Background: Sonoelastography is an ultrasound imaging technique able to assess mechanical properties of tissues. Strain elastography (SE) is a qualitative sonoelastographic modality with a wide range of clinical applications, but its use in brain tumor surgery has been so far very limited.
Objective: To describe the first large-scale implementation of SE in oncological neurosurgery for lesions discrimination and characterization.
Methods: We analyzed retrospective data from 64 patients aiming at (i) evaluating the stiffness of the lesion and of the surrounding brain, (ii) assessing the correspondence between B-mode and SE, and (iii) performing subgroup analysis for gliomas characterization.
Results: (i) In all cases, we visualized the lesion and the surrounding brain with SE, permitting a qualitative stiffness assessment. (ii) In 90% of cases, lesion representations in B-mode and SE were superimposable with identical morphology and margins. In 64% of cases, lesion margins were sharper in SE than in B-mode. (iii) In 76% of cases, glioma margins were sharper in SE than in B-mode. Lesions morphology/dimensions in SE and in B-mode were superimposable in 89%. Low-grade (LGG) and high-grade (HGG) gliomas were significantly different in terms of stiffness and stiffness contrast between tumors and brain, LGG appearing stiffer while HGG softer than brain (all P < ·001). A threshold of 2.5 SE score had 85.7% sensitivity and 94.7% specificity in differentiating LGG from HGG.
Conclusion: SE allows to understand mechanical properties of the brain and lesions in examination and permits a better discrimination between different tissues compared to B-mode. Additionally, SE can differentiate between LGG and HGG.
Keywords: Brain tumor; Glioma; Intraoperative imaging; Intraoperative ultrasound; Sonoelastography; Strain elastography; Tumor stiffness.
Copyright © 2018 by the Congress of Neurological Surgeons.