| Objective 1. We retrospectively analyzed epidemiological characteristics, etiology, the emergency endoscopy and clinical treatment from patients hospitalized with nonvariceal upper gastrointestinal bleeding from 8 hospitals of China, preliminary understanding the problems in the diagnosis and treatment of NVUGIB in current China, and evaluating the before endoscopy and complete Rockall score and Blatchford score in further bleeding, death, surgery, endoscopic intervention and transfusion.2. On the basis of retrospective study, we prospective compared the efficacy and safety of a restrictive transfusion strategy with those of a liberal transfusion strategy in NVUGIB.Methods 1.8 hospitals attended the study, we retrospectively analyzed the patients who were diagnosed NVUGIB by endoscopy between 2008-2012, age 18 years old or more, to develop questionnaire, and design EpiData software visual data entry interface, and record detailed clinical basic data and test results; According to Rockall score standard and Blatchford score standard, we calculate the score of patients with complete data respectively.2. We enrolled 132 patients with NVUGIB, recorded detailed the baseline data, endoscopic diagnosis, the level of hemoglobin before transfusion, transfusion amount and whether death, rebleeding, complications and length of hospital stay and follow-up data, and analyzed the relationship between transfusion and clinical prognosis.Results 1.The study included data from 2977 patients from 8 hospitals of China. Less than or equal to 60 age group (n=1771),60-79 age group (n=999), more than or equal to 80 age group (n=207), the mean age was 54.65±14.34 years, and the male to female ratio was 3.25:1.95.47%(2842/2977) in common ward,3.53%(105/2977) in the emergency ward,1.00%(31/2977) in ICU. Pepetic ulcer is the common causes of NVUGIB, there is a general declining trend in the constituent ratio of peptic ulcer disease in recent years (17.2%,16.1%,14.1%,13.0%,12.8%).14.41%(429/2977) patients with comorbidity,58.4%(1740/2977) patients with the Rockall score,60.46% (1800/2977) patients with Blatchford score,92.85%(2764/2977) patients were treated by proton pump inhibitors before endoscopy, tranexamic acid was used to stop bleeding in 18.95%(564/2977) patients, erythromycin and metoclopramide was used to improve the endoscopic diagnosis rate in 1.24%(37/2977) and 1.24%(9/2977) patientsrespectively.1.28%(38/2977) patients with nasogastric tube,19.65%(585/2977) patients with emergency endoscopy,23.45%(698/2977) patients with transfusion, 5.34%(159/2977) patients with endoscopic treatment,16.9%(96/568) high-risk peptic ulcer with endoscopic intervention,2.65%(79/2977) patients with surgery, 0.50%(15/2977) patients with angiographic embolization. Bleeding was due to peptic ulcer in 2181 patients (73.26%),159 patients found adhensive thrombus,10 cases were rinsed by saline and exposed basement lesions.61.71%(1346/2181) patients with helicobacter pylori detection.2.92%(87/2977) patients with rebleeding,1.81% (54/2977) patients with complications caused by bleeding.237 cases took aspirin before hospitalization, of 32.50%(77/237) restarted, the average time of 25.92±11.31 days (5 to 60 days), which have much difference with the international consensus(7-10 days). The median length of hospital stay was 8 days (IQR,5-11), the mortality was 1.71%(51/2977); Pre- endoscopic Rockall score (AUC=0.842) and complete the Rockall score (AUC=0.804) to predict mortality were superior to the Blatchford score (AUC=0.622). Pre-endoscopic Rockall score(AUC=0.658) predicting rebleeding was better than complete Rockall score (AUC=0.548) and the Blatchford score (AUC=0.528). Three scoring system to predict the operation ability were poor (0.589 vs.0.547 vs.0.504). Blatchford score (AUC= 0.698) predicting transfusion was superior to pre-endoscopic Rockall score (AUC=0.608) and complete Rockall score (AUC=0.610).2. As compared with a liberal transfusion strategy, a restrictive strategy significantly improved rebleeding in patients with NVUGIB (HR=0.34,95%CI=0.1180.988, p=0.05). However, there were no differences in mortality rates、hospital stay and transfusion amount.Conclusion 1. Pepetic ulcer is the common causes of NVUGIB, there is a general declining trend in the constituent ratio of peptic ulcer disease in recent years and high-risk peptic ulcer with lower endoscopic intervention (16.9%). Time to restart the antithrombotic drugs have much difference with the international consensus(7-10 days). During taking NSAIDS or antiplatelet aggregation drugs, the proportion of the combined use of PPI is 0.91%. Emergency endoscopy ratio is only 37.25%, ratio of application of surgery and angiographic embolization is 2.65% and 0.50%, respectively.2. Pre-endoscopic Rockall score and complete Rockall score has a better predictive value to the mortality, pre-endoscopic Rockall score predicts bleeding is better than the complete Rockall score and Blatchford score. Three scoring system to predict the operation ability were poor. Blatchford score predicting transfusion was superior to pre-endoscopic Rockall score and complete Rockall score.3. As compared with a liberal transfusion strategy, a restrictive strategy significantly improved rebleeding in patients with NVUGIB. |
PDF Full Text Request