Format

Send to

Choose Destination
Clin Neuroradiol. 2014 Jun;24(2):129-34. doi: 10.1007/s00062-013-0234-x. Epub 2013 Jul 12.

Orbital masses: the usefulness of diffusion-weighted imaging in lesion categorization.

Author information

1
Department of Radiology, University of Yamanashi, 1110 Shimokato, 409-3898, Chuo-shi, Yamanashi, Japan.

Abstract

INTRODUCTION:

Diffusion-weighted imaging (DWI) produces contrast among different kinds of tissues according to their diffusibility characteristics. The purpose of our study was to evaluate the role of DWI including measurement of apparent diffusion coefficient (ADC) values in recognizing benignancy or malignancy of orbital masses.

METHODS:

A total of 39 orbital masses were evaluated visually for signal characteristics on DWI and ADC maps. ADC values were calculated for each lesion. Visual signal characteristics were compared using the Fisher exact test. Receiver operating characteristic (ROC) analysis was carried out to determine sensitivity and specificity for distinguishing malignant from benign lesions using ADC values. The Mann-Whitney U test was applied to compare the ADC values between orbital lymphomas and idiopathic orbital inflammatory (IOI) lesions, and between optic nerve sheath meningiomas and gliomas.

RESULTS:

Visual assessment revealed no significant difference between benign and malignant lesions on DWI (p-value = 0.66). However, visual assessment of ADC maps revealed a statistically significant (p-value ≤ 0.0001) between benign and malignant lesions. ROC analysis showed a sensitivity of 83.33 % and a specificity of 85.71 % when using an optimal cut off ADC value of 0.84 × 10(-3) mm(2)/s for differentiating malignant from benign lesions. Significant differences in mean ADC values were observed between lymphomas and IOI lesions (p-value = 0.05), and between optic nerve sheath meningiomas and gliomas (p-value = 0.03).

CONCLUSION:

DWI is useful for differentiating malignant and benign orbital tumors if accompanied by visual assessment of ADC maps and ADC value calculations.

PMID:
23846020
DOI:
10.1007/s00062-013-0234-x
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center