Visibility can be an issue during emergency endoscopy for upper GI bleeding, but it can be improved by nasogastric lavage or prokinetics such as erythromycin. Factors predicting poor endoscopic visibility—and therefore the potential need for erythromycin—have been highlighted in a new prospective study.
Acute GI bleeding is a common major medical emergency. In a 2007 UK-wide audit, the overall mortality of patients admitted to hospital with acute GI bleeding was 7%.1 By contrast, the mortality of patients who bled during admission to hospital for other reasons was 26%.
Emergency upper GI endoscopy is indicated for patients with significant Rockall or Blatchford scores.2,3 However, the effectiveness of upper GI endoscopy as an emergency treatment for patients with significant GI bleeding can be limited by poor visibility due to the presence of luminal blood and other gastric contents. Nasogastric (NG) lavage and prokinetics such as erythromycin have been shown to improve visibility and decrease the need for a second-look endoscopy.4–7 Administering 3mg/kg erythromycin IV slowly over 30 minutes and then waiting up to 90 minutes for the infusion to take effect can, however, delay the emergency procedure. Furthermore, the risk of inadequate visibility is generally not high in this setting, so knowing who is most likely to benefit from IV erythromycin beforehand would be desirable.
Jung et al. from Yonsei University College of Medicine, Seoul, South Korea, conducted a prospective observational study to identify factors affecting visibility during emergency endoscopy for upper GI bleeding (within 24 hours of presentation).8 The study authors also integrated these factors in a regression model that indicate the likelihood for poor visibility and this could have potential for predicting the need for erythromycin.
The primary endpoint of this study was a ‘visibility score’ applied to four parts of the upper GI tract: the fundus, body, antrum, and duodenal bulb. A score from 0 to 2 was given to each part, depending on the degree of mucosal visibility (a score 0 was given for areas where less than 25% of the mucosa was visible), with a total score between 0 and 8. Good visibility was a score of 7 points or more.
Multivariate logistic regression analysis was used to determine the factors independently associated with visibility, including age, sex, smoking, aspirin and anticoagulant use, rectal examination findings, blood pressure, heart rate and blood test results.
The appearance of the NG tube aspirate (blood or coffee grounds), the emergency endoscopy being scheduled within 8.5 hours of presentation to hospital and blood urea levels were the variables that had a statistically significant association with endoscopic visibility. However, only the first two variables were incorporated in an algorithm that could predict visibility prior to endoscopy—the CART model.
Using the CART model, good visibility was predicted when the NG aspirate was clear. If the NG aspirate was not clear, patients scheduled for an emergency endoscopy more than 8.5 hours after presentation were also likely to have good visibility
The study does have some limitations: single centre, observational, small sample size (114 patients), and variability in calculating the visibility score amongst different endoscopists. Furthermore, the aim of the study was simply to score the visibility. The study did not address the real question of how many patients required a second endoscopy because the site of bleeding could not be identified at the index examination.
This study confirms that IV erythromycin is most likely to be of benefit for those patients who have more severe bleeding. Paradoxically, these are the patients for whom we may feel the least confident delaying the emergency procedure whilst a nasogastric tube is passed and IV erythromycin is prescribed, administered and then allowed to take effect.
It would be simple enough to incorporate erythromycin into our routine practice when it comes to the endoscopic treatment of acute upper GI bleeding, but randomised prospective studies are needed before we do!
- British Society of Gastroenterology. UK comparative audit of upper gastrointestinal bleeding and the use of blood (2007, accessed from http://www.bsg.org.uk/clinical/general/uk-upper-gi-bleeding-audit.html April 2015).
- Blatchford O, Murray WR, Blatchford M. A risk score to predict need for treatment for upper-gastrointestinal haemorrhage. Lancet 2000; 356: 1318–1321.
- Rockall TA, Logan RF, Devlin HB, et al. Risk assessment after acute upper gastrointestinal haemorrhage. Gut 1996; 38: 316–321.
- Altraif I., Handoo FA, Aljumah A, et al. Effect of erythromycin before endoscopy in patients presenting with variceal bleeding: a prospective, randomized, double-blind, placebo-controlled trial. Gastrointest. Endosc. 2011; 73: 245–250.
- Coffin B, Pocard M, Panis Y, et al. Erythromycin improves the quality of EGD in patients with acute upper GI bleeding: a randomized controlled study. Gastrointest. Endosc. 2002; 56: 174–179.
- Carbonell, N., Pauwels, A., Serfaty, L. et al. Erythromycin infusion prior to endoscopy for acute upper gastrointestinal bleeding: a randomized, controlled, double-blind trial. Am. J. Gastroenterol. 2006; 101: 1211–1215.
- Huang, E.S., Karsan, S., Kanwal, F. et al. Impact of nasogastric lavage on outcomes in acute GI bleeding. Gastrointest. Endosc. 2011; 74: 971–980.
- Jung S, Kim E, Kim H, et al. Factors that affect visibility during endoscopic hemostasis for upper GI bleeding: a prospective study. Gastrointest Endosc Epub ahead of print 12 March 2015. DOI: 10.1016/j.gie.2014.12.024.