Background: Acute lower gastrointestinal bleeding (LGIB) is a major cause of morbidity and mortality. Multidetector row computed tomography (CT) with gastrointestinal (GI) bleed protocol is a novel diagnostic technique for detecting and localizing LGIB. Being rapid and noninvasive, it may be useful as a first-line modality to investigate cases of acute LGIB. Goals: To assess and compare diagnostic accuracy of 99mTechnetium (Tc)-labeled red blood cell (RBC) scintigraphy and multidetector row CT with GI bleed protocol for detection and localization of source of acute LGIB. Study: Requirement of informed consent was waived for this retrospective study. Seventy-six patients had undergone either RBC scintigraphy, CT with GI bleed protocol, or both, followed by conventional angiography for evaluation of acute persistent LGIB between January 2010 and February 2014 at our institution. Accuracy of both modalities was assessed using conventional angiography as reference standard and compared using the 2-tailed, Fisher exact test. A P-value of <0.05 was considered statistically significant. Results: Fifty-one, 20, and 5 patients had undergone RBC scintigraphy only, CT with GI bleed protocol only, and both modalities, respectively. Fourteen of 25 patients in the CT group had angiographic evidence of active bleeding as compared with 32 of 56 patients in the scintigraphy group. CT with GI bleed protocol had higher accuracy (96%) than 99mTc-labeled RBC scintigraphy (55.4%, P < 0.001). Conclusions: CT with GI bleed protocol was more accurate in detecting and localizing the source of acute LGIB as compared with 99mTc-labeled RBC scintigraphy.
- Digital subtraction angiography
- Gastrointestinal hemorrhage
- Multidetector computed tomography
- Radionuclide imaging