Upper Gastrointestinal Bleeding Mortality Risk Changes in Patients With Comorbidities Over Three Decades

Abstract

Purpose: Our objective was to show the changes in the mortality risk of each comorbidity used in the Rockall score in patients with upper GI bleeding (UGIB) during three decades, using a nationwide representative data. Methods: We analyzed the National Hospital Discharge Sample (NHDS) for UGIB outcomes through the last three decades from 1979 to 2009. We identified the patients with the primary ICD-9 codes representing a diagnosis of UGIB and then calculated the mortality risk, crude odds ratio and adjusted odds ratio of each comorbidity used in the Rockall score in each decade. Results: There is a dramatic decline in mortality risk of patients with renal failure, starting from 50% in the first decade and ending up 4.0% in the third decade. Mortality risk of patients with CHF also had a substantial decrease from 17.9% to 5.2% in the first and third decades, respectively. Patients with IHD, cancer, and liver failure all showed decrease in the mortality risk too. Overall adjusted odds ratio of UGIB mortality indicates that liver failure, CHF, renal failure and cancer are significant risk factors for UGIB. The decreasing trend of odds ratio for patients with CHF (P<0.001), IHD (P<0.02) and renal failure (P<0.001) were statistically significant in this period. In esophageal bleedings, there was an increase in the mortality risk for patients with renal failure, while other comorbidities had a decreased risk from the second to third decade. Mortality risk of gastric bleeding for patients with renal failure has had a dramatic change, starting with 37.5% in the first decade, and ending up 3.7% in the third decade. Mortality risk of patients with CHF had a significant decline from 15% in the first decade to 4.1% in the third decade. Other comorbidities did not show such significant changes in their mortality risk in this period. Duodenal bleeding mortality risk in patients with renal failure plummeted from 47.4% in the first decade to 3.4% in the third decade. CHF patients showed a significant decline in this period, with 20.7% to 8.4% in the first and third decades, respectively. Patients with liver failure, IHD and cancer did not show significant changes in the mortality risk in this period. Conclusion: Critical care improvements in patients with comorbidities may have reduced mortality risks, particularly in renal and heart failure cases. These reductions are less in other studied comorbidities, which might be related to the pharmaceutical and procedural risks for causing UGIB. Despite reduction in the mortality risk, extra caution should be taken in the management of UGIB cases with the studied comorbidities due to higher risk of mortality in them.Table: Table. UGIB mortality risk and adjusted odds ratio for different Rockall score comorbidities