Conventional MRI Features of Central Nervous System Embryonal Tumor, Not Otherwise Specified in Adults: Comparison with Glioblastoma

Acad Radiol. 2022 Mar:29 Suppl 3:S44-S51. doi: 10.1016/j.acra.2021.01.010. Epub 2021 Jan 25.

Abstract

Rationale and objectives: The purpose of this study was to explore conventional MRI features that can accurately differentiate central nervous system embryonal tumor, not otherwise specified (CNS ETNOS) from glioblastoma (GBM) in adults.

Materials and methods: Preoperative conventional MRI images of 30 CNS ETNOS and 98 GBMs were analyzed by neuroradiologists retrospectively to identify valuable MRI features. Five blinded neuroradiologists independently reviewed all these MRI images, and scored MRI features on a five-point scale. Kendall's coefficient of concordance was used to measure inter-rater agreement. Diagnostic value was assessed by the area under the curve (AUC) of receiver operating curve, and sensitivity and specificity were also calculated.

Results: Seven MRI features, including isointensity on T1WI, T2WI, and FLAIR, ill-defined margin, severe peritumoral edema, ring enhancement, and broad-based attachment sign, were helpful for the differential diagnosis of these two entities. Among these features, ring enhancement showed the highest inter-rater concordance (0.80). Ring enhancement showed the highest AUC value (0.79), followed by severe peritumoral edema (0.67). The combination of seven features showed the highest AUC value (0.86), followed by that of three features (ill-defined margin, severe peritumoral edema, and ring enhancement) (0.83).

Conclusion: Enhancement pattern, peritumoral edema, and margin are valuable for the discrimination between CNS ETNOS and GBM in adults.

Keywords: Diagnostic imaging; Embryonal tumor; Glioma; Magnetic resonance imaging; ROC curve.

MeSH terms

  • Adult
  • Brain Neoplasms* / diagnostic imaging
  • Brain Neoplasms* / pathology
  • Central Nervous System / pathology
  • Glioblastoma* / diagnostic imaging
  • Glioblastoma* / pathology
  • Humans
  • Magnetic Resonance Imaging / methods
  • Margins of Excision
  • Retrospective Studies