Introduction The mortality of the cirrhotic patients with shock induced by esophageal varices (EV) rupture haemorrhage remains high despite all medical advances. The clinical study aims to exhibit the efficacy of endoscopic therapy using esophageal varicose ligation (EVL) in treating hypotension caused by upper gastrointestinal haemorrhage, as well as compare the long-term difference in effectiveness of single-band and multi-bands techniques for the patients.

Methods sixty-seven patients with clinical diagnosis of cirrhosis were hospitalised in an emergency for massive hematemesis. The blood pressure (Bp) of the patients was between 60–82/30-55 mmHg, and the haemoglobin (Hb) was in the range of 50–87 g/L. As well as preparing blood protects, we take fluid resuscitation and apply somatostatin or terlipressin to reduce portal vein pressure of the patients, whose Hb was lower than 60 g/L were taken blood transfusion. When their Bp could stabilise between 78–86/50-55 mmHg and the Hb between 55–60 g/L, the emergency endoscopy examination and treatment were performed without narcosis. The patients with EV bleeding were randomly divided into single-band and multi-band ring ligation groups, while accompanying GV treated by injecting in the bleeding gastric varicose vein with sequential sclerosing agent, tissue adhesive, sclerosing agent, which is named sandwich method of GV injection therapy. The sixty-seven patients were found EV by gastroscoy, in whom there were fifty-three accompanying GV. In multi-bands treatment group with thirty-five patients, two, three, or four bands were applied on the varices according to the severity of EV. The single-band treatment group was consisting in thirty-two patients.

Results Sixty-six cases were rescued successfully by emergency endoscopic treatment within 24 h of bleeding. Unfortunately, one person was still bleeding although EVL and sclerotherapy both being applied the bleeding oesophagus because it had been more than 24 hours since the bleeding and he had been treated with Sengtaken-Blakemore tube for three days, causing the erosion and ulcer of oesophagus. Two days after the endoscopic therapy, the patient died for the poor control of haemorrhage as he continually accepted comprehensive treatment. In the multi-bands ligation group, the number of patients with EV completely eradicated and mostly eradicated was eight and twenty-one respectively. The disappearance percents of varicose veins were 82.8%. In the single-band ligation group, the number of people with EV completely eradicated and mostly eradicated were five and fourteen, respectively, which the disappearance percent of the varicose veins was 59.4%. There were great significant differences between the two groups (p < 0.05). During one year of follow-up the EV recurrence percent of the multi-bands ligation group was 5.7%, which was lower than that of the single-band ligation group 18.7% (p < 0.05). No esophageal stenosis was observed in both two groups.

Conclusion The patients with EV and (or) GV rupture haemorrhage in bleeding and hypotension state can be rescued successfully by endoscopy and endoscopic therapy. The long-term efficacy of the multi-band ligation is superior to single-band EV ligation (EVL).

Disclosure of Interest None Declared