The value of qualitative and quantitative assessment of lesion to cerebral cortex signal ratio on double inversion recovery sequence in the differentiation of demyelinating plaques from non-specific T2 hyperintensities

• Published in: European radiology 2017-02, Vol.27 (2), p.763-771
• Main Authors: Hamcan, Salih, Battal, Bilal, Akgun, Veysel, Oz, Oguzhan, Bozkurt, Yalcin, Tasdemir, Serdar, Sari, Sebahattin, Tasar, Mustafa
• Format: Article
• Language: English
• Subjects: Adult; Case-Control Studies; Cerebral Cortex - diagnostic imaging; Cerebral Small Vessel Diseases - diagnostic imaging; Demyelinating Diseases - diagnostic imaging; Diagnostic Radiology; Female; Humans; Hypertension - diagnostic imaging; Image Processing, Computer-Assisted; Imaging; Internal Medicine; Interventional Radiology; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Medicine; Medicine & Public Health; Middle Aged; Migraine Disorders - diagnostic imaging; Multiple sclerosis; Multiple Sclerosis - diagnostic imaging; Neuro; Neuroradiology; Observer Variation; Radiology; Retrospective Studies; Sensitivity and Specificity; Spinal cord; Ultrasound; Vasculitis - diagnostic imaging; Young Adult
• ISSN: 0938-7994; 1432-1084
• DOI: 10.1007/s00330-016-4379-2

Objectives To assess the usefulness of the visual assessment and to determine diagnostic value of the lesion-to-cerebral cortex signal ratio (LCSR) measurement in the differentiation of demyelinating plaques and non-specific T2 hyperintensities on double inversion recovery (DIR) sequence. Material and methods DIR and fluid-attenuated inversion recovery (FLAIR) sequences of 25 clinically diagnosed multiple sclerosis (MS) patients and 25 non-MS patients with non-specific T2-hyperintense lesions were evaluated visually and LCSRs were measured by two observers independently. Results On DIR sequence, the calculated mean LCSR ± SD for demyelinating plaques and non-specific T2-hyperintense lesions were 1.60 ± 0.26 and 0.75 ± 0.19 for observer1, and 1.61 ± 0.27 and 0.74 ± 0.19 for observer2. LCSRs of demyelinating plaques were significantly higher than other non-specific T2-hyperintense lesions on DIR sequence. By using the visual assessment demyelinating plaques were differentiated from non-specific T2-hyperintensities with 92.8 % sensitivity, 97.5 % specificity and 95.1 % accuracy for observer1 and 92.8 % sensitivity, 95 % specificity and 93.9 % accuracy for observer2. Conclusion Visual assessment and LCSR measurement on DIR sequence seems to be useful for differentiating demyelinating MS plaques from supratentorial non-specific T2 hyperintensities. This feature can be used for diagnosis of MS particularly in patients with only supratentorial T2-hyperintense lesions who are categorized as radiologically possible MS. Key Points • Demyelinating plaques and non-specific T2-hyperintensities have different SI on DIR images . • These differences can be assessed by LCSR measurement or visual assessment . • There is an excellent inter-observer agreement for both methods . • This feature can be used in radiologically possible MS cases .