 Okay. Good morning everybody. Today I'll be presenting an approach for upper GI bleed. Before starting, I would like to wish DJ and Leon for passing their intermediate exam. And I wish all the best for Anthony next time to join us. Okay. So this is quite a big topic and it's a very common problem we face in our emergency department. So upper GI bleed is very common, medical condition with high mortality and morbidity. And commonly they present here with, usually with hematemesis. Very rare you get them with melina, sorry, you get them with melina, but very rare you get them with hemategesia. So these are some common causes which I'm not going to discuss, but these are some common causes in upper GI bleed, which were discussed I think previously in the previous presentation with Anthony. So gastric and duodenal ulcer is figal varicis, esophagitis, malary waste, angiotysplasia, delifoil lesion, and sometimes we know no lesion identified. Regarding delifoil lesion, this is a submucosal dilatation of vessels which tend to erode into the epithelium and bleeds. This is a picture shows the delifoil lesion. This is small, as you see in the center, this dilated small blood vessel. Then this is gastric entral vascular lactacea, which is known also as watermelon gastritis. Again, it's dilatation of blood vessels which gives watermelon look like. Okay, so some other less common causes are hemabilia, which is presence of blood into biliary system. Usually there will be some liver or biliary pathology like liver tumor or instrumentation to the biliary system. Hemococcus pancreaticus, presence of blood into the pancreatic duct. Again, it could be because of tumor pancreas or pseudocyst or any cause of bleeding into the pancreas. Our two intracrystallis and chameron legion. Our two intracrystallis is most common in the third part of the odenum or fourth part of the odenum. Chameron legion is just an ulcer or erosive legion in the sac of hiatal hernia, which is very less than 5%. Okay, this is general approach to upper GI bleed. The initial evaluation of patient includes assessment and diagnostic study. In assessment of the patient, you are looking for stability of the patient, whether he's stable or not. Then you go for diagnostic study that the choice is in the scope to check where is the bleed from and to intervene. Okay, coming to the initial evaluation, you assess the severity of the bleed, identify the potential source and determine if there are any conditions which can be managed by in the scope. Okay, part of that management is your history, history taken from the patient. 60% of the patients tend to bleed from similar legion. If they have history of previous bleed, they tend to actually bleed from the previous legion. And your history can't suggest the source of bleed. Like patient with portal hypertension, you suspect these figal varicis. Patient with like aortic repair previously, like aortic graft or aortic aneurysm can't suspect our two intracrystallis. Then other things like epigastric pain can't suggest like optical cell disease and so on. Okay, in your history, you are supposed to take as well, you should consider your comorbid condition because that will affect your management. Like patient who have coronary artery disease, you should take care of their hemoglobin concentration that you keep it high up, plus they should get some oxygen. Patient with CCF or some renal failure, you should be very cautious with your fluid. Okay, then patient with hepatic dysfunction, your management be affected by giving them like fresh frozen plasma, you should consider it. Then some patient may need intracal intubation if they have dementia or encephalopathy. Okay, then other part of history is your medication. You should know whether your patients are taking NSAIDs, which can be the reason for their ulcers. Maybe they are taking warfarin for some or any antiplated therapy for some condition that could be the reason for their bleed. And other part of history, you should know like if patients taking iron and the color of the melina is just an iron related. Okay, this already I have mentioned symptoms assessment of the patient. Okay, one more point in the symptoms is regarding whether the patients are in clinical condition, whether they have hypotension, dizziness or they have reduced mental status. So that can indicate severity of bleed. Okay, coming to physical examination, as everybody knows signs of hypovolemia, if it's moderate or mild, the patient it will be tachycardic only. If it's more than 50% of blood loss, orthostatic hypotension, and more than 50% or 40%, there will be hypotension supine hypotension. So examination, while doing PR, you examine the stool color, but this is not very reliable thing. Then presence of abdominal pain or guarding or anything which suggests prodenitis can suggest that patient may have perforation and should be worked up for that, like doing an x-ray, looking for air under the diaphragm, and maybe CT scan. Okay, lab tests includes your blood counts, serum chemistries, liver tests, coagulation studies. Okay, it's a patient with coronary heart disease or risk of infection, get electrocardiogram. Then it's important to get the urea nitrogen and urea, because having a high ratio of BU and to keratinine or urea to keratinine is very suggestive of upper GI bleed. Nezogastric lavage is controversial in diagnosis of upper GI bleed. There was a study done regarding that, looked whether there would be any clinical benefit in angi lavage. 632 patients were admitted with the bleed, and some of them went into angi lavage and some of them didn't. The angi lavage was associated with shorter endoscopy time, but there was no difference in regard of mortality and length of hospital stay or surgery, or even transfusion requirement. Okay, then coming to risk stratification, endoscopic clinical and lab features may be useful for risk stratification. There are certain factors used for that, like hemodynamic instability, like systolic blood pressure less than 100 or heart rate more than 100, hemoglobin less than 10. If there is any active bleed during endoscopy, ulcer size if it's ulcer is more than 3 cm, then ulcer location if it's duodenal bulb or it's indelizer curvature. Okay, I'll mention three scoring system, that's Rokal system and BlackScore system and AIMS 65. Coming to the first one, which took in concentration that endoscopic finding during the endoscopy and as well as age of the patient, hemodynamic shock and major comorbid illnesses. So each factor was given some points and a score of 2 or less was associated with low risk of further bleed or death. But I have no idea if the score comes like 10 where we stand, I'm not sure. Okay, then that's the BlackScore system. In this one, they didn't take any endoscopic finding in scoring and the score of 0 was associated with low risk of need of endoscopy intervention. So they took in concentration in blood urinary region, hemoglobin, systolic BP and other markers. This score system looks easier to implement in emergency than the other one. Okay, coming to AIMS 65, which is the easiest to implement, the studies has showed that it has high accuracy. Five factors were taken that include albumin and AIMS 65 actually is kind of mnemonic for the factors we are using. Albumin less than 3 gram per deciliter and INR more than 1.5, altered mental status, systolic blood pressure of 90 or less and age less than 65. So this is the scoring system and interpretation of this system is like this. If you get zero risk factor, so chance of mortality is 0.3%, one risk factor is 1% and so on. Easy way to remember this is 0 into 0 is equal to almost 0. Then 1 into 1 is almost 1. 2 into 2 is 4, it's almost 3. Then 3 into 3, 3 into 3 is almost 9. Then 4 into 4 is almost 16 and 5 into 5 is almost 25. So you can't remember, if you remember it in this way, it's the easiest to remember that. And that's my discovery by the way. I didn't read it. Okay, general management. So this is general points you have to do when you manage a patient of GI BLEE. First you have to try out him whether he needs an ICU or ward admission. Then what kind of support he needs? If he has coronary artery disease, he needs oxygen, then you should have two large board cannulas whether he needs intubation or not. Then fluid recess, he gets like any other hypotensive patient, he gets fluid recess accordingly. Blood transfusion, medication and upper endoscope. So I'll go through last three points more extensively now. Coming to blood transfusion, the slide looks very congested but it's very easy. This blood transfusion decision is individualized. Okay, so usually the recommendation is to initiate transfusion if your hemoglobin falls less than 7 gram per DC liter. Okay, except in cases with patients who have comorbid illnesses or coronary artery disease, that time you fall below 9 should initiate blood transfusion. Okay, you should avoid over transfusion in various bleeds. Okay, what are the studies done in this regard? RCT then suggests that using lower hemoglobulin threshold for initiating transfusion improves outcome. The trial had 221 adults with acute bleed, so they divided them into two groups. One was restrictive group and one was the liberal group. So restrictive group, they never transfused them except if the hemoglobin fell below 7 and liberal group, they transfused them if the hemoglobin fell less than 9. And this was the result. Restrictive group received fewer, less units. Mortality was lower in restrictive group and it was less likely to have further bleed. Okay, then patient with cirrhosis, the death and further bleeding were less in restrictive group. Okay, but we have to keep in mind something that all the patients got in the scope within the first five hours, so that was very early in the scope. So theoretically speaking, if we wait like two days, most probably they wouldn't do well. Medication. Asset separation, that's an era of PPI's high dose of anti-security therapy, our advice, which significantly reduced the rate of bleed compared with the standard treatment. Okay, as you know, we give usually pentaprosal 20 milligram bottles, then we will be followed by infusion 8 milligram per hour, and which last for, supposed to last for 72 hours. Then we can't ship them into oral dose, that will be 40 milligram per day or on the prosal 20 milligram twice daily. Prokionetics. They have studied erythromycin and metaclopromycin regarding prokionetic. And the goal of prokionetic is to improve gastric visualization by gastric emptying. Okay, so reasonable dose was 3 milligram per kg, 20 to 30 minutes before the endoscopy. Okay, many studies were done in this regard, and most of them should improve visibility, shorten endoscopy time, reduce the need of second relook, or a second scope. So meta-analysis, which included five trials, 316 patients, suggested that the using of pre-kinetics decreased the need for second relook, but did not affect the number of unit transfusion transfused and length of hospital stay or need of surgery. And the subgroup analysis showed that erythromycin showed a benefit in the regard of second relook endoscopy, but metaclopromycin didn't. Then there was another meta-analysis with examined four trials and had 335 patients, and the meta-analysis found that the patient who received erythromycin were more likely to have an empty stomach during endoscopy. And they had reduction in the time of endoscopy and volume of blood transfusion and the length of hospital stay. I'm not sure about the volume of blood transfusion, because the first meta-analysis didn't show that. Then finally there was strength towards short endoscopy time and decreased mortality. Again, that other meta-analysis didn't show decreased mortality. Then erythromycin was compared to nezogastric lavage. So they divided, it was RTC, which divided into three groups. One was compared erythromycin alone. Then with gastric, sorry compared erythromycin alone with nezogastric lavage alone and nezogastric lavage plus erythromycin. So there were three groups, one erythromycin alone and one was NGT alone and one was both of them together. Quality of visualization did not differ. And there was no difference among all groups in this regard. So it's NG lavage was not recommended. Okay, using somatostatin or ectorotite is not recommended in case of non-very cell bleed. Antibiotics. Antibiotics were studied in that and it showed it has multiple trial wear then and it showed that they are beneficial in patient with cirrhosis because 20% of patients present to hospital with patients of cirrhosis present to hospital with infection and 60% develop infection while being in the hospital. So it was recommended to give cirrhotic patients, they get the antibiotic pre-scope. Okay, antifibrillinotic agent, which is trans-examic acid, meta-analysis was done in this regard also. Seven trials examined and benefit with regard to mortality. It showed benefit in regard to mortality but there was no benefit in regard to bleeding, surgery or, sorry, re-bleeding surgery or transfusion requirement. So it has no role in upper GI bleed management. Okay, upper endoscopy. Upper endoscopy is diagnostic modality of choice. Okay, it has high sensitivity and specificity. Early endoscopy, which is within 24 hours, is recommended. During endoscope you can classify your ulcer according to forest classification, which I'll explain very soon. In patient with blood to obscure the source they may need second endoscopy but second endoscopy is not routinely recommended. This is forest classification. If you see in the left column it is divided into five, sorry, six groups, which is forest 1A, 1B, 2A and so on. So 1A is actively bleeding ulcer. Second 1B is oozing without visible vessel and 2A is non-bleeding visible vessel, then adherent clot, flat spot and clean base ulcer. The prevalence of these ulcers is in second and middle column, which shows that forest 2A, sorry, clean ulcer base, which is forest 3, is the most common followed by forest 2A. And I think last week we were discussing whether what is the risk of reblead in the forest classification. So you see in the right side column that 90% of the patient with the forest 1A tend to reblead and 50% of the patient with forest 2A. But the patient with forest 1B only 10% reblead, though it's in forest 1 and they were oozing initially. So visibility of the vessel seems that it played a role in reblead patient. Joseph Indoscopy, currently most patients are treated with either thermal coagulation or hemostatic clip with or without additional injection therapy. Okay, why they are doing so because a meta-analysis was done and the analysis included 74 trials, which is, I think it's a big number. What did it show? The following were the major conclusions. Compared with epinephrine monotherapy, the risk of further bleed was significantly lower in the other patient which treated with other modalities, like thermal coagulation. Hemoclip were more effective than epinephrine alone. Then the efficacy of endoscopic therapies for adherent growth was uncertain. The choice between hemoclip and thermal therapy was dependent upon the other factors like location of the ulcer and the endoscopist themselves. Okay, the standard approach when you treat patient with endoscopy is thermal coagulation and hemoclip, which also can be combined with injection, sorry. So injection therapy should not be used alone. It should be used in conjunction and should not be ever used alone because the chances of rebleed are higher. Thermal coagulation is contact probe achieve acute hemostasis and prevent recurrent bleed by coactive coagulation. That's compressing the probe against the blood vessel and doing the coagulation so it seals off the vessels and stops the bleed. Other way is organ plasma coagulation, but in this one there is no you don't compress it against the blood vessel, so there is no sealing as in the first one. Hemoclip, endoscopic application of hemoclip is alternate method, which is similar to surgical ligation. If that hemoclip didn't stop the bleeding vessels, it leaves a marker for further things like when you go for interventional radiology or angioembolization. So if you have that marker, the clip will be like a marker for you later on. So it has disadvantage. Alternate approach is these are other things other than the clips is fibrin sealant that you inject fibrin sealant to initiate the hemostasis which prevent recurrent bleed. And there is something called hemostatic nanopowder, which gets adherent to moist surface and stop bleed. I'm not sure about the mechanism exactly, but it forms a barrier to the second relook endoscopy. Second relook endoscopy refers to practice performing a plan endoscopy, generally within 24 hours of initial endoscopy. I'll come to the reason why we do second relook endoscopy, but if you have no reason, usually it's not recommended to go for second endoscopy. So when you are going to go for second relook, if visualization during initial endoscopy was limited due to blood or debris, if there is a concern about part of the endoscopy that the previous endoscopy therapy was self-optimal. Okay. How you manage recurrent or persistent bleed? Persistent bleed refers to the active bleed that does not stop despite endoscopic therapy or bleeding that developed during endoscopic therapy of non-bleeding legion. Precurrent bleed refers to the bleeding that occurs following spontaneous homostasis of the previous successful endoscopic homostasis. Okay. We have options like surgery, which was discussed in previous presentation last week. I just mentioned that type of surgery we can do. So surgical treatment of pyrtic ulcer, bleeding pyrtic ulcer include overseeing the artery with tranquil begatomy, and dractomy with gastrogenostomy, which is billaric too. Highly selected begatomy. This can be done in a stable patient, and you can do it with laparoscopic. In addition to failure of endoscopic therapy, other indication for surgery is hemodynamic in a stable patient despite vigorous resus, like the patient we had last week, when shock associated with recurrent hemorrhage and perforation. Okay. Intervention and geography is less invasive. Okay. And it's equally effective, but it should be considered for patient with high risk. Okay. Best technique of treatment of bleeding into the biliary tree and pancreatic duct. And it's less likely to be successful in patient with impaired coagulation. Follow up patients. Once your patient is ready to discharge, you have to follow him up to treat the basic, the main reason for his ulcer, whether it's a biliary related or NICID or whatever.