Magnetic Resonance Imaging, Vol. 15, No. 2, pp. 155-162, 1997 Copyright 0 1997 Elsevier Science Inc. Printed in the USA. All rights reserved 0730-725XB7 $17.00 + 40 ELSEVIER PII 50730-725X( 96 ) 00340-5 COMPARISON OF FOUR POTENTIAL MR PARAMETERS FOR SEVERE TISSUE DESTRUCTION IN MULTIPLE SCLEROSIS LESIONS JAN HEIN T. VAN WAESBERGHE, *$ JONAS A. CASTELIJNS, *§ PHILTP SCHELTENS,~~ Luc TRUYEN,~~ GEERT J. LYCKLAMA A NIJEHOLT, *$ FRANK G. HOOGENRAAD,$ CHRIS H. POLMAN,@ JAAP VALK$ AND FREDERIK BARRHOF,~ Departments of *Diagnostic Radiology, INeurology, $Clinical Physics & Medical Engineering, and the §MR Center for MS Research, Academic Hospital “Vrije Universiteit,” Amsterdam, The Netherl~ds The purpose of this study was, first, to evaluate correlations between four potential magnetic resonance (MR) parameters for severe tissue destruction in multiple sclerosis (MS) lesions. Second, to evaluate the effect of incidental magnetization transfer (MT) effect on hypointense lesions in multislice T1 spin echo (SE) imaging. In 49 lesions, from 10 MS patients, MT ratio (MTR), T1 relaxation time, signal intensity (SI) on TX-weighted SE images normalized to normal-appearing white matter (NAWM), and SI normalized to eerebrospinal fluid (CSF) were measured. Differences in contrast of hypointense lesions were measured between single slice and multislice imaging. MTR correlated sign~~n~y (p < .OOl) with longitudinal relaxation rates ( l/T1) (r = 0.84)) SI normalized to NAWM (r = 0.78), and SI normalized to CSF (r = 0.75). The degree of reduction in contrast, caused by multislice imaging, correlated significantly with MTR of lesions (r = 0.60, p < .OOl ), leading to reduction of hypointense lesion load (p < .OI) . We conclude that all four MR markers for severe tissue destruction are highly correlated when applied to selected hypointense lesions on T,-weighted SE images. Due to “incidental” off-resonance excitation, contrast and hypointense lesion load will be reduced in multislice TV-weights SE images. 0 1997 Elsevier Science Inc. Kevwords: Sclerosis,multiule: Brain, MR; Magnetization transfer contrast; T1 relaxation time; Magnetic resonance;Tissue charact&ization. ’ ’ - INTRODUCTION Magnetic resonance (MR) imaging is the most sensi- tive paraclinical parameter in diagnosing multiple scle- rosis (MS) .I,’ In patients suspected of MS, Tz- weighted spin-echo (SE) sequences are very sensitive in demonstrating dissociation in spacea In patients fol- lowed from onset, a moderately strong correlation (r = 0.62) exists between T2 lesion load at presentation and development of disability at 5-yr follow-up.4 How- ever, in patients with established MS, increase in lesion load as shown by Tz-weighted SE images is poorly correlated with increase in clinical disability, as mea- sured by the expanded disability status scale (EDSS) scoring method?T6 Although there are several possible explanations for this clinicoradiologic paradox,7 one important factor is that Tz-weighted MR imaging is not very specific. Inflammation, edema, demyelination, gliosis, and axonal loss are all represented as areas of high signal intensity (ST) on T,-weighted SE images, while not all of these processes are clinically relevant in the long term. Possibly, even remyelinated plaques, which show an abnormal myelin structure,8*9 may still produce abnormal SI on Tz-weighted SE images, while ~nction~ly they are competent. Persistent demyelin- ation and axonal loss most likely cause disability, whereas other pathologic features like edema and in- flammation do not. In addition to T,-weighted SE im- aging, more specific MR parameters are, therefore, needed that selectively reflect clinically relevant histo- pathological features. Recent studies have focussed on the presence of RECEIVED 5123/96; ~CEPTED 9115196. Address correspondence to J.H.T.M. van Waesberghe, M.D., ~p~ent of Diagnostic Radiology, MR Center for MS Research, Academic Hospital “Vrije Universiteit,” De Boelelaan I 117, 1081 HV Amsterdam, The Netherlands. 155