The Comprision Between The AIMS65 And Blatchford Scoring Systems In Non-Variceal Upper Gastrointestinal Bleeding

Objective:The aim of this study was to compare the predictive ability and applicability of AIMS65 and Blatchford risk scoring systems (BRS) about non-variceal upper gastrointestinal bleeding (NVUGIB) in clinical intervention, re-bleeding and death. Then to provide the basis in choosing a suit NVUGIB score system for clinician and elevate the accuracy of the prognosis.Methods:A retrospective observational study was performed in all patients with NVUGIB who attended the First Affiliated Hospital of Guangxi Medical University from June 2012 to October 2015. Then to Calculate the AIMS65 and BRS score for each patient and divide the patients into different groups by clinical intervention, re-bleeding and death. Then to compare their clinical value by plotting their ROC and calculating their AUC.Result:1. Scores of the AIMS65 and BRS after non-variceal upper gastrointestinal bleeding The AIMS65 score and BRS score of the clinical intervention group, re-bleeding group and death group were all higher than the no clinical intervention group, no re-bleeding group and survival group. All of the differences were statically significant (P<0.05). The patients who need clinical intervention, re-bleeding and in-hospital death were increasing as the AIMS65 score increased. And the same as the BRS score. AIMS65 and BRS were all suitable to estimate the clinical intervention and predict the re-bleeding and in-hospital death.2. The comparison between the AIMS65 and BRS in estimating the need of the clinical intervention The AUC of AIMS65and BRS in estimating the clinical intervention were0.730 (95%CI 0.684～0.772, P=0.023),0.814(95%CI 0.773-0.851, P=0.020), respectively; The differences between their AUC were statically significant (P=0.003). AIMS65 and BRS can predict the clinical intervention well and BRS was superior than AIMS65; 3. The comparison between the AIMS65 and BRS in predicting re-bleeding The AUC of AIMS65 and BRS in predicting re-bleeding were 0.716(95%CI 0.670～ 0.760, P=0.031)、0.768 (95%CI 0.675～0.750, P=0.029), respectively;But there was no statistically significant difference between their AUC (P=0.167). AIMS65 and BRS can predict re-bleeding well, but it still cannot say who was superior; 4. The comparison between the AIMS65 and BRS in predicting death The AUC of AIMS65 and BRS in predicting death were 0.893 (95%CI 0.859～ 0.794, P=0.023)、0.794 (95%CI 0.752-0.832, P=0.041), respectively; The differences between their AUC were statically significant (P=0.002). AIMS65 and BRS can predict in-hospital death well and AIMS65 was superior than BRS.Conclusion:1.AIMS65 and BRS were suitable for NVGIB to predicting the clinical intervention, re-bleeding and death.2. AIMS65 was not accurate as BRS in clinical intervention but superior than it in death.3. Although AIMS65 was not accurate as BRS in clinical intervention but superior than it in death and it was a simple method with accuracy that deserved clinical verifications.