|
Back to 2014 Annual Meeting Posters
Repeat Endoscopy and Angiography for Persistent Lower Gastrointestinal Bleeding: Is There an Upper Limit?
Allan Mabardy*, Lerna C. Ozcan, Joseph J. Coury, Akeel Merchant, Julia Tassinari, Alan W. Hackford, Haisar E. Dao
Saint Elizabeth's Medical Center, Boston, MA
Introduction: Patients presenting with lower gastrointestinal bleeding may undergo multiple endoscopic and angiographic procedures. Identification of the source of bleeding may be elusive, and the end result may be a total colectomy. For those patients who have many nonoperative interventional procedures, we sought to investigate any differences that might exist with regards to overall mortality, the rate of operative intervention, and the type of operative procedure.
Methods: All patients were identified nationwide who were admitted to the hospital between 2006 and 2010 with lower gastrointestinal bleeding using the Nationwide Inpatient Sample (NIS) database. Patients not requiring blood transfusions were excluded. Patients were grouped according to the number of colonoscopies, flexible sigmoidoscopies, and angiographies that were performed. We then compared the rate of bowel resection, the type of procedure, and the overall mortality for patients who underwent more than 2 nonoperative interventional procedures with those patients who had 2 or fewer procedures.
Results: From 2006 through 2010, 289,501 patients were admitted with a diagnosis of lower gastrointestinal bleeding requiring blood transfusion. Of these patients, 0.33% (n = 954) underwent more than 2 nonoperative interventional procedures. The rate of large bowel resection was significantly higher for this group of patients (15.0% vs. 3.5%, p<0.0005). There was no difference in mortality (1.7% vs. 1.7%, p=0.962). For those patients who underwent large bowel resection, patients who had undergoing more than 2 nonoperative interventional procedures were significantly more likely to undergo total colectomy (20.3% vs. 8.0%, p<0.0005).
Conclusions: Patient who undergo more than 2 nonoperative interventional procedures are more likely to require a bowel resection and more likely to undergo a total colectomy. Overall mortality for this group, however, is not higher. These patients may have persistent bleeding that cannot be localized, yet a stable enough clinical picture to permit repeating endoscopy or angiography and delaying resection. There is no evidence from our study to suggest that such patients should undergo resection sooner, nor is there evidence that multiple endoscopies and angiographies adversely affect survival.
Back to 2014 Annual Meeting Posters
|
