Value of the Cinematic Rendering From Volumetric Computed Tomography Data in Evaluating the Relationship Between Deep Soft Tissue Sarcomas of the Extremities and Adjacent Major Vessels: A Preliminary Study

• Published in: Journal of computer assisted tomography Vol. 43; no. 3; p. 386
• Main Authors: Li, Kun, Yan, Ruiying, Ma, Huan, Zhang, Da-Fu, Ding, Yingying, Li, Zhen-Hui
• Format: Journal Article
• Language: English
• Published: United States 01.05.2019
• Subjects: Adult; Aged; Cone-Beam Computed Tomography - methods; Extremities - diagnostic imaging; Extremities - pathology; Female; Humans; Imaging, Three-Dimensional - methods; Male; Middle Aged; Retrospective Studies; Sarcoma - diagnostic imaging; Sarcoma - pathology; Sensitivity and Specificity; Young Adult
• ISSN: 1532-3145
• DOI: 10.1097/RCT.0000000000000852

The aim of the study was to assess the value of cinematic rendering (CR) from volumetric computed tomography data in evaluating the relationship between deep soft tissue sarcomas (STSs) of the extremities and the adjacent major vessels. Preoperative contrast-enhanced axial imaging (CEAI) in the arterial phase with three-dimensional volume rendering (VR) and CR of contrast-enhanced computed tomography were used to assess adjacent vascular invasion in 43 cases of deep STSs of the extremities. The imaging assessments were compared with surgical findings and interpreted as negative (no vascular invasion) or positive (vascular invasion was present). Intrareader and interreader agreement were assessed using Cohen κ statistics. The diagnostic performance of CEAI, VR, and CR was evaluated by receiver operating curve analysis and compared using the DeLong test. Thirty-four and nine cases were classified as negative and positive, respectively, in surgery. Intrareader agreement values for the CEAI, VR, and CR assessments were all excellent (0.984, 0.934, and 0.914, respectively), whereas the interreader agreement for CEAI assessments was greater than that for VR and CR (0.969 vs 0.804 and 0.761). Cinematic rendering showed lower accuracy (0.698), sensitivity (0.778), specificity (0.676), positive predictive values (0.389), and negative predictive values (0.920) for vascular invasion diagnosis than CEAI or VR; the accuracy, sensitivity, specificity, positive predictive values, and negative predictive values increased to 0.767, 0.889, 0.735, 0.471, and 0.962 for both CEAI and VR. The results were not statistically significant (all P > 0.05). Cinematic rendering has the potential to be used to evaluate vascular invasion in cases of deep STSs of the extremities, but it should be used alongside the traditional methods such as CEAI.