Introduction: Patients hospitalized with myocardial infarction (MI) are at risk of gastrointestinal bleeding because of the need for antiplatelet agents and/or anticoagulation. The data regarding the safety of endoscopy after MI are limited. This study sought to assess mortality rates of patients hospitalized with acute MI who require esophagogastroduodenoscopy or colonoscopy using the National Inpatient Sample (NIS) database.
Methods: A retrospective cohort analysis of all adult inpatients in the NIS from 2016 admitted for ST-elevation infarction myocardial infarction (STEMI), non-STEMI, or type II non-STEMI was conducted. Data were collected including patient demographics and indication for endoscopy per ICD-10 coding. HCUPnet was used to query NIS to obtain all inpatient mortality. The primary methods included adjusted χ for categorical outcomes, adjusted linear regression for continuous outcomes, and adjusted logistic regression for multivariable analysis.
Results: A total of 1,281,749 patients were admitted for acute coronary syndrome in 2016, and 55,035 of these patients underwent endoscopy In the multivariable regression analysis, those who underwent a GI procedure (odds ratio [OR] 0.80, P value < 0.002) and angiogram (OR 0.48, P value < 0.001) had lower in-hospital mortality, after adjusting for age, Elixhauser index, need for angiogram, sex, race, and hospital type. Endoscopy postcatheterization was not associated with a difference in mortality compared with preangiogram (OR = 0.84, 95% confidence interval 0.60-1.19).
Discussion: Patients who underwent endoscopy are sicker and have higher mortality rates than those who do not undergo endoscopy, but after adjusting for comorbidities, mortality is actually lower. This suggests that endoscopy is safe and should be performed when clinically indicated despite recent cardiac ischemia.