Open Access
Digestive Disease Interventions 2017; 01(S 01): S111-S142
DOI: 10.1055/s-0037-1603722
Oral Presentations
Thieme Medical Publishers 333 Seventh Avenue, New York, NY 10001, USA

Predictors for Positive CT Angiography in Lower Gastrointestinal Bleeding

Talal Akhter
1   Department of Radiology, Temple University Hospital, Philadelphia, Pennsylvania
,
Victor Rivera
1   Department of Radiology, Temple University Hospital, Philadelphia, Pennsylvania
,
Kyle Panzner
1   Department of Radiology, Temple University Hospital, Philadelphia, Pennsylvania
,
Joseph Panaro
1   Department of Radiology, Temple University Hospital, Philadelphia, Pennsylvania
,
David Pryluck
1   Department of Radiology, Temple University Hospital, Philadelphia, Pennsylvania
,
Emily Cuthbertson
1   Department of Radiology, Temple University Hospital, Philadelphia, Pennsylvania
,
Dmitry Niman
1   Department of Radiology, Temple University Hospital, Philadelphia, Pennsylvania
,
Gary Cohen
1   Department of Radiology, Temple University Hospital, Philadelphia, Pennsylvania
,
Mark Burshsteyn
1   Department of Radiology, Temple University Hospital, Philadelphia, Pennsylvania
› Author Affiliations
Further Information

Publication History

Publication Date:
24 May 2017 (online)

 
 

    Diagnostic imaging options for localization of lower gastrointestinal bleeding (LGIB) prior to conventional catheter directed angiography have historically included a Tc99m tagged red blood cell scan, and more recently, computed tomography angiography (CTA). It has been well documented that CTA sensitivity in identifying LGIB is inferior to nuclear scintigraphy (0.1 ml/min for scintigraphy versus 0.35 ml/min for CTA). As a result, CTA can frequently be negative in patients who are clinically found to have a GI bleed.

    We performed a retrospective, IRB approved analysis of all CTAs performed for LGIB at Temple University Hospital from March 2014 through May 2016 to analyze parameters that may help to predict active extravasation on CTA. A total of 71 patients were analyzed. 34 patients did not demonstrate active bleeding on CTA, and were designated as the negative group. 37 patients demonstrated active bleeding on CTA, and were designated as the positive group.

    Patients in the positive group were found to have more likely received vasopressors (p = 0.01) and pRBCs (p = 0.08) prior to CTA. In addition, a trend was observed, showing higher average hematocrit (HCT) in the positive CTA group prior to CTA (27.86 vs. 25.45, p = 0.1). Therefore, patients who have not received transfusions or vasopressors may benefit from nuclear scintigraphy rather than CTA as the initial diagnostic study. Stratification of patients into subgroups based on these predictors may reduce unnecessary imaging, improve efficiency, decrease costs, and most importantly, reduce morbidity, specifically in patients with underlying renal insufficiency.


    No conflict of interest has been declared by the author(s).