The Application Value Of Several Scoring Systems In Patients With Acute Non-variceal Upper Gastrointestinal Bleeding

Objective:To compare the predictive of AIMS65,Pre-endoscopy Rockall Risk Score(PRS),Pre-endoscopy Baylor Bleeding Score(BBS)and Glasgow Blatchford Score(GBS)and shock index(SI)in the emergency hall for clinical intervention,rebleeding and death in patients with acute non-variceal upper gastrointestinal Bleeding(ANVUGIB).Methods:A retrospective analysis was conducted on 105 ANVUGIB patients who visited the emergency hall of our hospital from January to December 2018,and the laboratory examination results and medical history data completed after the patients were collected.According to the above results,AIMS65,PRS,GBS,pre-endoscopy BBS and SI values of all ANVUGIB patients were calculated respectively.Divided into two groups: AIMS65 low-risk group and high-risk group by 2 points;divided into PRS low-risk group and medium high-risk group by 3 points;divided into pre-endoscopy BBS low-risk group and high-risk group by 6 points;divided into GBS low-risk group and high-risk group by 12 points.Clinical intervention(blood transfusion,surgical or endoscopic intervention),rebleeding,and death in patients with different risk stratification under the same scoring system were compared,respectively.According to the patients with and without clinical intervention,and bleeding and discharge end,divided into clinical intervention group and non clinical intervention group,group of bleeding and have bleeding again and again death and survival groups,draw four rating scale,SI and SI + AIMS65 ROC curve,compare four kinds of scale,SI and SI +AIMS65 for ANVUGIB patients clinical intervention,rebleeding and the predictive value of death.Results: The clinical intervention rate,rebleeding rate and mortality rate of patients in the low-risk group of AIMS65 were lower than those in the high-risk group(P < 0.05);the clinical intervention rate and rebleeding rate in the high-risk group were higher than those in the low-risk group(P < 0.05),but there was no statistical difference in mortality rate(P > 0.05).With AIMS6565,PRS,GBS,pre-endoscopic BBS,SI and SI +AIMS65,the results showed that the AUC and 95% confidence interval(CI)of the four scoring methods were AIMS65 = 0.723(95% CI,0.627-0.806),PRS = 0.641(95% CI,0.542-0.732),GBS = 0.826(95% CI,0.740-0.893),pre-endoscopy BBS = 0.649(95%CI,0.549-0.739),SI=0.615(95%CI,0.515-0.708),SI+AIMS65=0.742(95%CI,0.648-0.823).GBS was better than other scoring methods in predicting the needs of clinical intervention,but there was no difference compared with SI + AIMS65.For the prediction of rebleeding,the AUC values of four scoring systems and SI,SI+AIMS65were as follows: AIMS65 = 0.771(95% CI,0.679-0.848),PRS = 0.779(95% CI,0.732-0.8886),GBS = 0.786(95% CI,0.695-0.860),pre-endoscopy BBS = 0.799(95%CI,0.710-0.871),SI=0.571(95%CI,0.471-0.667),SI+AIMS65=0.783(95%CI,0.692-0.858),and there was no statistical difference among the four scales and SI+AIMS65.(P > 0.05);for the prediction of death,AUC value was AIMS65 = 0.861(95% CI,0.780-0.921),PRS = 0.724(95% CI,0.628-0.806),GBS = 0.719(95% CI,0.623-0.802),pre-endoscopy BBS = 0.723(95% CI,0.627-0.806),SI=0.553(95%CI,0.452-0.650),SI+AIMS65=0.893(95%CI,0.818-0.945),AIMS65 was superior to GBS scoring system,there was statistical difference(0.861 vs 0.719,P< 0.05).However,there was no significant difference between AIMS65 and pre-endoscopy BBS and PRS and SI+AIMS65 in predicting death(P > 0.05).Conclusion: AIMS65,PRS,pre-endoscopy BBS,GBS score and SI + AIMS65 can predict the clinical intervention,rebleeding and death of ANVUGIB patients.SI +AIMS65 is better than GBS in predicting death and GBS is better than AIMS65,pre-endoscopy BBS and PRS in predicting clinical intervention.In predicting rebleeding,the abilities of AIMS65,PRS,pre-endoscopy BBS,GBS and SI + AIMS65 were similar.