Discussion
Acute MI and UGIB alone account for significant morbidity and mortality. When UGIB occurs with acute MI, risk of a poor outcome is elevated and a decision regarding therapeutic procedures, such as endoscopy or cardiac catheterization, becomes difficult. Studies during the past 20 years have examined the safety of endoscopy in the setting of MI.
In 1993, Cappell found that endoscopy in clinically stable patients with MI was safer than in unstable patients with MI. In 1999, Cappell and Iacovone matched controls with patients undergoing EGD with MI and healthy patients undergoing EGD and discovered that EGD in stable MI patients (APACHE score ≤15) resulted in a complications rate of only 2%; however, in seriously ill patients (APACHE score ≥16) who underwent EGD with MI, the complications rate was 21%. Lin et al concluded that urgent EGD in the setting of acute MI appears to be relatively safe, even in patients who are hemodynamically unstable and require intensive care or mechanical ventilation. Cappell and Iacovone suggested that the severity of illness, measured by a higher APACHE II score, identified patients at a greater risk of complications from endoscopy after MI.
This study strongly suggests that EGD is relatively safe for the management of UGIB in patients with acute MI with no major and only minor complications observed. We found that patients with APACHE II scores >16 were more likely to develop minor complications; however, we observed that the type of MI (STEMI or NSTEMI) before cardiac catheterization and peak troponin I do not seem to predict a higher incidence of complications. These findings are consistent with studies showing an increased severity of illness and a higher APACHE II score identified patients at greater risk of complications from endoscopy from MI. Furthermore, this study also demonstrated that patients who underwent cardiac catheterization do not have a higher incidence of complications than those who did not. Despite these results, given the medical comorbidities of UGIB and acute MI together, a baseline risk must be considered.
The strengths of this study include the inclusion of all patients who underwent EGD within 30 days of MI at our institution during the past 11 years, multiple parameters assessed (pre- or post-cardiac catheterization, types of MI, APACHE II scores), and separation of major and minor complications. As with any study, limitations are apparent. First, this is a retrospective, single-center study with a limited number of endoscopies performed during the past 11 years. Many other studies on the subject are retrospective, single-center studies and have a smaller number of patients. Ideally, a randomized controlled trial should be performed on this topic to fully assess safety. Despite this limitation, the information presented by this study may enhance the knowledge regarding this subject.
Second, given the limited number of patients, it is possible that the power of the study may not be adequate to detect a difference within the two populations (type II error). This is an adherent limitation in a single-center, low-volume institution; however, this study is one of the largest on the subject. In the future, a multicenter retrospective study would reduce the chances of type II error bias.
Third, no control group was used in this study. Ideally, a control group of patients with MI would be age matched with patients undergoing endoscopy and MI. Fourth, limited endpoints were evaluated. Other clinical parameters such as left ventricular function, furthering ischemia, and recurrent MI were not included in this article. Further studies may be required to examine these outcomes.
Finally, selection bias may be present given the fact that the endoscopist may elect not to pursue endoscopy in critically ill patients, affecting our results; however, our study did include 27% of patients with APACHE II scores ≥16, indicating that many sick patients underwent EGD.