Stratification of Grade of Spinal Cord Non-Ependymal Gliomas by Magnetic Resonance Imaging

World Neurosurg. 2019 May:125:e902-e908. doi: 10.1016/j.wneu.2019.01.205. Epub 2019 Feb 10.

Abstract

Objective: The purpose of this study was to demonstrate the usefulness of conventional magnetic resonance imaging (MRI) in stratification of spinal cord non-ependymal gliomas.

Methods: Forty-one patients (mean age, 29.9 ± 16.8 years; 27 men) with first-diagnosed spinal cord non-ependymal gliomas underwent conventional MRI including T1-weighted, T2-weighted (T2W) and contrast-enhanced T1-weighted imaging sequences. The tumors were removed by surgery and processed by histology 1 week after MRI. Patients were divided into low-grade and high-grade groups according to histological findings. The MRI features of gliomas including signal-to-noise-ratio (SNR), margins, heterogeneity, edema, enhancement, tumor size, longitudinal location, and presence of syrinx were measured and compared between low-grade and high-grade groups using the χ2 test and the Mann-Whitney U test.

Results: The SNR of low-grade gliomas on T2W images was significantly higher than that of high-grade gliomas (165.0 ± 108.7 vs. 98.6 ± 44.4; P = 0.016 by the Mann-Whitney U test). The margins of low-grade gliomas were found to be clearer compared to that of high-grade gliomas (54.5% vs. 5.3%; P = 0.001). Receiver-operating-characteristic analysis revealed that the area under the curve of 1/SNR, margin and the combination of 1/SNR and margin was 0.722 (95% confidence interval [CI], 0.564-0.880), 0.736 (95% CI, 0.578-0.894), and 0.845 (95% CI, 0.717-0.973), respectively. No significant differences were found in the heterogeneity, edema, enhancement level, enhancement SNR, size, longitudinal location, cyst, hemorrhage, and intramedullary location between low-grade and high-grade non-ependymal gliomas (all P > 0.05).

Conclusions: The characteristics of signal intensity and margin on T2W imaging are valuable in stratification of spinal cord non-ependymal gliomas.

Keywords: Gliomas; Grade; Magnetic resonance imaging; Spinal cord.

MeSH terms

  • Adult
  • Female
  • Glioma / diagnostic imaging*
  • Glioma / pathology
  • Humans
  • Image Interpretation, Computer-Assisted / methods
  • Magnetic Resonance Imaging / methods*
  • Male
  • Middle Aged
  • Neoplasm Grading / methods*
  • Neuroimaging / methods*
  • Spinal Cord Neoplasms / diagnostic imaging*
  • Spinal Cord Neoplasms / pathology