Is the Modified AIMS65 Score Suitable for Distinguishing the Low Risk Group and the High Risk Group for Patients with Upper Gastrointestinal Bleeding?

ABSTRACT

PURPOSE: The goal of this study was to increase the performance of the AIMS65 score in the prediction of outcomes in upper gastrointestinal bleeding by modifying the AIMS65 score. METHODS: Data were collected retrospectively between January 2015 and June 2015. A total of 212 adult patients, who visited the emergency department with an upper gastrointestinal hemorrhage during this period were included for analysis. High risk patients were defined as follows: those who needed an endoscopic or surgical hemostasis, suffered rebleeding, hospitalized in an intensive care unit, and those who were deceased within 30 days or required a blood transfusion. The seven parameters of the modified AIMS65 score were as follows: Albumin levels, international normalized ratio (prothrombin time), altered mental status, systolic blood pressure, age>65 years, hemoglobin levels, and heart rate. RESULTS: The high-risk group was comprised of 163 patients, while the low risk group was comprised of 49 patients. The areas under the curve for AIMS65 and modified AIMS65 scores were 0.727 (95% confidence interval, 0.662-0.786) and 0.847 (95% confidence interval, 0.791-0.892), respectively, which were significantly different (p<0.001). The AIMS65 score had a sensitivity of 53.0% and a specificity of 78.5% at a score of 0. The modified AIMS65 score had a sensitivity of 22.4% and a specificity of 99.3% at a score of 0. For the modified AIMS65 score of 3 or lower, the sensitivity was 97.9% with a specificity of 21.4%. CONCLUSION: The modified AIMS65 score was effective in distinguishing between the low-risk group and the high-risk group among patients with upper gastrointestinal bleeding.