Upper Gastrointestinal Bleeding (ugib): Can We Improve Our Management?

Upper Gastrointestinal Bleeding (Ugib): Can We Improve Our Management?

Stephanie Ossowski^*¹; Alexander Bullen, MD²; Dr. Michael Maloy, MD² and Dr. James F Hanley III, M.D.², (1)The University of Texas Health Science Center - San Antonio, San Antonio, TX, (2)University of Texas Health Science Center San Antonio- RAHC, Harlingen, TX

Meeting: Hospital Medicine 2016, March 6-9, San Diego, Calif.

Abstract Number: 254

Category: Research

Sub-Category: Quality Improvement

Keywords: Gastrointestinal Bleeding Management, Quality Improvement and Restrictive Transfusion

Background: UGIB is a common cause of inpatient admissions often associated with morbidity and an overall mortality rate approaching 10%. Since therapy is different, it is important to differentiate between variceal and non-variceal GI bleeding. While both will often need volume resuscitation, proton pump inhibitors (PPI), judicious transfusion and early endoscopy, patients with variceal bleeding will need other intervention to include antibiotics and vasoactive agents. We were interested in whether we followed the guidelines and whether we could make recommendations to improve our care.

Methods: A retrospective analysis was performed on admission and discharge data from January 1, 2013 to January 31, 2014 of patients admitted to our hospital with a diagnosis of UGIB. Patients with a history of malignancy were excluded. Data regarding source of GI bleeding, number of patients transfused, hemoglobin/hematocrit upon presentation, history of previous gastrointestinal bleeding, number of patients who developed complications post-transfusion, number of patients who received PPIs, number of patients who received vasoactive agents, number of patients with hepatic cirrhosis who received antibiotic prophylaxis, and number of patients in which overall management was adequate, was obtained.

Results: A total of 100 patients met criteria. 54% were male, and the average age for men was 67 years old and 69 for women. 12% of the patients had hepatic cirrhosis. 37% of the patients had CAD. 62% of the patients were transfused. 4% of the patients experienced complications to transfusions. 26% had a history of prior GI bleeding. Only one (9.0%) of the patients with hepatic cirrhosis presenting with UGIB received antibiotics. Five (15%) of the patients with CAD who were transfused had an Hb greater or equal to 10 g/dL. Only ten (28.6%) of the patients without CAD were transfused according to the current guidelines.

Conclusions: We found that our management was suboptimal 39% of the time. The principle cause of suboptimal care was overuse of blood transfusion and underuse of antibiotics in cirrhotic patients. Current guidelines emphasize restrictive transfusions, since it leads to reduced rates of further bleeding and need for rescue therapy. Appropriate use of transfusion is also cost-effective, with the cited cost of a platelet depleted red blood cell unit to be as much as $2,400. Among cirrhotic patients, the prophylactic use of antibiotics was low. There have been multiple trials that have shown an overall reduction of infectious complications and decreased mortality (NNT=16) with the use of prophylactic antibiotics. These findings have been reported to the Quality Improvement Committee, with recommendations on the use of transfusions and antibiotics in these patients.