Abstract
Scintigraphic studies of gastrointestinal bleeding are usually performed in static acquisition mode and without continuous imaging over a longer period. The aim of this study was to evaluate the diagnostic accuracy of continuous dynamic 99mTc red blood cell scintigraphy (BQS) and cine-mode display in the assessment of unexplained gastrointestinal bleeding. We performed BQS in 40 patients (mean age 57 years) with gastrointestinal bleeding from an undetermined source. All these patients had negative findings of previous examinations. Blood transfusions were required in 16 patients. Continuous dynamic scintigraphic acquisition was performed up to 6 h post injection (p.i.). The scintigraphic data were reviewed without knowledge of the patient's final diagnosis. For this purpose we used a cine-mode display. Dynamic 99mTc red blood cell scintigraphy correctly identified the site of active bleeding in 22 of 23 patients with positive scintigraphic findings. The sensitivity of the scan was 95%, the specificity 94%, the positive and negative predictive values 95% and 94%, respectively. Forty-three percent of the scans became positive within 1.5 h and 82% within 5 h. Positive scans were more likely in patients who required blood transfusions than in patients without transfusion requirement (16 vs 7 patients), and the bleeding site was identified earlier in the former group (mean 2.83 vs 5.28 h p.i.). The scintigraphic identification of the bleeding lesions enabled the performance of a more limited surgical approach in the colon as well as in the distal small intestine, while exact scintigraphic localization was not possible if the bleeding site was located in the proximal small intestine. Continuous dynamic 99mTc red blood cell scintigraphy interpreted by cine-mode display is highly accurate in the identification of a bleeding site. When performed correctly, scintigraphy enables usually segmental resection of these lesions.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have