Manage an Upper GI Bleed Like a Boss

Put down that NG tube and pick up some M-M-Maybach Music!

I recently had an acute upper GI bleed case and various questions/arguments arose regarding management. In an effort to keep things evidence-based (what else?) the posting below summarizes some of the most salient points from the International Consensus Upper Gastrointestinal Bleeding Conference Group in 2010. Access the link provides complete information--including guidelines unchanged from the previous conference in 2003 such as abandonment of the NG tube as standard procedure for all UGIB patients.

Italics are cut from the article.

Prognostic scales are recommended for early stratification of patients into low- and high-risk categories for rebleeding and mortality (Agree 97% Grade: low 1c, "do it")

Unchanged from previous (2003), patients presenting to the ED with non-variceal upper gastrointestinal bleeding (UGIB) should be stratified into a risk category based on one of few accepted criteria-based scaling systems. Perhaps the best one--most objective and pre-endoscopic--is the Glasgow Blatchford scale (GBS). There's a convenient plug-n-click version available on mdcalc.com. Essentially, GBS offers almost a binary system where those who earn a total score of 0 are low risk and may be considered for outpatient management, and everyone else (score above 0) warrants a full work-up and some sort of intervention (transfusion, endoscopy, surgery, etc). OK, simple enough. What next?

Blood transfusions should be administered to a patient with a hemoglobin level of 70 g/L or less (Agree 100% Grade: low, 1c "do it")

The unanimity of this entry comes at somewhat of a surprise, considering that the data sourced for this decision comes from anesthesiology, cardiothoracic, and ICU studies of non-UGIB patients. Certainly what is well documented and shared opinion is that essentially all patients with a Hb of 6 or lower need blood, whereas those with 10 or greater almost never do. If your patient displays signs of end organ or tissue dysfunction (troponins, for example) the recommendation is to transfuse regardless of initial CBC Hb levels (bleeding is dynamic and lab values may take time to catch up to current states). Given that UGIB patients are critical, we can draw a parallel to the Surviving Sepsis Campaign: transfuse critical (septic) patients to a target level higher than 7 to 9 (greater than 10 showed no benefit). What if your patient's level is 8 or 9? A trial in Transfusion in 1999 showed no difference in M&M for post-CABG patients being transfused at levels less than 9 and 8--and these are patients with coronary artery disease!
Final word: give PRBCs to those with Hb less than 7 and/or any level with signs and symptoms of end-organ dysfunction.

In patients receiving anticoagulants, correction of coagulopathy is recommended but should not delay endoscopy (Agree 97%, Grade: low, 2c "probably do it")

First, this is in regards to those patients receiving coumadin and such at home, not those with documented liver disease coagulopathy. The data for this proposal seems scanty and underwhelming. In short, for patients with INR greater than 2.5 it may be feasible to administer FFP to facilitate endoscopy and reduce a (not well supported) rebleed risk. This however should NOT in anyway delay endoscopy, especially with an INR less than 2.5. In such patients, ship them to endscopy as soon as possible after consulting with your gastroenterologist. FFPs take time to arrive and administer and time is of the essence.

Pre-endoscopic PPI therapy may be considered to downstage the endoscopic lesion and decerase the need for endoscopic intervention but should not delay endoscopy (Agree 94%, Grade moderate 1b "do it")

Well, do it if you can. PPIs in UGIB have never been shown to benefit the patient in terms of mortality, rebleeding, or need for surgery. The weak plea for its use stems namely from the Winter's criteria, an endoscopy-based criteria used when looking at a lesion that maintains a poor kappa score. Dr Newman and Dr Shreeves summarize this literature quite thoroughly and convincingly. My current approach is similar to theirs: give 40 mg (IV or PO--equally effective) to appease the endoscopist, but keep all lines open for fluids, blood, pressors, and any other therapy aside from a protonix drip. It doesn't benefit anyone, is costly and resource draining in the ED. Oh yeah, and hold off on the "high dose" of 80 as well, it's never been shown to have benefit either.

Now let's pull out the two-seater, baby.