Various vasoactive drugs are recommended to use in combination with endoscopic variceal ligation (EVL) in treating acute esophageal variceal bleeding (EVB). The efficacy and drug choice of vasoactive agents under Taiwan's National Health Insurance program remain to be validated.

The aim of this prospective cohort study was to assess the efficacy of somatostatin, compared with terlipressin in cirrhotic patients who have acute EVB and received EVL and the preference of vasoactive agents in real-world clinical practice.

From April 2010 through April 2015, cirrhotic patients with significant upper gastrointestinal bleeding were screened. Eligible patients with acute EVB were non-randomly assigned to receive early administration of somatostatin (group S) or terlipressin (group T) infusion, followed by EVL. A decision to use vasoactive drugs depended on the physician's favorite. In group S, somatostatin by intravenous bolus (250 μg) followed by 250 μg/hour was continued for 3 days. In group T, terlipressin was started with 2mg bolus injection and followed by 1 mg infusion every 6 hours for 3 days.

Introduction Rupture of esophageal varices that results in variceal hemorrhage is a major complication of cirrhotic patient leading to high mortality rate. Despite the control of initial bleeding, the early rebleeding may be as high as 30-50% among the survived patients and the mortality rate be up to 40%. Therefore it is recommended that the goals should be placed at bleeding arrest and prevention of early rebleeding.

Vasoactive drugs may reduce portal hypertension which leads to a reduction in variceal pressure to achieve better control of hemorrhage. In general, vasoactive agents profoundly improve the control of variceal bleeding if compared to placebo. On the other hand, the efficacy of somatostatin for early rebleeding control is similar to terlipressin. In suspected acute variceal bleeding (EVB), vasoactive drugs should be started as soon as possible and are generally prescribed for 5 day to prevent against very early rebleeding. Endoscopic variceal ligation (EVL) is recommended in patients with EVB and is best used in combination with vasoactive drugs according to the Baveno IV consensus. In contrast to consensus recommendation, short-course administration (2-day or 3-day) of vasoactive drugs are as effective as a 5-day course for the control of acute EVB when used as an adjuvant therapy to EVL.

The management of acute EVB in real world clinical practice may be much different from the recommendation of guidelines. In Taiwan, the National Health Insurance (NHI) program uses a single-payer system, covers 99.9% of the nation's population and provides physicians with a high degree freedom in choosing various medications. Nevertheless, the NHI system to date provides only a short-course (3 days) administration of vasoactive drugs in treating acute EBV. This leads us to conduct this study, with the aim of this prospective cohort study to assess the efficacy and safety of 3-day somatostatin compared with 3-day terlipressin in acute EVB patient receiving EVL in real world situation in Taiwan. The preference of doctors towards vasoactive drugs and the efficacy of vasoactive agents in preventing rebleeding were analyzed.

Materials and Methods Study Patients From April 2010 through April 2015, cirrhotic patients admitted to Kaohsiung Veterans General Hospital (tertiary referral center) with significant upper gastrointestinal bleeding were screened. Eligible patients with acute EVB were non-randomly assigned to receive early administration of somatostatin (group S) or terlipressin (group T), followed by EVL within 24 hours.

Inclusion criteria were: (i) acute hemorrhage from esophageal varice(s); (ii) portal hypertension attributed by cirrhosis; (iii) age was between 20 and 80 years old. Acute bleeding of esophageal varices was defined as when oozing or spurting were directly observed endoscopically, or when red color signs and/or white nipple sign were seen on esophageal varices without other potential sources of bleeding identified.

Patients were excluded if they met the exclusion criteria as follows : (i) being associated with hepatocellular carcinoma (HCC) of Barcelona-Clinic Liver Cancer (BCLC) > C; (ii) being associated with severe illness such as chronic obstructive pulmonary disease (COPD), septic shock, cerebral vascular event, acute coronary syndrome and uremia; (iii) with current gastric variceal bleeding; (iv) ever underwent endoscopic injection sclerotherapy (EIS), EVL within 6 month prior to current bleeding episode; (vii) ever received shunt or transjugular intrahepatic porto-systemic stent shunt (TIPS) procedure; (viii) allergic to and/or with contraindications of vasopressors; (ix) pregnancy.

Informed consent was given to each of all participated patients and obtained before the administration of vasoactive agents. This study conformed to the Declaration of Helsinki of the World Medical Association (2008) and was approved by the Institutional Review Board at Kaohsiung Veterans General hospital, Taiwan, Republic of China (IRB No. VGHKS99-CT4-20).

Study Design In this prospective cohort study, eligible patients were non-randomly assigned into two groups (group T: terlipressin and group S: somatostatin). A decision to use vasoactive drugs depended on the favorite of participating attending physicians who were requested to fill out the enclosed questionnaires immediately after the administration of the first dose for further understanding the underlying reason of the chosen medication. Convenient administration, safety profile, random selection and others were the four selections given. The result of selected choice was documented.

Patients suspected of acute variceal bleeding were administered with vasoconstrictors, either terlipressin (Glypressin, Ferring GmbH, Kiel, Germany) or somatostatin (Somatosan, BAG Health Care GmbH, Lich, Germany) and continued at the first time. After having been pre-medicated with hyoscine-N-butylbromide intramuscularly (20 mg), eligible patients underwent emergent endoscopy (index endoscopy) within 12 hours on admission. Two experienced endoscopists performed emergent EVL for the enrolled patients with acute EVB, who had performed EVL for more than 9 years. Pneumoactive ligation device (Sumitomo Bakelite Co., Ltd, Tokyo, Japan) and endoscopes (GIF XQ260; Olympus Co. Ltd, Tokyo, Japan) were applied. The targeted varix was suctioned and entrapped by a cap on endoscope. Ligation was aimed at active bleeding sites, red color spots or white nipple signs during the procedure. One to two rubber bands were applied to ligate the targeted varix. Patients were routinely placed on liquid diet in the following 3 days and subsequently shifted to semisolids and solids. Somatostatin was given by intravenous bolus (250 μg) followed by 250 μg/hour and continued for 3 days in group S. Terlipressin was started at 2mg bolus injection and followed by 1 mg infusion every 6 hours for 3 days in group T.

Both study groups were administered with 40mg pantoprazole intravenously after EVL for 3 days, and following oral form for 12 days to hasten the process of healing of ulcers induced by ligation. Oral administration of non-selective β-blocker was considered to start on day 5. Prophylactic antibiotic was initiated and continued for 5 days. Elective EVL was undertaken at the interval of three to four weeks. The period of follow-up was 42 days.

Study End Points Control of initial bleeding was defined as when treatment failure did not occur within 48 hours of recruitment. One criterion defined failure, whichever occurs first: new fresh blood vomitus > 2 hours after start of vasoactive drugs or therapeutic endoscopy, hemoglobin drop > 3 grams per deciliter in patients not transfused and death. An episode of clinically significant bleeding is being defined by blood transfusion > 2 units of packed red blood cells. Very early rebleeding was defined as when acute variceal bleeding occurred from 48 to 120 hours after having controlled of initial acute hemorrhage. Early rebleeding was defined as variceal hemorrhage occurs from day 3 till day 42 after initial bleeding arrest. The end points were the control of initial bleeding, very early rebleeding, early rebleeding, mortality at 6 weeks, blood transfusion, hospital stays and adverse events in real world clinical practice.

Statistical Analysis By using the Student's two-tailed t-test, quantitative data were expressed as mean+ standard deviation and compared. X2 and Fisher's exact test were used for the comparison in qualitative variables wherever appropriate. The reported p values were implemented against two sided tests, wherever appropriate. P-value < 0.05 was considered significant. All analyses were conducted using SPSS 12.0.1C (SPSS Inc., Chicago, IL, USA).

Inclusion Criteria:

• acute hemorrhage from esophageal varice(s)
• portal hypertension attributed by cirrhosis
• age was between 20 and 80 years old

Exclusion Criteria:

• being associated with hepatocellular carcinoma (HCC) of Barcelona-Clinic Liver Cancer (BCLC) > C
• being associated with severe illness such as chronic obstructive pulmonary disease (COPD), septic shock, cerebral vascular event, acute coronary syndrome and uremia
• with current gastric variceal bleeding
• ever underwent endoscopic injection sclerotherapy (EIS), EVL within 6 month prior to current bleeding episode
• ever received shunt or transjugular intrahepatic porto-systemic stent shunt (TIPS) procedure
• allergic to and/or with contraindications of vasopressors
• pregnancy.