RESEARCH PAPER Improved detection of cortical MS lesions with phase-sensitive inversion recovery MRI Varun Sethi, 1,2 Tarek A Yousry, 1,3,4 Nils Muhlert, 1,2 Maria Ron, 1,2 Xavier Golay, 1,3 Claudia Wheeler-Kingshott, 1,2 David H Miller, 1,2 Declan T Chard 1,2 ABSTRACT Objective Cortical grey matter lesions are common in multiple sclerosis (MS), but usually not seen on MRI. The authors compared the performance of double inversion recovery (DIR, currently considered the best available imaging sequence for detecting cortical lesions) with phase-sensitive inversion recovery (PSIR, a sequence allowing much higher resolution scans to be obtained in a clinically feasible time). Methods Sixty MS patients and 30 healthy controls underwent MRI scanning on a 3 Tesla scanner. The authors compared intracortical (IC) and leucocortical (LC) lesion counts obtained with a standard DIR sequence (13133 mm resolution, obtained in 4 min) and a PSIR sequence (0.530.532 mm, 11 min). Lesions were marked separately on DIR and PSIR scans. Results In the whole MS cohort, more cortical lesions were seen on the higher-resolution PSIR than the DIR scans (IC mean6SD: 18.169.8 vs 5.964.5, p<0.001; LC mean6SD: 13.4612.9 vs 7.368.0, p<0.001). On PSIR, $1 IC lesion was seen in 60/60 MS patients and 1/30 controls, and $1 LC lesion in 60/60 patients and 6/ 30 controls. On DIR, $1 IC lesion was seen in 50/60 patients and 0/30 controls, and $1 LC lesion(s) in 60/60 patients and 5/30 controls. Conclusions Compared with DIR, using PSIR the authors are able to detect a significantly greater number of cortical grey matter lesions. The presence of at least one IC lesion in every MS patient, but very few healthy controls, suggests that it may be a useful adjunct to conventional MRI when a diagnosis of MS is suspected but not confirmed. INTRODUCTION Cortical grey matter lesions are a common finding in neuropathological studies of multiple sclerosis (MS), 1e4 and in some studies are more extensive than lesions seen in white matter. However, conventional MRI using T2-weighted, or fluid- attenuated inversion recovery (FLAIR) sequences, reveals few cortical lesions, and is considerably less sensitive to cortical than white matter lesions. 256 A sensitive method to detect cortical lesions in vivo might assist the diagnosis of MS and provide a new way to monitor the course of MS and its modifi- cation by treatment. Double inversion recovery (DIR), a sequence designed to suppress signals from white matter and cerebrospinal fluid, detects more cortical lesions than FLAIR sequences. 2578 Several studies have demonstrated associations of cortical lesions load on DIR with physical and cognitive impair- ment, 1 2 9e14 suggesting that they are clinically relevant. However, DIR imaging prospectively detects only 18% of pathologically confirmed cortical lesions. 5 15 Furthermore, DIR sequences are limited by low signal-to-noise ratio (SNR), and image intensity, non-uniformity and blood flow artifacts may be misinterpreted as lesions. DIR scans can be read in combination with another image to improve accuracy and sensitivity. 9 Phase-sensitive inversion recovery (PSIR) is a T1- weighted sequence with phase-sensitive recon- struction that provides a greater range of signal intensity than conventional T1-weighted inversion recovery sequences by additively combining nega- tive and positive longitudinal magnetisation in the image. 10 It provides better grey matterewhite matter contrast than other T1-weighted sequences, 16 and when acquired at the same reso- lution as DIR, was useful in confirming the pres- ence of MS cortical lesion and avoiding ‘ false positives’. 10 The combination of high SNR and contrast has facilitated development on a 3 Tesla scanner of a 2- dimensional (2D) PSIR sequence that, at 0.530.5 mm 2 in-plane and 2 mm slice thickness (equal to a voxel size of 0.5 mm 3 ), provides up to 6 times greater resolution than 2D DIR acquired with a typical resolution of 131 mm in-plane and 3 mm slice thickness 2 17 (voxel size of 3 mm 3 ), and 2e4 times more than previously reported 3D DIR sequences (with voxel sizes ranging between 1 and 2 mm 3 ). 5 18e20 Except at ultra-high field strengths (eg, 7 Tesla 20 ), sub-millimetre voxel sizes cannot be obtained using DIR sequences without prohibi- tively long scan times, but are achievable using the PSIR sequence in clinically acceptable times (in this study, 11 min). The higher resolution on PSIR improves definition of the cortex and its surroundings, and should facilitate detection of grey matter lesion. The main aim of our study was to compare the higher-resolution PSIR versus a standard resolution DIR sequence when used independently to detect cortical lesions in MS. 2 For comparison, we chose a 2D DIR sequence that has been extensively studied in MS populations. 2 Additionally, we investigated: (1) the frequency of cortical lesions in MS patients and healthy controls to explore their specificity for MS and (2) cortical lesions in MS clinical subgroups, and their associations with clinical measures. < Additional tables and figure are published online only. To view these files please visit the journal online (http://dx.doi.org/ 10.1136/jnnp-2012-303023/ content/early/recent). 1 Queen Square MS Center, NMR Research Unit, Department of Neuroinflammation, UCL Institute of Neurology, London, UK 2 Departments of Neuroinflammation, UCL Institute of Neurology, London, UK 3 Brain Repair and Rehabilitation, UCL Institute of Neurology, London, UK 4 Lysholm Department of Neuroradiology, National Hospital for Neurology and Neurosurgery, London, UK Correspondence to Dr Varun Sethi, Queen Square MS Center, NMR Research Unit, Department of Neuroinflammation, UCL Institute of Neurology, Queen Square, London WC1N 3BG, UK; varun.sethi.09@ucl.ac.uk For author footnote see end of the article. Received 18 April 2012 Revised 1 June 2012 Accepted 5 June 2012 Sethi V, Yousry TA, Muhlert N, et al. J Neurol Neurosurg Psychiatry (2012). doi:10.1136/jnnp-2012-303023 1 of 6 Multiple sclerosis JNNP Online First, published on July 17, 2012 as 10.1136/jnnp-2012-303023 Copyright Article author (or their employer) 2012. Produced by BMJ Publishing Group Ltd under licence. group.bmj.com on June 26, 2013 - Published by jnnp.bmj.com Downloaded from