Predictors of Rebleeding and Mortality in Patients with High-Risk Bleeding Peptic Ulcers
- 193 Downloads
Background and Aim
Patients with bleeding ulcers can have recurrent bleeding and mortality after endoscopic therapy. Risk stratification is important in the management of the initial patient triage. The aim of this study is to identify the clinical and laboratory risk factors for recurrent bleeding and mortality.
A prospective study was conducted in 390 consecutive patients with bleeding peptic ulcers and high-risk endoscopic stigmata, e.g., active bleeding, a non-bleeding visible vessel, adherent blood clot, and hemorrhagic dot. We tested 13 available variables for association with recurrent bleeding and 15 were tested for association with mortality. A logistic regression model was used to identify individual correlates associated with these adverse outcomes.
Bleeding recurred in 46 patients (11.8%) within 3 days and 21 patients (5.4%) had in-hospital mortality. In the full-factor analysis model, the incidence of recurrent bleeding was significantly higher in five of the 13 investigated variables and mortality was significantly higher in two of the 15 variables. In the final analysis model, significant risk factors for recurrent bleeding within 3 days, with adjusted odds ratios (OR), were in-hospital bleeding (OR 3.3), initial hemoglobin level <10 g/dl (OR 3.3) and ulcer ≥2 cm (OR 2.0). In-hospital bleeding was the only independent risk factor for mortality (OR 8.3).
The study emphasizes the role of ulcer size, anemia and in-hospital bleeding as the determining high-risk predictors for adverse outcomes for bleeding peptic ulcers.
KeywordsRisk factors Logic regression model Ulcer bleeding In-hospital bleeder
- 1.Barkun A, Sabbah S, Enns R, et al. The Canadian Registry of Nonvariceal Upper Gastrointestinal Bleeding and Endoscopy (RUGBE): endoscopic hemostasis and proton pump inhibition are associated with improved outcomes in a real-life setting. Am J Gastroenterol. 2004;99:1238–1246.CrossRefPubMedGoogle Scholar