[The Application of Neuroimaging Reconstruction Technique in Evaluating the Relationship Between Skeletal Musculature Tumors and Nerves Nearby].
Z Lü,J Zang,B Zhang,T J Chang,F Yang,M L Na
DOI: https://doi.org/10.3760/cma.j.issn.0376-2491.2020.05.013
2020-01-01
Abstract:Objective: To analyze the relationship between tumors of the musculoskeletal system and adjacent nerves by reconstructing images of peripheral nerves, and explore its value in surgical treatment. Methods: From May 2016 to April 2019, a total of 27 patients were collected in Department of Imaging,Shougang Hospital, who were with skeletal muscle system tumors, including 15 primary soft tissue tumors, 9 primary bone tumors, 3 metastatic tumors, all of them were closely related to nerves. There were 17 males and 10 females, aged 13-67 years, with an average age of 34 years. Before the operation, CT volume scanning was performed, and curved planar reconstruction (CPR) was used to reconstruct the peripheral nerves. All patients were operated within 2 weeks after the examination. According to the image characteristics before the operation, the nerve invasion was judged. The sensitivity, characteristic, positive predictive value and negative predictive value of the tumor invasion (compression) nerve were calculated according to the intraoperative findings as the gold standard. Result: Of the 27 cases, 25 cases (25/27, 92.6%) could show the relationship between the tumors and the adjacent nerves at the same level, and 22 cases (22/25, 88.0%) had the same preoperative image judgments as the intraoperative findings. In the reconstructed images, the peripheral nerve was a continuous strip-like structure on the same level with the tumor. The invaded nerve became thicker and the edge was blurred. Enhanced scan showed enhancement. The sensitivity, specificity, positive predictive value and negative predictive value of neuroimaging reconstruction were 100.0%, 89.5%, 75.0% and 100.0% respectively. The sensitivity, specificity, positive predictive value and negative predictive value of the nerve compression were 92.3%, 100.0%, 100.0% and 80.0% respectively. Conclusions: Neurological reconstructed images can help clinicians evaluate the relationship between lesions and adjacent nerves quickly and intuitively. They can guide the selection of surgical methods, reduce the risk of intraoperative nerve injury, and have high sensitivity and specificity for nerves invasion.