-up period of117patients was9.6months.60patients(48.7%) were presented with relapse of bleeding. The cumulative1-year,2-year and5-year rebleeding rate was46.6%,65.6%and59.8%.65patients(55.6%) were dead during the follow-up, with overall1-year,2-year and5-year cumulative mortality of59.8%,46.2%and24.1%. Different endoscopic procedures did not influence the risk of rebleeding or death. Hyponatremia(P=0.028, HR=6.660,95%CI1.225-36.197) and diameter of portal vein>12mm(P=0.001, HR=12.340,95%CI2.692-56.558) were independent predictor for relapse of bleeding, while AFP>20ng/ml(P=0.004, HR=3.851,95%CI1.522-9.743) and thrombin time>21s(P=0.001, HR=38.207,95%CI6.196-235.620) predicted shorter survival time.Conclusion:1. The results of the retrospective study suggested that there was no significant difference in terms of rebleeding, mortality, or eradication rate between ligation+cyanoacrylate group and combined ligation and sclerotherapy+cyanoacrylate group among cirrhotic patients with both esophageal and gastric varices. But the latter group was associated with higher risks of complications.2. The primary results of the randomized controlled trial demonstrated that among cirrhotic patients with both esophageal and gastric patients, there was no significant difference observed in relation to risk of rebleeding, mortality, incidence rate of complications or portal hypertensive gastropathy between ligation+cyanoacrylate group and combined ligation and sclerotherapy+cyanoacrylate group. However, the results of the stratified analysis suggested a superiority of ligation+cyanoacrylate injection for the prevention of rebleeding among such patients with concomitant use of propranodol or without diabetes.2. Patients who had esophagogastric variceal bleeding complicated by hepatic carcinoma had poor overall prognosis. Different endoscopic procedures were not related with increased risks of rebleeding or mortality. Hyponatremia and diameter of portal vein>12mm were independent predictor for relapse of bleeding, while AFP>20ng/ml and thrombin time>21s predicted shorter survival time in such patients.