The COVID-19 pandemic posed a difficult situation for the gastroenterology field due to safety concerns of endoscopy related to aerosolization of the virus. Due to this, many COVID-positive patients were treated conservatively, if possible, when presenting with gastrointestinal (GI) bleeding. As we have now learned, COVID appears to be with us for the foreseeable future. In “Impact of COVID-19 infection on patients admitted with nonvariceal upper gastrointestinal bleeding,” an important question regarding the possible connection between mortality of patients with nonvariceal upper GI bleeding (NVUGIB) and COVID-19 infections was posed.1 From their paper they concluded that for patients admitted with NVUGIB, COVID is an independent predictor of mortality.1 I propose there may be a relationship due to some confounding variables that were not able to be elicited from the retrospective database data, but not necessarily causation.
Using a retrospective database, the authors describe a population that had a principal diagnosis of NVUGIB on hospital admission. It has been noted that GI bleeding in COVID patients ranges from presenting with bleeding and coincidentally being found to have COVID, to presenting with more traditional COVID symptoms of respiratory distress leading to more protracted and complicated courses that include GI bleeding.2 Their study seems to imply the study population was the former. However, due to the retrospective nature, we do not know the clinical presentation of the COVID infections in these patients, such as cough, shortness of breath, or supplemental oxygen use. The authors note that a potential limitation of the study is the lack of COVID severity information. Although they did use the Charleston comorbidity index, having a COVID severity index would have made the findings stronger. For instance, other studies have found the COVID‐19 severity index and Charleston comorbidity index to be high in patients with GI bleeding versus patients without GI bleeding, suggesting that severity of COVID may play a role in the bleeding risk.3
The authors did adjust for confounders such as anticoagulation, organ failure, and preexisting conditions, but I believe some major potential confounders were missed. For example, some of the treatments for COVID can affect bleeding risk, such as corticosteroids, anticoagulation, and tocilizumab.2 Having information on the percentage of patients on these medications would help better understand the risk of GI bleeding associated with these agents and exclude them as the potential causes of the bleeding. This is especially true given we already know steroid use places patients at increased risk of GI bleeding.4
Lastly, the most common diagnosis for patients with NVUGIB in the COVID-19 group was hematemesis (19.70%).1 This, however, is not an etiology, but rather a clinical symptom. There are many etiologies of hematemesis in the GI tract, but many times non-GI causes, such as hemoptysis, can be mistaken for hematemesis. Using data from a retrospective database with discharge diagnosis as a surrogate for cause of death has limitations, as it does not give one all the clinical history and is dependent on the person filling out the discharge summary. Overall, the study provides some interesting data, but due to the inherent limitation of database data, I think some important factors could provide more context to the relationship of NVUGIB.
—Ericka Howard, MD
Division of Digestive and Liver Diseases, Department of Medicine, University of Texas Southwestern Medical Center, Dallas, Texas, USA
Ericka.N.Howard@gmail.com
http://orcid.org/0009-0005-3494-9710
References
- 1.Elfert K, Malik M, Aboursheid T, et al. Impact of COVID-19 infection on patients admitted with nonvariceal upper gastrointestinal bleeding: an analysis from the National Inpatient Sample. Proc (Bayl Univ Med Cent). 2024;37(1). doi: 10.1080/08998280.2023.2260280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cappell MS, Friedel DM.. Gastrointestinal bleeding in COVID-19–infected patients. Gastroenterol Clin North Am. 2023;52(1):77–102. doi: 10.1016/j.gtc.2022.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zellmer S, Hanses F, Muzalyova A, et al. Gastrointestinal bleeding and endoscopic findings in critically and non-critically ill patients with corona virus disease 2019 (COVID-19): results from Lean European Open Survey on SARS-CoV-2 (LEOSS) and COKA registries. United European Gastroenterol J. 2021;9(9):1081–1090. doi: 10.1002/ueg2.12165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Narum S, Westergren T, Klemp M.. Corticosteroids and risk of gastrointestinal bleeding: a systematic review and meta-analysis. BMJ Open. 2014;4(5):e004587. doi: 10.1136/bmjopen-2013-004587. [DOI] [PMC free article] [PubMed] [Google Scholar]