RESUMEN
OBJECTIVE: 99mTc-labelled white-cell scintigraphy is sensitive and specific for detecting inflammatory bowel disease (IBD). A retrospective analysis was carried out on 99 consecutive studies to compare results from planar imaging and single photon emission computed tomography (SPECT). MATERIALS AND METHODS: Two observers blindly and independently reported all planar images followed by the SPECTs. Both sets of scans were reported in random order. All possible abnormalities were assigned a confidence rating and specific, preagreed bowel location and their extent measured. RESULTS: The observers disagreed on whether studies were normal in 17 of 99 planar scans (κ=0.56) but only eight of 99 SPECTs (κ=0.83). Consensus reporting, where there was disagreement between observers, yielded the following results: planar and SPECT normal in 56 of 99 cases; planar and SPECT positive in 29 of 99 cases; planar normal but SPECT positive in 13 of 99 cases; planar positive but SPECT normal in one of 99 cases. In the 14 of 99 cases where there was discordance between planar and SPECT, confirmation of active IBD was sought from other investigations (e.g. histology). Relevant other investigations were available in eight of these discordant cases and showed agreement with the SPECT result in seven of eight patients. Where the same abnormalities were identified in both planar and SPECT, there was a significant increase (P<0.001 mean: 57 mm) in the extent of abnormalities as measured from SPECT compared with planar. CONCLUSION: Reporting from SPECT improves inter-operator variability and appears to improve sensitivity for detecting IBD. Planar imaging also appears to underestimate the extent of active disease.