Conference Coverage

Management of Patients With Upper Gastrointestinal Bleeding

In this podcast, Alan Barkun, MD, speaks about the management of patients with upper gastrointestinal (UGI) bleeding, including risk stratification, pre-endoscopic assessment of patients, endoscopy, and the management of patients following endoscopy. Dr Barkun also spoke about these topics during his session at The American College of Gastroenterology Annual Scientific Meeting 2022 titled “State of the Art Lecture: Risk Stratification and Treatment of UGI Bleeding.”

Additional Resource:

  • Barkun A. State of the art lecture: risk stratification and treatment of UGI bleeding. Talk presented at: ACG 2022; October 21-26, 2022; Charlotte, NC. Accessed October 12, 2022.

Alan Barkun, MD, is a gastroenterologist and clinical epidemiologist at McGill University (Montreal, Canada).



Jessica Bard: Hello, everyone and welcome to another installment of Podcasts360, your go-to resource for medical news and clinical updates. I'm your moderator, Jessica Bard with Consultant360, a multidisciplinary medical information network.

Dr. Alan Barkun is here to speak with us today about his session at ACG 2022 titled, "State-of-the-Art Lecture: Risk Stratification and Treatment of UGI Bleeding." Dr. Barkun is a gastroenterologist and clinical epidemiologist at McGill University in Montreal, Canada.

Thank you for joining us today, Dr. Barkun. Can you please provide us with an overview of your session?

Dr Alan Barkun: Sure, it'd be my pleasure. So perhaps what I'll be doing during the session is talking briefly about risk stratification, and then going over selected aspects of the management of patients who present with upper GI bleeding. Now with regards to risk stratification, I think the key is, and we've known this probably for the last five, six years. That the risk stratification skills are limited in what they can bring to the clinician. However, they are underused, specifically with regards to whom we could go ahead and admit versus actually send home. So actually, there are a number of scales that have been looked at. The Blatchford Score is based on a number of clinical criteria and laboratory criteria that would include patients presenting with melena, with syncope, presence of associated hepatic disease or cardiac failure, whether the patient is tachycardic when they come in, and in addition to that, the systolic blood pressure, hemoglobin and blood urea values. Based on these components, a Blatchford scale score can be derived.

And importantly, patients who are found to be at low risk with a Glasgow-Blatchford Score of zero or one can, in fact, be sent home and scoped electively and not admitted to the hospital. This is important, and indeed the false-negative rate is very low. It's less than 1% actually misclassifying a given patient as a patient who would require a hospital-based intervention. So, it's a very safe, very good risk stratification tool to be able to decide who needs to stay in the hospital for either stabilization and or an urgent endoscopy or what we call an early endoscopy versus someone who can go home and come back to be scoped thereafter.

With regards to the high-risk end of the scale, unfortunately, the risk stratification scores haven't done as well. We have the Rockall Score, the Blatchford Score, the AIMS65, and an Italian score called the PNED, as well as a more recent score called the ABC Score. Some of these actually are not bad in predicting negative outcomes for patients, but none actually has been found to influence the management of patients because they've been classified as high risk at this point.

Jessica Bard: And you also mentioned the management. Can you please give us an overview of the management of UGI bleeding that you'll speak to in your session?

Dr Alan Barkun: Right, very good. So of course, I basically divide a session in the pre-endoscopic assessment and management of the patients, then the endoscopy itself and what is done to handle the patient at gastroscopy. And then, finally, look at the considerations following endoscopy. So common questions would be what about the nasal gastric tube? And we now know that routine insertion of nasal gastric tube is not indicated, mainly because the clot is too thick to be absorbed through the nasogastric tube. And unfortunately, the yield that an NG tube or NG tube lavage provides is not discriminant enough to be able to allow us to manage patients. So that's the answer from that point of view. Another common question is whom to transfuse? We now know based on five randomized trials, and almost 2000 patients, that a restrictive policy of red blood cell transfusion is the best and we recommend as threshold for transfusion a hemoglobin of seven grams per deciliter, aiming for nine grams per deciliter, which results in lowering all cause mortality and further bleeding without increasing the need for interventions.

With regards to patients who have active coronary disease based on very little data, we actually recommend a transfusion threshold of 10 grams per liter. I remind you that there is also, as you're preparing to do the gastroscopy, a role for prokinetic administration, typically erythromycin; there's now a meta-analysis of almost 600 patients and eight studies suggesting that giving erythromycin prior to the gastroscopy results in decreasing the need for repeating an early gastroscopy to find a bleeding lesion because you have improved visualization. The role of PPI's prior to endoscopy remains a little bit controversial. And more recently, in fact, a group of authors who developed guidelines decided they could not make a recommendation on it; much more important is adequate resuscitation of the patient. Indeed, we now know that patients who are high risk, typically American Society of Anesthesiology class three to five, the sickest patients, actually need to be adequately resuscitated and that there is a U-shaped mortality curve with increased mortality over the first six to 12 hours.

So, you must not rush into scoping these patients. These patients must first adequately be resuscitated and then the sweet spot, the window for scoping them, is probably closer to 24 hours, which is an old message the so-called early gastroscopy. With regards to doing an endoscopy in high-risk patients, a recent study published just a year and a half ago or so suggested that even in high-risk patients, you do not gain from doing an even more urgent gastroscopy as early as within six hours or so. So really the take home message, which is an old one, is that you should gastroscope patients within 24 hours and this remains the same. And that includes the patients who are particularly a high risk for a negative outcome at GI bleeding. So much for pre-endoscopy management. With regards to the actual endoscopic component of management for these patients, I remind you that we go ahead and only provide endoscopic hemostatic options to patients who have high-risk lesions.

These include active spurting, active oozing, and non-bleeding visible vessels, as well as adherent clots. All of these will benefit from some intervention. Endoscopic therapy is indicated in all, although the data for adherent clots is not as strong and you may decide to only provide high-dose PPIs for those patients. As to the mechanisms of how to proceed in performing endoscopic hemostasis, there are different tools that can be used. The best data in the literature relate to bipolar electrocoagulation, heater probe, and injection of absolute ethanol, although the latter occurred mainly in Asian studies. And then we still have good data to suggest that we would consider offering through-the-scope clips, argon plasma coagulation, or soft monopolar electrocoagulation in patients presenting with non-variceal upper GI bleeding. And then finally, epinephrine injection can be used but is not as good if used alone as if it's used in association with one of the other hemostatic modalities.

These are not new messages, but are old messages that bear repeating as such. One of the newer kids on the block is the hemostatic powder TC325 or gels. Recent data suggest Tc325 can be used in the context of ulcer disease as well as other conditions that cause non-variceal upper GI bleeding. And in a recent guideline, Loren Laine and colleagues from the American College of Gastroenterology did recommend using it in that context. It probably has an even more useful role in the context of malignant bleeding, but that is a rare cause of non-variceal upper GI bleeding. And more recent than that even is the use of an over-the-scope clip. We now have four randomized trials and a meta-analysis. Of the RCTs, two are published studies, two studies published in abstract form and in fact a third has recently been published as well.

And these all suggest that the over-the-scope clip has a role to play. However, the strongest data are when using these for patients who have recurrent refractory bleeding to initial endoscopic therapy. The data using an over-the-scope clip as primary endoscopic method in patients with non-variceal upper GI bleeding unfortunately is right now based on randomized trials that had some methodological limitations to them. And we have not recommended this option as of now as such. That's about it for the endoscopic therapy, for alternatives you would consider. Following endoscopic therapy, I remind you that PPI's post endoscopic therapy have been shown in high doses to actually result in not only decreased re-bleeding and need for surgery, but also mortality. The best data are for the high-dose PPI bolus followed by infusion for three days. However, the bolus dosing with regards to repeat IV administration of the treatment have also been recommended.

And you can go ahead and give 40 milligrams of a PPI anywhere from two to four times a day with improved outcomes as well. Once the patient leaves the hospital, we recommend for the very high-risk patients to have double oral PPI dosing for the first two weeks after discharge from hospital and then single daily PPI dosing for a duration of time that is proportional or relevant to what was found at the time of endoscopy with regards to the bleeding lesion. Following discharge from hospital or while the patient is in hospital, if a patient re-bleeds, we do recommend before considering other options to go ahead and repeat endoscopic therapy. And this is where the over the scope clip may be useful in select cases in patients who continue to fail endoscopic therapy; of course transcatheter arterial embolization or more rarely surgery should be considered for these patients.

And then finally, before finishing, I would remind everyone that absolutely key with evolving information that has come about and recent guidelines that were just published by the American College of Gastroenterology is the issue of secondary prophylaxis. So first in patients who have bleeding ulcers, always think of looking for H. pylori and H. pylori should be tested for and eradicated to make sure that bleeding does not recur with an ulcer. I remind you that the H. pylori testing has an increased false negative rate if done in the acute setting of bleeding and should be repeated outside the acute setting. And finally, without going into details for this today, any patient who has bled while on single or dual antiplatelet therapy or on an anticoagulant would need to be put on a PPI and needs to stay on the PPI for as long as they will need to be given the single or dual antiplatelet therapy or the anticoagulant therapy. So I think in a short whirlwind tour, this kind of summarizes for you the key aspects of managing patients with non-variceal upper GI bleeding. I hope the presentation is clear enough.

Jessica Bard: Absolutely. That was comprehensive. It was prepping the patient, performing endoscopy, managing the patient following endoscopic therapy, new guidelines. What would you say are the overall take-home messages that if you could sum it up into a couple sentences that you want the audience to know?

Dr Alan Barkun: Sure, Sure. So I think the take-home messages are that the mortality of the patients has decreased due to adequate management of those patients. Key aspects of management would include risk stratification for discharging patients at low risk, going ahead and managing patients appropriately prior to the endoscopy, which includes adequate resuscitation, which is absolutely key, using appropriate endoscopic approaches to perform hemostasis and after admission to hospital, not forgetting the antithrombotic use for patients and secondary prophylaxis.

Jessica Bard: Is there anything else that you'd like to add today that you think that we missed?

Dr Alan Barkun: I could talk forever, unfortunately, but I will not. I think those are the key issues and those are the ones I'll be going over during the postgraduate course.

Jessica Bard: Well, thank you so much for joining us on the podcast today.

Dr Alan Barkun: Thank you very much for having me