Utility of 70-kV single-energy CT in depicting the extent of breast cancer for preoperative planning.

PURPOSE: To evaluate the detectability of breast cancer and visibility of the tumor extent using 70-kV single-energy contrast-enhanced (CE) breast computed tomography (70-kV CECT) compared with CE breast magnetic resonance imaging (CEMR). METHODS: Between 2013 and 2015, 110 patients with 112 breast cancer lesions who underwent breast surgery after undergoing both 70-kV CECT and CEMR were enrolled. The major axis lengths of the breast lesion were measured and compared with the pathologically determined major axes. Agreement in the measured major axes was evaluated using the intra-class correlation coefficient (ICC). RESULTS: Both 70-kV CECT and CEMR depicted all breast cancer lesions. The mean major axis was 3.0 (95% confidence interval [CI], 2.5-3.4) cm on CECT and 2.9 (2.6-3.3) cm on CEMR. The mean differences between the pathologically and radiologically measured major axes on 70-kV CECT and CEMR were 0.9 (0.7-1.1) and 1.0 (0.8-1.2) cm, respectively. The accuracy of the radiological major axes compared with the pathological major axes was 82.1% and 80.4% on CECT and CEMR, respectively (p = 0.81). The major axes on the two modalities demonstrated moderate agreement (ICC = 0.69, 95% CI 0.58-0.77). Pathologically and radiologically measured major axes on 70-kV CECT and CEMR demonstrated excellent agreement (ICC = 0.91, 95% CI 0.93-0.96). CONCLUSIONS: Low-tube voltage (70-kV) CECT is the preferred modality to identify breast cancer lesions and tumor extent for preoperative planning because it has a similar diagnostic ability to CEMR and can be performed in the supine position.

Useful computed tomography features for differentiating between focal atelectasis and pleural dissemination on preoperative evaluations of thymic epithelial tumors.

PURPOSE: Distinguishing between focal atelectasis (FA) and pleural dissemination (PD) is important for determining the optimal therapeutic strategy for thymic epithelial tumors (TET). This study aimed to identify useful computed tomography (CT) features for distinguishing between these two conditions. MATERIALS AND METHODS: We retrospectively analyzed preoperative CT images of 27 TET, which included 40 PD and 40 FA lesions. Two radiologists independently interpreted the pleural lesions without knowing the final diagnosis. The CT images were evaluated to assess the lesion location, size, and shape, presence of a spinous shadow and ground glass opacities (GGO) near to the lesion, and the shortest distance from the lesion to the nearest peripheral pulmonary artery (PA). RESULTS: FA lesions tended to occur on the dorsal side (90%, P = 0.024); have shorter major and minor axes (P < 0.001), a triangular shape (43%, P = 0.002), a spinous shadow (45%, P = 0.001) and GGO (28%, P = 0.006); and be close to a peripheral PA (P = 0.007). Ninety percent of PD lesions were located in the left thorax, and all of them were ipsilateral to the tumor (both P < 0.001). The 9 examined factors exhibited sensitivity, specificity, positive predictive, and negative predictive values of 85%, 95%, 94%, and 86%, respectively, for diagnosing FA (when ≥3 factors were present), and 90%, 48%, 63%, and 83%, respectively, for diagnosing PD (when ≥4 factors were present). CONCLUSION: The site, size, and shape of a lesion; the presence of a spinous shadow/GGO; and the distance to the nearest PA are useful for distinguishing between PD and FA.

Utility of the variable flip angle 3D fast-spin echo (isoFSE) sequence on 3T MR for diagnosing vertebrobasilar artery dissection.

We aimed to investigate the utility of the isoFSE sequence, one of the variable flip angle 3D fast-spin echo sequences, on 3T-MR for displaying vessel walls and diagnosing vertebrobasilar artery dissection (VAD). We retrospectively evaluated 12 initial and 28 follow-up images from 12 patients diagnosed with either intracranial VAD or carotid artery dissection. The image quality for displaying the vessel wall was scored using a five-point scale (1 poor, 5 excellent) on initial T1-weighted isoFSE images for each region of the arteries. The intracranial artery dissection findings assessed at time points after onset were evaluated on initial and follow-up T1/T2-weighted isoFSE images. For small arteries, including the anterior/posterior inferior cerebellar artery, similar high scores were obtained on both unenhanced and contrast-enhanced T1-weighted isoFSE images (average: 4.7-5.0, p > 0.2). On unenhanced images, dissected vertebral arteries showed significantly lower scores than non-dissected vertebral arteries for both readers (p = 0.017 and 0.015, respectively), but the scores were high (3.9 and 4.0, respectively). Definitive findings of VAD were observed on the initial images except in one case. For all cases, definitive findings were seen on at least one of the initial or follow-up images. Temporal changes in the findings could be observed for all cases. In conclusion, we showed favorable wall visualization on T1-weighted isoFSE images and the utility of follow-up imaging using unenhanced-T1/T2-weighted and contrast-enhanced T1-weighted isoFSE sequences with acceptable scan times, which could promote the regular use of 3D black-blood vessel wall imaging.

Usefulness of 3D balanced turbo-field-echo MR sequence evaluating the branching pattern of the intrahepatic bile ducts: comparison with drip infusion CT cholangiography.

PURPOSE: For liver surgery, it is crucial to preoperatively examine the course of the right posterior bile duct. While MR cholangiopancreatography (MRCP) can only visualize the bile ducts, 3D balanced turbo-field-echo (BTFE) sequence clearly depicts the bile ducts and portal veins as well as drip infusion CT cholangiography (DIC-CT), without contrast media. We evaluated whether BTFE could substitute for DIC-CT. MATERIALS AND METHODS: Thirty patients undergoing MRCP and BTFE on 1.5-T MR and DIC-CT were evaluated. Two readers retrospectively evaluated the branching pattern (supra-type: A-C or infra-type: D-E) and scored the degree of confidence and motion artifacts using a 3-point scale for the three 2-mm-thick reconstructed images. RESULTS: The bile duct diameter did not differ between DIC-CT and MRCP (p = 0.07). Five patients (17%) had intrahepatic biliary dilatation (>3 mm). The A, B, C, D, and E types were diagnosed in 21, 6, 1, 1, and 1 patient, respectively (28 supra-types and 2 infra-types) on DIC-CT. For DIC-CT, MRCP, and BTFE, the mean motion artifact scores were 3.0/3.0, 2.7/2.6, and 2.9/2.8, respectively. The mean diagnostic confidence scores were 2.9/2.9, 2.4/2.4, and 2.9/2.8, respectively, with no difference between DIC-CT and BTFE. The concordance between DIC-CT and BTFE was high (infra- or supra-type: κ = 1.00/1.00, A-E: κ = 0.86/0.66), but it was poor between DIC-CT and MRCP (infra- or supra-type: κ = 0.35/-0.05, A-E: κ = 0.33/0.41) for both readers. CONCLUSIONS: Similar to DIC-CT, the BTFE MR sequence had high diagnostic accuracy regarding the branching pattern of the intrahepatic bile duct, especially for the supra/infraportal type.