Abstract
BACKGROUND
There is a risk of rebleeding after acute non-variceal upper gastrointestinal bleeding (ANVUGIB), and rebleeding increases the risk of poor prognosis. At present, there is no effective predictive method for such rebleeding. The changes of red blood cell distribution width as well as hemoglobin and gastrin levels are all related to bleeding diseases, which can provide a reference for clinical improvement of relevant mechanisms and prediction of bleeding.
AIM
To investigate the relationship of the red blood cell distribution width (RDW), hemoglobin (Hb), and gastrin (GAS) with Glasgow-Blatchford score (GBS) and AIMS65 score in patients with ANVUGIB and to explore their predictive value for rebleeding.
METHODS
A prospective cohort study was conducted on 122 ANVUGIB patients admitted to our hospital from August 2018 to February 2021, and they were divided into either a rebleeding group (n = 31) or a no-rebleeding group (n = 91) according to whether rebleeding occurred 7 d after onset. Baseline data, RDW, Hb, GAS, GBS score, and AIMS65 score were compared between the two groups. Pearson correlation analysis was performed to explore the relationship of each index with GBS and AIMS65 scores, and multi-factor logistic regression analysis was performed to identify the influencing factors of rebleeding. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) were used to analyze the value of each index in predicting rebleeding, and to compare the incidence of rebleeding in patients with different levels of each index.
RESULTS
The proportion of patients with gastrointestinal bleeding in the rebleeding group was higher than that of the no-rebleeding group. The RDW, GAS, and GBS and AIMS65 scores were higher and Hb was lower in the rebleeding group than in the no-rebleeding group (P < 0.05). RDW and GAS were positively correlated with GBS and AIMS65 scores, and Hb was negatively correlated with GBS and AIMS65 scores (P < 0.05). Multivariate logistic regression analysis showed that after controlling for past gastrointestinal bleeding, GBS, and AIMS65 score, RDW, Hb, and GAS were still influencing factors of rebleeding (P < 0.05). The performance of RDW and Hb combined with GAS for predicting rebleeding (AUC = 0.850) was better than that of RDW (0.721), Hb (0.721), or GAS (0.806) alone. The rebleeding rate in patients with high levels of RDW and GAS was higher than that of patients with low levels, but patients with high levels of Hb had a lower rebleeding rate than those with low levels (P < 0.05).
CONCLUSION
RDW, Hb, and GAS are related to the risk of disease and rebleeding in patients with ANVUGIB. Combined detection of them may be a reliable method to assess the risk of disease and predict rebleeding.
Collapse