Lisa L Cook1, Paula J Foster, Stephen J Karlik. 1. Department of Physiology and Pharmacology, University of Western Ontario, 1151 Richmond Street, London, Ontario N6A 5C1, Canada.
Abstract
PURPOSE: To directly correlate spinal cord pathology of guinea pigs with experimental allergic encephalomyelitis (EAE) to the MRI data obtained at 1.5T. MATERIALS AND METHODS: Spinal cords from EAE animals were imaged in vivo with the following MRI sequences: T2-FSE, PD-FSE, fluid-attenuated inversion recovery (FLAIR)-FSE, T2-CSE, T1-CSE, T1-CSE + gadolinium-DTPA (Gd-DTPA), PD-CSE, and short-tau inversion recovery (STIR)-FSE. The spinal cords were removed and the lesions with specific pathological compositions were identified by histological analysis. Regions of interest (ROIs) were drawn on the corresponding MR images, and signal-to-noise ratios (SNRs) were measured for each MR sequence and compared with controls. RESULTS: The receiver operating characteristic (ROC) analysis of STIR-FSE and PD-CSE was able to differentiate tissue that contained cellular infiltrates with a high degree of accuracy. The SNRs of T2-FSE, STIR-FSE, T2-CSE, PD-CSE, and T1-CSE + Gd-DTPA were elevated in lesions that contained cellular infiltrates alone, whereas the SNRs of PD-CSE and T1-CSE + Gd-DTPA were reduced in demyelinated lesions that also contained inflammation. CONCLUSION: The SNR difference between the two lesion groups suggests that the combination of STIR-FSE, PD-CSE, and T1-CSE + Gd-DTPA sequences may be useful for differentiating inflammatory lesions containing demyelination from lesions with inflammation alone. (c) 2005 Wiley-Liss, Inc.
PURPOSE: To directly correlate spinal cord pathology of guinea pigs with experimental allergic encephalomyelitis (EAE) to the MRI data obtained at 1.5T. MATERIALS AND METHODS: Spinal cords from EAE animals were imaged in vivo with the following MRI sequences: T2-FSE, PD-FSE, fluid-attenuated inversion recovery (FLAIR)-FSE, T2-CSE, T1-CSE, T1-CSE + gadolinium-DTPA (Gd-DTPA), PD-CSE, and short-tau inversion recovery (STIR)-FSE. The spinal cords were removed and the lesions with specific pathological compositions were identified by histological analysis. Regions of interest (ROIs) were drawn on the corresponding MR images, and signal-to-noise ratios (SNRs) were measured for each MR sequence and compared with controls. RESULTS: The receiver operating characteristic (ROC) analysis of STIR-FSE and PD-CSE was able to differentiate tissue that contained cellular infiltrates with a high degree of accuracy. The SNRs of T2-FSE, STIR-FSE, T2-CSE, PD-CSE, and T1-CSE + Gd-DTPA were elevated in lesions that contained cellular infiltrates alone, whereas the SNRs of PD-CSE and T1-CSE + Gd-DTPA were reduced in demyelinated lesions that also contained inflammation. CONCLUSION: The SNR difference between the two lesion groups suggests that the combination of STIR-FSE, PD-CSE, and T1-CSE + Gd-DTPA sequences may be useful for differentiating inflammatory lesions containing demyelination from lesions with inflammation alone. (c) 2005 Wiley-Liss, Inc.
Authors: Archana A Gupta; Di Ding; Richard K Lee; Robert B Levy; Sanjoy K Bhattacharya Journal: Invest Ophthalmol Vis Sci Date: 2012-02-13 Impact factor: 4.799