Main Article Content

Differentiation between various types and subtypes of intracranial meningiomas with advanced MRI


Mousam Panigrahi
Narendra K. Bodhey
Saroj K. Pati
Nighat Hussain
Anil K. Sharma
Arvind K. Shukla

Abstract

Background: Meningiomas are the most prevalent of all intracranial tumours. Although they are mostly benign, about 20% of meningiomas are atypical or malignant. Knowledge of their histologic grade can be clinically useful while planning surgery.
Objectives: To differentiate between various grades and subtypes of meningiomas with advanced MR parameters.
Method: We assessed the advanced MR imaging characteristics of 27 histopathologically confirmed meningiomas on a 3T MRI, of which 23 were grade I meningiomas (2 fibroblastic, 9 meningothelial, 9 transitional, 3 unspecified) and 4 were grade II/III meningiomas (2 atypical, 1 papillary, 1 anaplastic). Analysis of the ADC, FA, λ1, λ2, λ3 and mean diffusivity was performed using standard post-processing software.
Results: The mean size of atypical meningiomas (5.9 cm ± 0.7 cm) was significantly higher (p = 0.038, 95% confidence interval [CI]) than that of typical meningiomas (4.6 cm ± 1.6 cm) with a cut-off value of 6.05 cm (75% sensitivity and 87% specificity). The mean cerebral blood flow (CBF) (ASL) of atypical meningiomas (286.70 ± 8.06) was significantly higher (p = 0.0000141, 95% CI) than that of typical meningiomas (161.09 ± 87.04) with a cut-off value of 276.75 (66.7% sensitivity and 75% specificity). Among the typical meningiomas, transitional subtypes had the lowest ADC. High FA and planar coefficient (CP) values and low λ3 and spherical coefficient (CS) values were seen in fibroblastic meningiomas. Fibroblastic meningiomas also showed the lowest vascularity among typical meningiomas.
Conclusion: Tumour size and ASL perfusion are two parameters that could differentiate between typical and atypical meningiomas while ADC, FA, λ3, CP, CS, rCBF and rCBV may be helpful in distinguishing different subtypes of typical meningiomas.


Journal Identifiers


eISSN: 2078-6778
print ISSN: 1027-202X