1Department of Radiology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea. yoonchoi99@yahoo.com 2Department of Surgery, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea. 3Department of Pathology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea.
Published online: March 1, 2009.
ABSTRACT
PURPOSE: To compare the diagnostic efficacy of high-resolution sonography (HRS) and multidetector computed tomography (MDCT) in determining the presence of extraglandular invasion of thyroid papillary cancer and to define ultrasound (US) features of perithyroidal invasion that correlate with histopathological findings. MATERIALS AND METHODS: We prospectively evaluated extracapsular invasion in 177 thyroid cancer patients using both HRS and MDCT. Receiver operating characteristics (ROC) were assessed with a four-point confidence scale (0 = no extracapsular invasion; 1 = possible invasion; 2 = probable invasion; 3 = definite invasion) by two reviewers for each imaging modality. Sensitivity, specificity, and accuracy were analyzed for each modality, along with interobserver variability. RESULTS: MDCT had a mean area under the ROC curve larger than that of HRS (HRS = 0.733, MDCT = 0.807, p < 0.05). HRS and MDCT were significantly different with regard to diagnostic sensitivity, specificity, and accuracy for extrathyroidal extension (p < 0.05; HRS = 75.7%, 66.1%, and 69.8%, respectively; MDCT = 86.7%, 69.7%, and 76%, respectively). Interobserver reliability was greater for MDCT than for HRS (kappa value, 0.861 versus 0.429). The cutoff value used in HRS for estimating the status of perithyroidal invasion was 2. CONCLUSION: HRS may be useful for preoperative investigation of thyroid papillary carcinoma extension, but it was inferior to MDCT because of lower diagnostic accuracy and lower interobserver reliability.
Print
facebook
twitter
Linkedin
google+
Download Citation
PDF Links
Full text via DOI 
