Acute Myocardial Infarction and Gastrointestinal Bleeding in an African American Inner City Population

Abstract

Purpose: Patients with concomitant gastrointestinal bleeding (GIB) and acute myocardial infarction (AMI) have been well described in literature; however the majority of these studies were performed in non-minority populations. The aim of our study was to characterize patients with GIB and AMI in a predominantly underserved, minority, African American inner city population. Methods: A retrospective chart review was conducted at two large teaching hospitals of a major university medical center in Brooklyn, NY. Patients were identified by using ICD-9 codes for GIB (occult and overt) and AMI, admitted 2001 through 2007. A total of 115 patients were identified, of which 102 (88.7%) were African American. This subset of patients was analyzed for patient characteristics, co-morbidities, endoscopy results as well as clinical outcomes. Results: The median age was 71.4 years, 59.8% were women. The co-morbidities most frequently identified were: arterial hypertension (73.5%), diabetes mellitus (35.3%), congestive heart failure (21.6%), history of CVA (16.7%), history of coronary artery disease (36.3%), and chronic kidney disease (22.5%). 36.3% of patients were taking aspirin, 22.6% plavix, and 4.9% coumadin. 74.5% of patients presented with overt GIB. Median peak troponin was 54.1 and the median of the lowest hemoglobin was 7.6 g/dl. 62 patients had endoscopic procedures (52 upper endoscopies and 18 colonoscopies). Endoscopy revealed 16.7% gastric ulcers, 6.7% duodenal ulcers, 3.3% Mallory Weiss tears, 16.7% esophagitis, 40% gastritis, 11.7% duodentitis, 1.7% gastric cancer, 1.7% variceal bleed and 1.7% Dieulafoy lesion were found. Ulcers were described as: clean base (62.5%), adherent clot (25.0%), visible vessel (6.3%), active bleeding (6.2%). 7 repeat endoscopies and 2 surgeries were required for rebleeding. The majority lesions found at colonoscopy included internal hemorrhoids, rectal ulcers, vascular malformations, diverticulosis. There were 19 cardiac catheterizations performed. Occult GIB was associated with significantly less mortality than overt GIB (3.8% versus 19.7%, P < 0.05). Although our patient population had a significant amount of chronic kidney disease, this did not affect the mortality rate (P-0.28). Conclusion: Inner city, African Americans presenting with acute MI and GIB have a significant amount of arterial hypertension and chronic kidney disease, but a surprising low rate of variceal bleeding. The type of GIB, rather than comorbid conditions significantly affected patient mortality. The endoscopic findings were consistent with the trend recorded in the literature showing a decrease in the prevalence of peptic ulcer disease.