The yield of bleeding scans in acute lower gastrointestinal hemorrhage

Previous studies have demonstrated low yield and conflicting results as to the ability of technetium-99m labeled red blood cell (TRBC) scintigraphy to accurately localize the site of bleeding in acute lower gastrointestinal hemorrhage. Our objective was to evaluate the accuracy of TRBC scintigraphy in clinical day-to-day practice at our institution and to determine predictors of a positive test.

A retrospective medical records review of all patients who underwent TRBC scintigraphy over a 5-year period was conducted at a single academic medical center. The site of bleeding found on TRBC scintigraphy was correlated with the endoscopic, angiographic, surgical, and pathologic findings in each patient.

A total of 127 scans were performed in 115 patients. Forty-nine scans were positive (39%) and 78 were negative (61%). Forty-two patients had further evaluation after a positive scan, and the bleeding site was confirmed to be accurate in 20 of 42 patients (48%). A contradictory bleeding site was found in 5 of 49 positive scans (10%). Patients transfused greater than 2 units of packed red blood cells within 24 hours preceding the TRBC scan were twice as likely to have a positive scan (64%) than those transfused 2 units or less (32%) (P = 0.002). Multivariate analysis demonstrated that the number of units of blood transfused within 24 hours prior to the TRBC scan and the lowest recorded hematocrit were significantly different between patients with positive and negative scans. There was no difference in the rate of endoscopy between patients with a positive or negative scan result.

TRBC scintigraphy has a relatively low yield in the evaluation of acute lower gastrointestinal bleeding. Objective selection criteria such as number of units of packed red blood cells transfused prior to obtaining the scan may increase the overall yield.