Abstract: Objective By retrospectively analyzing the transfusion threshold used in the clinical transfusion practice for acute upper gastrointestinal bleeding (AUGIB) patients admitted to our hospital, we evaluated the prevalence and utility of restrictive transfusion strategies for such patients. Methods Patients diagnosed with AUGIB and transfused with red blood cells in our hospital from July 2020 to June 2022 were selected as research subjects. The frequency of the hemoglobin (Hb) threshold used in the transfusion patients was investigated. According to the Hb threshold of blood transfusion, the patients were divided into a restrictive transfusion group (the hemoglobin threshold for transfusion was 70 g/L, and 80 g/L in the patients with coronary heart disease) and a non-restrictive transfusion group. Then, the differences in clinical outcomes between the two groups were compared. Results Among the 384 transfusion patients, Hb 70g/L transfusion threshold was used in 62.0%, 70~80 g/L threshold in 32.8%, 80~90 g/L threshold in 3.6%, and the liberal transfusion strategy of Hb≥90g/L in only 1.6%. The main transfusion strategy for 384 patients was restrictive transfusion, accounting for 73.4%, and the ratio was up to 88.5% in the subgroup of patients with coronary heart disease; the non-restrictive transfusion strategy was taken only in 26.6% of 384 patients. The rate of re-bleeding during hospitalization was 4.6% in the restrictive transfusion group, and 10.8% in the non-restrictive transfusion group, with the difference between the two groups being statistically significant (P=0.027). All-cause mortality during hospitalization was 5.3% in the restrictive transfusion group, and 5.9% in the non-restrictive transfusion group; there was no significant difference in mortality between the two groups (P=0.830). Totally 54 of 384 (14.3%) patients failed to reach the target Hb value after transfusion and were at risk of undertransfusion, all of whom were in the restrictive transfusion group. 12.5% of patients had Hb concentrations below the target Hb value before discharge, and 45 of the 48 undertransfusion patients were in the restrictive transfusion group. Conclusion The restrictive transfusion strategy widely used in the clinical transfusion practice for AUGIB patients is safe and practicable. While avoiding over-transfusion, we should also be alert on the risk of undertransfusion stemming from the restrictive transfusion strategy.

Key words: Acute upper gastrointestinal bleeding, Hemoglobin, Restrictive transfusion strategy, Transfusion thresholds