Detection of upper tract urothelial neoplasms: sensitivity of axial, coronal reformatted, and curved-planar reformatted image-types utilizing 16-row multi-detector CT urography Jonathan R. Dillman, Elaine M. Caoili, Richard H. Cohan, James H. Ellis, Isaac R. Francis, Matthew J. Schipper Department of Radiology, University of Michigan Health System, 1500 East Medical Center Drive, Ann Arbor, MI 48109-5030, USA Abstract Background: The purpose of our study was to determine the sensitivity of 16-row multi-detector computed tomog- raphy urography (CTU) axial, coronal reformatted, and curved-planar reformatted image-types for upper tract urothelial neoplasm detection. Methods: Twenty-one CTU examinations were identified that contained one or more pathology-proven upper tract urothelial neoplasms during our study period. Two readers independently reviewed 1.25 and 2.5 mm axial, coronal reformatted, and curved-planar reformatted excretory phase CTU images. Each reader then docu- mented the location and appearance of lesions suspicious for upper tract urothelial neoplasm. These results were correlated with relevant endoscopic/surgical procedure notes as well as pathology results. Results: Reader #1 detected 72%, 63%, 75%, and 72% of the known 32 upper tract urothelial neoplasms using 1.25 mm axial, 2.5 mm axial, coronal reformatted, and curved-planar reformatted image-types, respectively. Reader #2 detected 72% of the known neoplasms using each of the four image-types. No significant difference in sensitivity between image-types was identified. After combining the results for all four image-types, readers #1 and #2 had overall sensitivities of 94% and 91%. Conclusions: Axial, coronal reformatted, and curved- planar reformatted image-types have similar sensitivities for the detection of upper tract urothelial neoplasm in 16-row multi-detector CTU. Reviewing multiple image- types increases the sensitivity of urothelial lesion detec- tion. Key words: Computed tomography urography (CTU)—Multi-planar reformatted imaging—Urinary tract—Urothelial neoplasm—Sensitivity In patients with a history of unexplained hematuria or urinary tract neoplasm, evaluation of the upper urinary tract (calices, pelvises, and ureters) urothelium is essen- tial. Traditional radiologic evaluation of these structures has utilized excretory urography (EU) [1–3]. Several re- cent studies, however, have established multi-detector row computed tomography urography (CTU) as a useful imaging modality in the evaluation of the upper urinary tract for neoplasm [4–7]. Most recent investigations employing multi-detector row CTU in the evaluation of upper urinary tract neo- plasms have utilized reconstructed axial images (with a slice thickness ranging from 1.25 to 5 mm) in conjunction with a variety of reformatted image-types, including maximum- intensity projection (MIP), average-intensity projection (AIP), coronal reformatted, and volume-rendered (VR) reconstructions [5–7]. Little information is available con- cerning the individual sensitivities of axial, coronal refor- matted, and curved-planar reformatted image-types in the detection of upper tract urothelial neoplasms. The purpose of this study was to determine retro- spectively the sensitivity of each of the aforementioned image-types in the detection of upper tract urothelial Correspondence to: Jonathan R. Dillman; email: jonadill@med.umich. edu ª Springer Science+Business Media, LLC 2008 Published online: 6 February 2008 Abdominal Imaging Abdom Imaging (2008) 33:707–716 DOI: 10.1007/s00261-008-9360-3