Management and Dispo of Upper GI Bleed
- Sep 2nd, 2014
- Chris Chase
Upper GI bleed (UGIB) is a common cause of presentation in the ED with an incidence of 102 per 100,000 hospitalizations. Mortality can range from 2-15%, with a rebleeding risk of up to 10-30%. It is important to identify high-risk patients that are candidates for early endoscopy or if monitored management is more appropriate. Low-risk patients may be identified and safely discharged from the ED.
It is first important to identify if the source of bleeding is upper or lower. UGIBs are more common and also more dangerous. Predictors of UGIB are hematemesis, history of melena, melenic stool on rectal exam, blood or coffee grounds on NG, and BUN/Cr ratio usually >20. The most common causes of UGI bleeds are PUD (55% with a 4% mortality) and esophageal varices (14% with a 50% mortality).
Management is multi-focal and varies depending on if ongoing bleeding is present or not. NG lavage has not shown to improve outcome, with no differences in mortality, LOS, surgery, or transfusion need. A negative lavage does not exclude UGI bleed. Benefits to lavage include shorter time to endoscopy, determine ongoing bleeding and can remove clot. In addition to general supportive measures, medications that are helpful include octreotide, PPIs, prokinetics, and prophylactic antibiotics. Octreotide reduces risk of persistent and rebleeding but showed no mortality difference in systematic review. Prokinetics improve endoscopic view by gastric emptying and reduces need for repeat endoscopy and LOS (erythromycin shown to work as well as NG lavage). Rocephin has a mortality benefit in cirrhotic patients (50% develop infection). Blood transfusions for Hgb <7 is associated with 45% mortality reduction. Vasopressin and balloon tamponade are potential options if other methods are exhausted, but are associated with greater complications.
Several scoring systems have been developed to identify high vs low-risk patients. The Glasgow-Blatchford Score (GBS) addresses whether or not a patient will need an urgent endoscopy by scoring clinical factors. These factors are pulse, systolic BP, BUN, hemoglobin, presence of melena or syncope, and presence of hepatic disease or CHF. A score of zero was associated with a very low likelihood of need for emergent endoscopy. A modified GBS, omitting subjective criteria (hepatic/cardiac disease, melena, and syncope), has performed as well as the full GBS in predicting clinical outcomes and may be easier to use in clinical practice. There are 5 risk factors that are associated with inpatient mortality: albumin <3.0 g/dl, INR>1.5, AMS, systolic BP <90, and age>65. Mortality risk increases, as expected, with more risk factors present (0 RF-0.3%- 5 RF-25% mortality).
- Cheng, D. W., Y. W. Lu, T. Teller, H. K. Sekhon, and B. U. Wu. “A Modified Glasgow Blatchford Score Improves Risk Stratification in Upper Gastrointestinal Bleed: A Prospective Comparison of Scoring Systems.” Ailment Pharmacol There 36.8 (2012): 782-89. Web. 4 Aug. 2014.
- Saltzman, John R. “Approach to Acute Upper Gastrointestinal Bleeding in Adults.” Uptodate.com. N.p., 16 May 2014. Web. 4 Aug. 2014.
- Vissers, Robert J. “Evaluating the Patient with GI Bleeding.” Advanced Practice Provider Academy. San Diego, CA. 16 Apr. 2014. Acep.org. Web. 4 Aug. 2014.