Our next speaker is going to give us a state-of-the-art lecture on risk stratification and treatment of upper GI bleed. We have the very best, Alan Barkin, who's a professor at McGill University from Montreal General Hospital. Thank you. Good morning, everyone. And I'd like to thank the course directors for the privilege of speaking with you this morning. These are my disclosures. I'm going to base... I don't know if I should be talking about guidelines anymore with all that we've heard on guidelines. I will nonetheless base my recommendations to you on guidelines. They include the 2019 Multi-Society, Multi-Country Guidelines published in the Annals of Natural Medicine, as well as more recent guidelines from the ESGE. But I'll focus most of my recommendations today based on the ACG guidelines that were headed by Lauren Lane, an excellent group of collaborators, and more recent guidelines that you may have seen in the Red Journal this year that included Nina Abram being the chair of the guidelines, the full guidelines that were published, as well as a dissemination tool that we published in addition to the guidelines. I'll explain to you why in a little while. And those are the algorithms that I'll show you today. So overall, non-vericellular upper GI bleeding mortality is actually no longer 10%. It's actually lower. Most of the large registries pin it at about 2% or so. It has decreased over the past 10 years. Mortality is principally related to comorbid illnesses. Interestingly, it's no longer cardiac and respiratory. It's malignancy and carotid kidney disease that appear to be related to it. And as you know, most people don't die of the actual bleeding itself, but from the comorbid illnesses. The management, as we all know, is based on endoscopic hemostasis, which is the cornerstone of therapy. PPIs should be shown to be an adjuvant for patients who first have undergone successful endoscopic hemostasis. So where do we stand today as 2022 is coming to an end with the endoscopic management of patients with non-vericellular upper GI bleeding, and in particular, peptic ulcer bleeding? And I'll also review with you the entire management, both prior to endoscopy and following endoscopy as well. So I want to spend a couple of minutes talking about the Glasgow Blatchford score. If you don't use it and you see patients in eMERGE who come in with bleeding, you should be using it. And the reason for that is the following. So first, we need to remember it relates to how patients present with regards to presence of melina and syncope, as well as whether they have comorbid illnesses. And finally, whether they're hemodynamically stable or not, so blood pressure as well as heart rate. And then the hemoglobin and the BUN. And based on this, you can actually devise a protocol or a score which has a direct impact on how you should be managing these patients. Note that just the presence of melina gives you one point. So as you're going to see in a second, a score of one or zero patients could be sent home. So if they have melina, that's it. Nothing else has to be perfect for them to be able to go home, but in fact, they can. So who to admit versus sent home? And we suggest that patients presenting to the ER with upper jab bleeding who are classified as very low risk, defined as a risk assessment score predicting a less than 1% false negative rate for the outcome of hospital-based intervention or death can indeed be discharged home with outpatient follow-up and scoped as an outpatient. And that would be a Blatchford score of zero or one. And unfortunately, very, very few people do this, and we all complain there are no beds in the hospital as well. You could save about 25 to 30% of patients' admissions related to jab bleeding with this simple approach. I will show you the conditional recommendation and level of evidence. You'll see this in the different guidelines that I show you, just to be true to the guidelines papers because those are important and part of the overall picture. We didn't have a guideline on this recently, but briefly, it's a common question we're asked, what about the NG tube? The NG tube is not required in patients with upper jab bleeding, or at least not routinely. Observational evidence does not suggest that the clinical benefit is there if a standard boron is a gastric tube. Why? Because it doesn't allow sufficient clearance of clots to be able to get adequate visualization, nor unfortunately does it allow clinically useful risk stratification. Indeed, the patients with a bloody NG aspirate presence, the negative predictive value of that, so the absence of it, is under 80%, so it doesn't allow us to be able to predict well enough. Who to transfuse? It's an old story, but I remind you because we often see it not being respected. We suggest a restrictive policy of red vessel transfusion with a threshold for transfusion at seven grams per deciliter amongst patients with upper jab bleeding. This is important because the data are very solid, you have five randomized trials, and the restrictive approach actually decreases mortality by a third and decreases further bleeding by about 40%, 50%, without risking differences in complications or an increased need of surgical or radiological intervention. So it's an important old message, but one, again, that is not followed as much as we'd like to see it there with clear benefits. With regards to prokinetic therapy, it's erythromycin, there are very few data on other prokinetic agents, and we do suggest an infusion of erythro before endoscopy is performed. This recent meta-analysis looked at eight randomized trials, almost 600 patients, and you can see that using erythro 45 minutes before you go scope the patient in the acute setting results in improvement in adequate gastro-mucosal visualization, which leads to a decreased need for repeating an endoscopy in the acute setting to be able to find the source of the bleeding lesion. It also results in decreased hospital stay, and there's no, however, improvement in the duration of the procedure, units of blood transfused, or the need for emergent surgery. So we do recommend this. I remind you prolonged QT, so get an EKG done, and of course, time your gastroscopy within 45 minutes of the infusion. Interestingly, pre-endoscopic PPI therapy were often called, and typically the emergentologist or primary care physician will tell you, but it's okay, I've started the PPI. Let's get out of that mindset. The PPIs part endoscopy do very little. In fact, we could not even reach a recommendation whether to give it or not. We actually did a meta-analysis for the guideline itself, and it shows that it does decrease the need for endoscopic hemostatic treatment, because if patients in whom a PPI had been dripping in by the time you go scope them, they have a lesser prevalence of high-risk stigmata, but this does not translate into improved outcomes. And if you run thousands of patients through cost-effectiveness models, it may be cost-effective to do this if your patient is most likely bleeding from a non-variceal etiology, or if it's going to be a while until the patient is endoscoped. But the relationship or the benefit of this is extremely low and should not be prioritized. I'll come back to that in a minute as well. When to endoscope? So this is a key question that comes back to us continuously, and usually we'll answer the patient is too stable so that we won't go in, or the patient is too unstable so we still don't go in acutely, right? This is the usual issue. But we do suggest that patients admitted to under observation in hospital undergo endoscopy within 24 hours of presentation. This is an old message. It relates to potential economic benefits by decreased length of stay, as well as increased benefits with regards to decreased mortality and decreased need for surgery, because these are older studies. However, and I would mention this, based on very little data, but you need to know this, the recommendations if there's a suspicion of variceal bleeding is that the patient should be scoped within 12 hours as such. So let me spend two seconds on the timing of endoscopy, because it's a recurrent issue and there is some new data in the past two years. First of all, who to endoscope? Be aware that there is a U-shaped mortality curve for the sickest patients. This is important. So if you look at the ASA, American Society of Anesthesiology, scores three to five, you can see that the patients are scoped within six to 12 hours, there's actually an increased mortality. So you can have a sweet spot 12 to 24 hours, and then the mortality goes up again. This presumably reflects patients who have not been adequately resuscitated. So do not go in, in the sickest patients, early on thinking that you need to go in until they're adequately resuscitated. Furthermore, there are also recent data, and two years ago, New Glen Journal by James Lau and colleagues from Hong Kong, looking specifically at patients who are at high risk in upper GI bleeding, a Blanchard scale greater than 12, so very sick patients. And they said, you know what, we have a tendency wanting to go in, so let's go ahead and compare them with early, what they called urgent endoscopy within six hours, versus the usual early, which is 24 hours. Over 500 patients mainly bleeding from ulcers. Time to endoscopy, this is pretty good. In real life, this is hard to achieve. Within 10 hours of randomization in the urgent group versus the early group, look, even more than 24 hours. So you're really comparing two very distinct groups amongst who are the sickest patients. No difference in mortality, no difference in further bleeding. Interestingly, we did see higher risk of ulcers present, and more endoscopic hemostasis because the patients were scoped earlier, but that does not translate into improved outcomes. So you do not need to feel pressured into going early on, particularly in sick patients, particularly if they have not been adequately resuscitated. Let's make that very clear to everybody. Moving to 2D endoscopy, this is interestingly an old story that needs to be reminded, is that active spurting, active oozing, or non-bleeding visible vessels, but we actually could not reach a recommendation for patients with adherent clots. So when you see a patient who has an adherent clot, you may want to just put the scope down and just give PPIs, that is acceptable, or you may want to go ahead and do endoscopic therapy based on older, randomized trial data, that is also acceptable. We could not reach a recommendation to clearly guide you on that. If you are going to treat patients endoscopically, what do you use, well, this is an older story, but I just want to go over it with you so that we're very clear. Bipolar electrocoagulation, you can see the first three have actually moderate level of evidence which allowed us to make a strong recommendation, that includes bipolar electrocoagulation, heater probe, and interestingly, mainly from Eastern countries, from Asia, injection of absolute alcohol, that works very well. The next group, the data are not as good, which is why we made a conditional recommendation, that would include a through-the-scope clip, the APC, as well, and more recently, the soft monopolar electrocoagulation, so the grasping force that you may want to use to raise and burn or to actually do coaptive electrocoagulation by putting pressure. And finally, I remind you that with regards to epi alone, it's better than doing nothing, but not as good as any of the others alone or combinations, all right? So that's an older message, but just be aware of that, because we still see that done in places that could do better. What about hemostatic powders and gels? They've been around for a while, but not everybody has a chance to use them. We suggest endoscopic hemostatic therapy with this TC325, which is the hemo spray, for patients who have actively bleeding ulcers. One of my take-home messages to you is it does not work in patients unless they have active bleeding, so do not put it on a patient who has a non-clearing visible vessel or a cloth. It'll just wash off. This was a conditional recommendation based on very little quality evidence. This is vexing, because we have a study that may never be done again that was well done, again, by James Lau in Hong Kong. It was a non-inferior randomized trial that compared TC325 to basically clipping or contact thermal coagulation with or without epi. Most of the patients, over half of them, were bleeding from ulcers, and they actually found that it was non-inferior, meaning you could use TC325 alone for ulcers. The problem is that there were methodological issues. There was an imbalance in the patients who bled from malignancies between both groups, and malignancy seems to respond very well to TC325. I'll come back to that in a short while. So that best study does not provide us a clear answer. I personally do not recommend using it alone, because it doesn't make biological sense, because the product washes off within 12 to 24 hours, and we know that the risk period for ulcers is at least within 72 hours following endoscopic therapy. But the data are there, and this is why the recommendation was made as it is, as it stands right now. There is an issue of cost as well, particularly in the United States, but also in a few other places, which is why we did not recommend it as initial modality. So I feel there's still some issues about hemostatic powders, particularly hemo spray that could be used. Be aware of it, but certainly in the context of rescue and in non-ulcer bleeds that have low risk of recurrent bleeding, absolutely. I thought I'd share with you, just to highlight a couple of things for those of you who may not have had much experience using it. So this is a spray. It's very finicky. If it touches water or humidity, it'll clot, so you have to be very careful and dry off the delivery catheter. You can see here a very large malignancy at the GE junction. We're in retroflex view, and the spray is very useful because you have multiple points of bleeding. Anywhere you're going to touch is going to bleed more. This is non-contact and can also cover a very large surface area quite quickly without causing mucosal damage. So those are the advantages of it, and this is why, and I'm hoping that you will hear within the coming year, year and a half, some very good results looking at TC325 or hemo spray in the context of malignant bleeding. We did not have very good endoscopic therapies available to us, and data from two studies have now shown that it's very good at achieving immediate hemostasis and actually surprisingly decreases re-bleeding at 30 days. So just be aware of that potential role in malignant bleeding. What about the over-the-scope clips? So this is really one of the key hot areas right now. So I'm going to be a party pooper, and I'm going to tell you what I'm going to tell you. So we suggest over-the-scope clips as a hemostatic therapy for patients who develop recurrent bleeding due to ulcers, 100%. One small study, but well done, clearly showed benefit. The question is whether we can use it as a primary therapy or not. So let me briefly tell you. One meta-analysis that looked at 10 studies, almost 1,000 patients. Problem is that they included randomized trials. There were four of them in this particular meta-analysis, as well as observational studies. And it does not allow us to conclude, even though it suggested that it had a role to play overall. If we look at rescue therapy, as I mentioned to you, a well-done study, a small study, but clearly showed improvement in the risk of further bleeding for patients. So clearly as a backup, by all means use it if the patients have had recurrent bleed. The issue is primary therapy. We now have four randomized trials. Two are fully published, two are not fully published. The two that were fully published had methodological issues. I'm happy to answer questions in the discussion section if you have them. I do not feel that they can recommend, based on these at this point, that we should be using over-the-scope clips routinely or even selectively, except for occasional issues for patients as a primary method of treatment, particularly for ulcers at this point. But I'm happy to discuss this with you further. And finally, there is a cost issue involved, and we're lacking some cost-effectiveness data if it is going to be used as primary therapy. So but of course I'll show you. This is from Amrita Sethi. It's a very nice video. It highlights a number of things. This is a patient who presented, had endoscopic therapy, had a duodenal ulcer, endoscopic therapy, re-bled, went to interventional IR with embolization, with coils. The patient then re-bled again. So they brought him back to endoscopic therapy, and you can see here, because I think it highlights a number of issues of the over-the-scope clip. These are actually the coils that were put in at interventional radiology with embolotherapy. So you can see here. It looks pretty benign, no problem at all, but you're going to see very quickly that in fact this is a very ticklish lesion and starts bleeding very actively. So as you know, for those of you who've used the over-the-scope clips, you can try and grab the lesion and bring it into the cat, or you can try and suction it in. Grabbing it did not work, but suctioning it, as you're going to see, is going to work very well, but it's going to highlight how some of these lesions, particularly posterior duodenal bulb, where the pancreatic duodenal artery runs in this case, is very, very sensitive. Just by going ahead and putting suction, you're going to see that bleeding is going to start and it's going to be quite significant. I remind you that even significant bleeding is not a huge issue. It's small. It looks impressive, but if you know what you're doing, you just take your time and proceed accordingly. So you can see quite a bit of bleeding here. Over-the-scope clip is going to be released, and it highlights very nicely the full thickness bite that it performs, because you can actually see the coils coming out even a little bit more towards us here. That's why I thought it'd be nice to kind of show this. I thank Amrita for this pretty amazing video. I'm just trying to move on. Sorry, I don't seem to be able to find... Oh, there we go. Perfect. Okay. A couple of messages before I finish. So PPI is post-endoscopic therapy. This is an old story. We recommend high-dose PPI therapy. We can argue forever as to whether you should give IV high-dose infusion continuously or intermittently for three days. I'm happy to discuss it with you. The best data are with the high-dose IV infusion, but there are good data as well with continuous infusion for three days. This is something you may or may not know about. Eventually, when your patient goes home, do they have to be on single or double oral daily dose therapy? There's actually one small randomized trial that looked at very high-risk patients, found that when the patient goes home after three days of PPIs, if the first 10 days is double PPI oral dose, then they actually do better than if they're on single oral daily dose. And then you complete the duration of either one month if it's a duodenal ulcer, two months if it's a gastric ulcer, two months if it's erosive esophagitis of PPI. So they're kind of a small message to give in this context. If a patient re-bleeds, I think it's very clear now, we do recommend to undergo a repeat attempt at endoscopic therapy. As you can see, we had a different type of data based on comparison with transcatheter arterial embolization or surgery, but the bottom line is you can actually effectively treat an additional 75% of patients if they re-bleed, and you do, again, endoscopic therapy, and that may be a role for the over-the-scope clip, absolutely. So just be aware of that. I would say to you, a small signal, if you've done, or if someone has done thermal therapy for the patient, and the patient re-bleeds, move away from thermal therapy as a second approach, if you have to repeat it, because there is an increased risk of perforation. And what about a patient who's failed endoscopic therapy? Usually you have to repeat endoscopic therapy at least a couple of times. We do recommend treating with transcatheter arterial embolization. The data actually is better with surgery than TAE, but these were in older studies per se, and TAE actually has significantly less complications. So this is, again, not something new. Okay, before I finish, special issues relating to peri-endoscopic management of patients with acute bleeding on an antithrombotic. So it's actually very simple, and it won't surprise anybody, but hopefully it may answer some of the questions that you have. So you have a patient who has an upper GI bleed, and this, by the way, it can be for upper or lower, but you'd look at the severity of bleeding. If patient has a life-threatening bleed, and the definition is below, and you can see it's basically a patient who's bled at least five grams per liter, who's had five units of blood, right, I kind of find it interesting to know that, or a bleed causing death. Well, we agree that's a significant life-threatening bleed, but kind of game is over by then. But anyways, those are the definitions that were adopted in certain guidelines. So life-threatening bleed, and the patient is on a VKA, consider PCC, so the prothrombin complex concentrate, you can consider it. If the patient is on a DOAC, you can go ahead and consider a specific reversal agent, and perhaps PCC as well. So these are patients who are massively bleeding, we don't have much time, and I respectfully argue for you that in this context, these patients will already have been tackled by the emergentologist, the intensivist, and so on by the time you see them. With a non-life-threatening bleed, however, for patients who are on a vitamin K antagonist, we don't recommend any of this, all right. So just be clear about that, and could not make a recommendation for PCC or not, but really recommend staying away from it. And for a DOAC, we did not recommend either. So there's very little to do for patients with regards to altering the coagulation status, except in rare circumstances. We did favor PCC over FFP if you are going to use it, although we didn't recommend to use it as a routine basis, because of issues in volume administered and other complications related to the transfusion. Imagine patients on antiplatelet agents, it's very straightforward. Do not do platelet transfusion, it actually increases complications. And if patients are on an aspirin, we recommend not to stop it, but you should continue it or restart it as soon as possible after endoscopic hemostasis. And finally, secondary prophylaxis, H. pylori eradication, don't forget that. Don't forget HP testing and the false negatives in acute context of bleeding. And for patients who are on dual antiplatelet therapy or on anticoagulant therapy, if they need to remain on this, they need to be on PPI for as long as they take this agent. So in conclusion, some recommendations, old ones and new ones. The Blatchford scale, I spent time discussing this with you, consider it for early discharge. Transfusion practice, transfusion threshold. Use of prokinetics, no PPI prior to endoscopy. The timing of the endoscopy can remain at 24 hours. Do not treat with injection of epinephrine alone. All the other modalities alone or in combination appear better. You can now use soft monopolar electrocoagulation if you want to. Possible limitation of TC325 for PUB, but keep in mind and you will see data, new data on malignant bleeding. The PPI story is an old one. The over-the-scope clip, I spent some time discussing it with you for recurrent bleed. Yes, for primary therapy, we will have to talk about it. Consider TAE for failures and remember the new updated management of patients who bleed on IT thrombotics. I thank you very much for your attention.

upper GI bleed

endoscopic hemostasis

Glasgow Blatchford score

outpatient follow-up

endoscopic treatment modalities

antithrombotic therapy