Validation of AIMS65 score in patients with acute upper gastrointestinal bleeding

Abstract

INTRODUCTION: Current guidelines on the management of acute upper gastrointestinal bleeding (AUGIB) suggested an early risk stratification of patients to predict prognosis of the disease. However, the current available scores are complex and have not been widely used in clinical practice. AIMS65 score is a simple score that can be used to risk stratify patients with AUGIB. OBJECTIVE: To validate AIMS65 score as a tool to predict mortality in AUGIB in Hospital Tengku Ampuan Afzan Kuantan Pahang. We also associate AIMS65 score with other outcomes of AUGIB. DESIGN and METHOD: This was a prospective cohort study in a single tertiary medical centre in Malaysia among adult patients presented with AUGIB from 1st January to 31st December 2016. The AIMS65 score was calculated upon first presentation or symptoms. Patients were monitored throughout their hospitalisation and were subsequently followed-up at 30 days. The outcomes that were measured were; in-patient mortality, 30-days mortality after discharged, re-bleeding, the need for blood transfusion and re-endoscopy. Area under the Receiving Curve (AUROC) for AIMS65 was calculated for each of the outcome. RESULTS: There were forty seven males and nine females enrolled into our study. The mean age of the subjects was 58.8 ±14.7 years old. There were eight in-patient deaths and nine deaths on 30-days follow up. The mean length of stay was 5.3±4.1 days. The AIMS65 score was statistically significant to predict in-patient mortality, AUROC of 0.941 (p<0.001); 30 days mortality AUROC of 0.922 (p<0.001); and the need for blood transfusion AUROC of 0.752 (p=0.005). There was also statistically significant positive association with length of hospital stay (P=0.033). However AIMS65 score was not statistically significant in predicting the need for ICU admission, the need of re-endoscopy and re-bleeding. CONCLUSION: The AIMS65 score is valid and can be use to risk stratify patients with AUGIB, particularly to predict in-patient mortality, 30 days mortality and the need of blood transfusion.