Is it safe to perform endoscopies in patients who have recently experienced an acute coronary syndrome?

In this national database study, there was no association between having an endoscopy while hospitalized for an acute coronary syndrome and increased in-hospital mortality. (LOE = 2b)

Hoffman GR, Stein DJ, Moore MB, Feuerstein JD. Safety of endoscopy for hospitalized patients with acute myocardial infarction: a national analysis. Am J Gastroenterol 2020;115(3):376-380.  [PMID:32022723]

Cohort (retrospective)

Self-funded or unfunded

Inpatient (any location)

These researchers used a national database of inpatients to identify all adults who experienced an acute coronary syndrome (ACS) and who also had an endoscopy (colonoscopy, esophagogastroduodenoscopy, or both) during that admission. In 2016, nearly 1.3 million patients were admitted for ACS, of whom 55,035 (4.3%) underwent an endoscopy. Among these patients undergoing an endoscopy, approximately 60% had a non-ST elevation myocardial infarction (NSTEMI) and nearly 9% had an acute myocardial infarction (AMI). The unadjusted (crude) in-hospital mortality rates for those undergoing an endoscopy was 9.9% compared with 7.7% in those who did not. However, after adjusting for age, comorbidities, and other factors that might influence the rate of deaths, the odds of dying in the hospital were actually lower after undergoing an endoscopy (odds ratio 0.80). The authors also performed an additional comparison of the 5050 patients who also underwent an angiogram and had an endoscopy with the 6185 who had an angiogram but did not undergo an endoscopy. They reported that of those who underwent endoscopy first 7.5% died compared with 6.7% of those who underwent catheterization first and 3.4% of those who did not have an endoscopy. After controlling for the same aforementioned factors in these patients, the authors found no association between endoscopy and in-hospital mortality. It is likely that more is going on in these circumstances than can be determined by this kind of analysis. On the surface, these patients appear to be "sicker" using the authors' measures; it is possible that these patients were highly selected by the endoscopist using other measures such as gestalt, cardiac ejection fraction, or other markers of fitness. Judgment isn't a bad thing, it's just hard to measure!