| [Objective]1.We retrospectively analyzed general characteristics form patients hospitalized with non-variceal upper gastrointestinal bleeding.and evaluating the ability of prediction for rebleeding,mortality,clinical intervention and the optimal thresholds of clinical outcomes with complete Rockall score,clinic Rocall score,Glasgow-Blatchford score,Modified Glasgow-Blatchford score and AIMS65 score.2.To explore the risk factors for rebleeding,mortality,clinical intervention for patients with NVUGIB,to improve the curative effect and patient's prognosis and making a foundation for the new scoring systems in the future.[Method]In this study,a retrospective analysis was performed on 647 hospitalized patients who were treated and endoscopic diagnosis with NVUGIB in the Digestive internal medicine ward of the second affiliated Hospital of Kunming Medical University from January 2014 to May 2019 according to the criteria of admission.Age 18 years old or more.record detailed of patient's general information,including name,gender,age and other general information and laboratory index,endoscopic diagnosis,disease outcomes,hospitalization days,cause of hemorrhage,the hospital clinical intervention therapy and so on.Then calculated using five scoring systems for each patient.Draw the receiver-operating characteristic curve(ROC),according to the area under the receiver-operating characteristic curve(AUROC)to evaluate the ability of prediction for rebleeding,mortality,clinical intervention for different scoring system and calculate the Optimal cut-off values for different scoring systems and outcomes.2.Using single factor and multiple factors logistic regression analysis to identify the risk factors for rebleeding,mortality,clinical intervention in NVUGIB.[Results]1.General data:Collect 647 patients with NVUGIB,there were 507 males(78.4%)and 140 females(21.6%),the males were 3.6 times as greater as the females.the average age was 58.89±16.30 years old,the average hospitalization time was 10.93±3.88 days.There were 459 cases(70.9%)with emergency endoscopy.According to the results of endoscopic examination,Peptic ulcer 528 cases(81.6%)is the common cause of NVUGIB.Overall,29 cases(4.5%)experienced rebleeding and 12 cases(1.9%)died in hospital.And a total of 322 cases(49.8%)needed at least one clinical intervention,of whom 273 cases(42.2%)needed blood transfusion,110 cases(1 7%)needed endoscopic hemostasis,5 cases(0.8%)needed interventional therapy.2.The scoring system to predict value2.1 rebleeding:The mean scores in patients with rebleeding was higher than the others without rebleeding outcomes(P<0.05).GBS(AUROC 0.73,95%CI,0.694-0.764),mGBS(0.781,95%CI,0.682-0.753)scores has good predictive value in rebleeding and were superior to the PRS(0.646,95%CI,0.608-0.683),CRS(0.628,95%CI,0.589-0.665),AIMS65(0.616,95%CI,0.577-0.654)scores(p<0.05),but the predictive value of GBS and mGBS score no difference(p 0.371).As to rebleeding,the CRS,PRS,GBS,MGBS,AIMS65 scoreS of more than 4,3,10,8 and 0 were the optimal cut-off values.2.2 mortality:The mean scores in patients with death was higher than the others without adverse outcomes,except for the PRS score(P=0.112).CRS(0.738,95%CI,0.702-0.771),GBS(0.861,95%CI,0.832-0.887),mGBS(0.827,95%CI,0.796-0.855),AIMS65(0.867,95%CI,0.839-0.893)scores has good predictive value in mortality,AIMS65 score was superior to the PRS(0.631,95%CI,0.592-0.668)and CRS scores(p<0.05),but the predictive value of GBS,mGBS and AIMS65score no difference.CRS score was superior to the PRS.As to mortality,the CRS,PRS,GBS,MGBS,AIMS65 scoreS of more than 3,3,11,10 and 1 were the optimal cut-off values.2.3 clinical intervention:The mean scores in patients with clinical intervention were higher than the others without adverse outcomes(p<0.05).GBS(0.747,95%CI,0.711-0.780),mGBS(0.740,95%CI,0,704-0.773)scores has good predictive value in clinical intervention and were superior to the AIMS65(0.674,95%CI,0.637-0.710),CRS(0.670,95%CI,0.632-0.706)and PRS(0.595,95%CI,0.557-0.634).But the predictive value of GBS and mGBS score no difference(p 0.184).AIMS65 and CRS scores were superior to the PRS score,but the predictive value of AIMS65 and CRS score no difference.As to clinical intervention,the CRS,PRS,GBS,MGBS,AIMS65 scoreS of more than 4,3,8,6 and 0 were the optimal cut-off values.3.The different clinical outcome related risk factors3.1 rebleeding:History of bleeding and of drug use easily bleeding,the age more than 65,heart rate,hemoglobin,albumin,INR,Platelets,white blood cells,blood urea and ALT,AST were associated with rebleeding.Hemoglobin is happening to the protection of bleeding risk factors.3.2 mortality:the age more than 65,hemoglobin,albumin,INR,white blood cells and blood urea were associated with mortality.The independent predictors for mortality were INR and white blood cells.Albumin is happening to the protection of bleeding risk factors.3.3 clinical intervention:History of drug use easily bleeding,the age more than 65,heart rate,hemoglobin,albumin,INR.white blood cells and blood urea were associated with clinical intervention.Hemoglobin and Platelets were happening to the protection of clinical intervention risk factors.[Conclusion]1.Peptic ulcer was still the main causes of NVUGIB.2.GBS and mGBS scores has good predictive value in rebleeding and clinical intervention and were superior to the PRS,CRS,AIMS65 scores.AIMS65 scores has good predictive value in mortality,GBS and mGBS scores after AIMS65 score were superior to the PRS,CRS.The predictive value of GBS and mGBS scoring system was not different in terms of NVUGIB rebleeding,death and clinical intervention.mGBS may can take the place of GBS.3.CRS scores were superior to the PRS score for NVUGIB death and clinical intervention,the predictive value for rebleeding no difference,but the AUROC was poor,it is not the ideal prediction tool.4.Hemoglobin is happening to the protection of bleeding risk factors.The independent predictors for mortality were INR and white blood cells.Albumin is happening to the protection of bleeding risk factors.Hemoglobin and Platelets were happening to the protection of clinical intervention risk factors. |
PDF Full Text Request