GI Emergencies: Massive Upper GI Bleeding Resuscitation

0:00Welcome to Board Pearls. This is episode one of four of the GI Emergencies chapter, in the Special Populations and Acute or Supportive Care module. In this episode we cover massive upper GI bleeding as a resuscitation problem: restrictive transfusion, massive transfusion protocol, anticoagulation reversal in the first hour, pre-endoscopic erythromycin, and the recognition traps of aortoenteric fistula and failed hemostasis.

0:27Massive upper GI bleeding runs on a clock. The clock starts when the patient arrives. The decisions made in the first hour shape the next twenty-four, and the wrong decision applied confidently is worse than no decision at all. The teaching for this episode follows the timeline the patient is actually moving through.

0:45Massive upper GI bleeding is a resuscitation problem before it is an endoscopic problem. The Forrest stigmata, the PPI strategy, and the variceal bundle all live in the upper GI bleeding chapter. What this episode adds is the resuscitation that precedes those decisions, and the parts of the first hour that the boards test as their own answers.

1:05The transfusion target is restrictive. The landmark trial randomized patients with acute upper GI bleeding to a restrictive strategy, a hemoglobin trigger of seven and a target of seven to nine, against a liberal strategy triggering at nine. Restrictive won, with lower mortality and less rebleeding, and the cirrhosis subgroup drove most of the benefit.

1:25That last point is the mechanism. Liberal transfusion in a cirrhotic with portal hypertension reopens partially clotted vessels and expands portal volume, which raises variceal pressure and starts the bleed again. Overzealous resuscitation in any bleeding patient raises mean arterial pressure across a fresh clot and dilutes the platelets and clotting factors the patient is making at the bleeding site. Restrictive works because permissive hypotension protects the clot.

1:54Two qualifications hold. The patient with active acute coronary syndrome is transfused at eight rather than seven, because the same anemia that protects clots starves an ischemic myocardium. And the patient with active uncontrolled hemorrhage in front of you is transfused empirically based on clinical assessment, because the laboratory hemoglobin lags the real-time loss by thirty to sixty minutes. The number you draw is from before the most recent surge.

2:21The second early decision is massive transfusion protocol activation. The threshold is anticipated need over ten units of packed red cells in twenty-four hours, or four units in one hour. The protocol delivers a balanced one-to-one-to-one ratio of packed red cells, fresh frozen plasma, and platelets. That ratio reproduces whole blood and avoids the dilutional coagulopathy that crystalloid-only or red-cell-only resuscitation produces. When a patient loses a whole blood volume, receiving red cells without plasma and platelets dilutes the clotting machinery and prolongs the bleed.

2:55Calcium repletion is the under-tested adjunct. Citrate in stored blood products chelates ionized calcium, and ionized calcium is required for coagulation cascade activation. Rapid transfusion drops ionized calcium fast enough that repletion belongs in the protocol itself rather than as an afterthought.

3:16Anticoagulation reversal runs in parallel with the transfusion. The full agent-specific algorithm lives in the endoscopy practice chapter, but the short version belongs here because it is part of the first hour. Warfarin reversal in life-threatening bleeding is four-factor prothrombin complex concentrate, not fresh frozen plasma, because PCC reverses faster and carries less volume burden in a patient already being volume-resuscitated. Vitamin K is added for sustained reversal but is not the acute-window answer because of its twelve to twenty-four hour onset. Dabigatran is reversed with idarucizumab five grams intravenously. Apixaban and rivaroxaban are reversed with andexanet alfa, with four-factor PCC as the alternative when andexanet is unavailable.

3:59Reversal pairs with intervention rather than delaying it. Endoscopic hemostasis can proceed with INR between one point five and two point five alongside reversal. INR above two point five gets reversed before endoscopy in most situations. The principle is that you do not let the INR hold up the EGD when the patient is exsanguinating.

4:23Pre-endoscopic preparation has two specific pieces beyond IV PPI. Erythromycin two hundred fifty milligrams intravenously over twenty to thirty minutes, given thirty to ninety minutes before endoscopy, improves visualization by promoting gastric emptying through motilin receptor agonism. Blood and clot clear from the gastric lumen so the endoscopist can find the source. The benefit is reduced need for second-look endoscopy and shorter length of stay, not mortality.

4:52Tranexamic acid is not given. A large trial showed no benefit and an increased risk of venous thromboembolism. The reason TXA fails in upper GI hemorrhage is that this is arterial spurting from a discrete vessel that requires mechanical hemostasis. It is not the fibrinolysis-driven coagulopathy that antifibrinolytic therapy is designed to address. TXA is the right tool for the wrong problem.

5:15Airway protection with endotracheal intubation is the answer in several scenarios. The encephalopathic cirrhotic with active hematemesis. Massive hematemesis with aspiration risk. Hemodynamic collapse. Any patient who cannot protect the airway during expected high-volume blood return at endoscopy. The board-favored move is elective intubation before EGD rather than waiting for an emergent airway in a positioned patient with blood in the field.

5:42Endoscopy timing for non-variceal upper GI bleeding is within twenty-four hours of presentation. A randomized trial in patients with high-risk bleeding scores compared urgent endoscopy within six hours against early endoscopy at six to twenty-four hours and found no significant difference in thirty-day mortality, rebleeding, or need for intervention. The instinct under stem pressure is to read sicker as more urgent, but pushing the EGD earlier did not help.

6:11Variceal hemorrhage gets endoscopy within twelve hours, and the full variceal bundle lives in the upper GI bleeding chapter. Pushing endoscopy below six hours is reserved for ongoing massive hemorrhage that cannot be stabilized, or for suspected aortoenteric fistula in a patient with a prior aortic graft.

6:31Aortoenteric fistula is the high-yield trap inside the massive UGIB recognition stem. The classic presentation is a herald bleed, a self-limited episode days to weeks before the catastrophic hemorrhage, in a patient with a prior aortic graft. The fistula classically communicates with the third part of the duodenum. CT angiography looking for periaortic gas or contrast extravasation is the imaging step that precedes both EGD and operative repair, because confirming the diagnosis changes the operation.

7:02Salvage angiography by interventional radiology is the threshold when endoscopic hemostasis fails despite appropriate dual therapy. The two arteries that get embolized most successfully are the gastroduodenal artery for posterior duodenal bulb ulcers and the left gastric artery for lesser-curve gastric ulcers. Those are also the two locations where endoscopic hemostasis fails most often, with rebleeding rates near half for posterior bulb ulcer and roughly a third for lesser-curve gastric ulcer. Both arteries course immediately deep to a relatively thin ulcer floor, which is why endoscopy struggles and why angiographic embolization works.

7:42So the upper GI hemorrhage bundle has a single shape. Resuscitate to a restrictive target rather than a liberal one, because permissive hypotension protects the clot. Activate the massive transfusion protocol at the balanced ratio and remember the ionized calcium that citrate chelates. Reverse anticoagulation in parallel with endoscopy rather than as a delay, with prothrombin complex concentrate for warfarin and the agent-specific antidotes for the direct oral anticoagulants. Clear the field with erythromycin so you can see the source, protect the airway when blood return threatens it, and recognize the herald bleed of an aortoenteric fistula and the two arteries where failed hemostasis sends you to interventional radiology.

8:25The next episode takes the lower GI and ischemic emergencies: acute lower GI bleeding with its hemodynamic split between CT angiography and colonoscopy, colon ischemia with pain before bleeding, and acute mesenteric ischemia where lactate lags and the four etiologies each demand their own intervention.

8:45For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode one of four of chapter thirty one, and I'll see you in the next one.