Incidence, Trends, Etiology, and Inpatient Mortality After Acute Upper and Lower Gastrointestinal Tract Bleeding in STEMI versus NSTEMI

Abstract

Introduction: To investigate the incidence, national trends, cause of bleeding and inpatient mortality after a gastrointestinal bleeding in STEMI versus NSTEMI in USA from 1993 to 2012. Methods: The Healthcare Cost and Utilization Project national database from 1993 to 2012 representing 20 % of the hospitals of United States was used for this study. Patients with gastrointestinal bleeding in STEMI and NSTEMI were compared in logistic regression models after adjusting for demographic, medical risk factors, hospital properties and acute myocardial infarction complications. Results: From the year 1993 to 2012 a total of 2,115,207 patients had STEMI and a total of 1,806,756 patients had NSTEMI. Among them 20,523 STEMI and 35,108 NSTEMI patients had Upper GI (UGI) bleeding and 17,697 STEMI and 32,926 NSTEMI patients had Lower GI (LGI) bleeding. There has been approximately a 700% and 50% increase in the incidence of UGI and LGI bleeding respectively in patients with NSTEMI from 1993 to 2012. Whereas there has been approximately a 50% decrease and a 200% increase in the incidence of UGI and LGI bleeding respectively in patients with STEMI from 1993 to 2012. The most common etiologies in order of frequencyes for UGI bleeding were chronic gastric ulcer, chronic duodenal ulcer and esophagitis and for LGI bleeding they were hemorrhoids, benign colonic polyps, diverticulosis, ischemic and ulcerative colitis. The odds ratio for UGI bleeding in NSTEMI versus STEMI was 1.469 (1.443 - 1.494) and the odds ratio for LGI bleeding in NSTEMI versus STEMI was 1.598 (1.569 - 1.628). Among STEMI patients with UGI bleeding, 4,364 (21.29%) died and among NSTEMI patients with UGI bleeding 3.954 (11.28%) died. However among STEMI patients with LGI bleeding 3.954 (16.94%) died and among NSTEMI patients with LGI bleeding, 2,934 (8.92%) died. The odds ratio for in patient mortality in STEMI versus NSTEMI for UGI bleeding was 2.127 (2.092 - 2.229) and the odds ratio for in patient mortality in STEMI versus NSTEMI for LGI bleeding was 2.083 (1.972 - 2.200).The odds ratio for in patient mortality in STEMI versus NSTEMI for combined Upper and Lower GI bleeding was 2.208 (1.687 - 2.890). Conclusion: Even after adjusting for potential confounders there was a clinical significant increase in the rates of Upper and Lower gastrointestinal bleeding during NSTEMI versus STEMI. The inpatient mortality rate was clinically significant in STEMI versus NSTEMI in both Upper and Lower gastrointestinal bleeding.