The AIMS65 Score is a Superior Predictor of Inhospital Outcomes in Patients with Nonvariceal Upper Gastrointestinal Bleeding of Benign Etiology

Aims and background: Upper gastrointestinal bleeding (UGIB) is defined as blood loss originating proximal to the ligament of Treitz.Incidence of UGIB in India ranges from 50 to 150 per 1,00,000 population annually.Patients from this region of India have not been well studied. Methods:We conducted a prospective study to evaluate the clinical profile, various etiologies, endoscopic findings, and outcomes in UGIB patients admitted to MKCG Medical College from April 2023 to December 2025.Results: A total of 294 patients with UGIB were included, with a median age of 48 years [interquartile range (IQR): 36-60].Regarding the etiology of UGIB, gastric and duodenal ulcers were the most common causes (59.9%; n = 160), followed by variceal bleeding and portal hypertensive gastropathy (26.6%, n = 71), Mallory-Weiss tears (4.1%; n = 11), esophageal ulcers (2.2%, n = 6), and gastric growths or polyps (7.2%; n = 19).Blood transfusion was required in 59.9% (n = 109) of patients, and 6.2% (n = 11) died during hospitalization.Admission Albumin, international normalized ratio (INR), mental status, systolic blood pressure, age >65 years (AIMS65) scores (p < 0.001), the Glasgow-Blatchford score (GBS) (p = 0.002), the clinical Rockall score (RS) (p = 0.014), and the complete RS (p = 0.025) were significantly higher among patients who died during hospitalization.The area under the receiver operating characteristic (AUROC) for the AIMS65 score (0.940) was superior to the GBS (0.761), the clinical RS (0.736), and the complete RS (0.725) in predicting inhospital survival.Similarly, AUROC for AIMS65 (0.734) and GBS (0.775) was superior to that of clinical RS (0.700) and complete RS (0.694) in predicting the need for blood transfusion.Conclusion: Gastric and duodenal ulcers were the most common causes of UGIB, followed by portal hypertension.Approximately twothirds of the patients required blood transfusions, whereas one-tenth died during hospitalization.The AIMS65 score was the most accurate predictor of inhospital mortality, whereas both the GBS and AIMS65 scores were better predictors of need for blood transfusion in nonvariceal UGIB of benign etiology.