Evaluation of Different Scoring Systems to Predict Outcomes of Non-Variceal GI Bleeding in the Elderly

Abstract

Introduction: Although several scoring systems have been developed to stratify risks and guide the management of patients presenting with GIB, their validity has not been established in the elderly population. Our aim was to assess the correlation of individual clinical/laboratory parameters as well as Blatchford, BLEED and AIMS 65 scores with the need for therapeutic endoscopic intervention, mortality and other clinical outcomes in elderly patients presenting with non variceal GIB. Methods: We conducted a retrospective analysis of patients over the age of 65 who presented to Florida Hospital, Orlando with lower or upper non variceal GI bleed between 2009 and 2012. We assessed Blatchford score on admission and the highest Blatchford score within the first 24 hours, AIM65 score and BLEED score on admission. Primary outcome was inhospital death or need for endoscopic/surgical intervention to control the bleed. Secondary outcome was a composite of inhospital death, need for any endoscopic/surgical intervention, blood transfusion during hospitalization and readmission within 30 days for GI rebleeding. Logistic regression and Pearson's chi-squared test were calculated to correlate individual clinical/laboratory parameters as well as Blatchford, BLEED and AIMS 65 scores with major clinical outcomes. Results: A total of 193 patients were included in the analysis, 4 of which, died in the hospital. Among them, 3 deaths were related to unstable comorbid conditions present on admission, while one patient died due to colonic perforation after colonoscopy. Primary outcome had the strongest correlation with lowest systolic blood pressure during the initial 24 hours of less than 100, pulse rate over 100, and with BLEED score. In contrast, there was only marginal correlation with Blatchford score (p = 0.075) and no correlation with AIMS 65 (p=0.88 ). AIM65 and BLEED scores did not correlate with the secondary outcome (p= 0.124 and 0.595 respectively), however, Blatchford score strongly correlated with the secondary outcome (p< 0.001). Conclusion: Mortality in elderly patients presenting with GIB is related to underlying comorbid conditions, rather than to GIB itself. Blatchford, BLEED and AIMS 65 scores, have limited utility in the elderly population. Among them, BLEED can be used to predict mortality and the need for urgent endoscopy while Blatchford score helps identify patients who might require transfusions and are at risk for recurrent GIB. AIM65 score had no significant use in elderly patients with GIB.Table 1: Association of scores and clinical parameters with primary outcomeTable 2: Association of scores and clinical parameters with secondary outcome