The Selection And Prognosis Of Transfusion Strategy In Acute Upper Gastrointestinal Bleeding

Background and Aims:Acute upper gastrointestinal bleeding (AUGIB) is a commonemergency condition. It is a frequent indication for blood transfusion,especially red-cell transfusion, because acute blood loss can cause anemiaand even peripheral circulatory failure. Transfusion may be lifesaving inpatients with massive exsanguinating bleeding. However, in most caseshemorrhage is not so severe, and in such circumstances the safest andmost effective transfusion strategy is controversial. We aim to comparethe selection and prognosis of different transfusion strategy in patientswith acute non-massive upper gastrointestinal bleeding.Methods:The retrospective data analysis comprised adult (≥18years) patientsadmitted to Digestive Department of the First Hospital of Jilin Universitybetween January1,2012and December31,2012, with the primarydiagnosis of acute non-massive upper gastrointestinal bleeding. Thisstudy included166patients and all the patients underwent endoscopy andreceived red-cell transfusion while hospitalized. The patients were assigned to two groups in accordance with the hemoglobin levels beforetransfusion: a restrictive strategy (transfusion when the hemoglobin levelfell below70g/L) and a liberal strategy (transfusion when the hemoglobinfell between70-90g/L).Results:1. Hemorrhage etiology: in all of166patients, the common etiologyof hemorrhage were: peptic ulcer (87cases,51%), esophagogastricvariceal bleeding (49cases,30%), erosive gastritis or esophagatis (16cases,10%), Mallory Weiss Syndrome (8cases,5%), gastric cancer (6cases,4%). The proportion between two groups was not significantlydifferent (p>0.05).2. Mortality: the mortality at4weeks was significantly lower in therestrictive strategy group than in the liberal strategy group (5.5%vs7.1%,p＜0.05).3. Risk of further bleeding: the rate of further bleeding wassignificantly lower in the restrictive strategy group than in the liberalstrategy group(9.1%vs19.6%，p＜0.05).4. In the subgroup of patients with cirrhosis, the risk of death and therisk of further bleeding were significantly lower in the restrictivetransfusion strategy than in the liberal transfusion strategy among patientswith Child–Pugh class A or B disease, whereas in the subgroup ofpatients with Child–Pugh class C disease, the risk of death and the risk of further bleeding were not differ significantly between the groups.5. Incidence of adverse events: the overall rate of adverse events wassignificantly lower in the restrictive strategy group than in the liberalstrategy group (31.8%vs46.4%, p＜0.05).Conclusions:In summary, a restrictive transfusion strategy, as compared with aliberal transfusion strategy, improved the outcomes among patients withacute non-massive upper gastrointestinal bleeding. The rate of survivalwas increased, and the risk of further bleeding, the need for rescuetherapy, and the rate of adverse events were all significantly reduced,with the restrictive transfusion strategy. As compared with a liberaltransfusion strategy, a restrictive strategy significantly improvedoutcomes in patients with acute non-massive upper gastrointestinalbleeding.