Journal of the Neurological Sciences
Volume 206, Issue 2 , Pages 175-179, 15 February 2003

MRI-clinical correlations: more than inflammation alone-what can MRI contribute to improve the understanding of pathological processes in MS?

Department of Pathology and Laboratory Medicine, University of British Columbia, Vancouver Hospital and Health Sciences Centre, Vancouver, BC, Canada V5Z 1L8

Article Outline

Abstract 

Magnetic resonance imaging (MRI) is a sensitive technique in the detection and follow-up of multiple sclerosis (MS) lesions. However, the pathologic basis of these changes is poorly understood. It is becoming increasingly apparent that routine MRI techniques do not represent specific histopathologic changes but reflect a variety of pathological changes in tissue. This review will outline the findings in MRI-pathology correlation in MS to date, and the current understanding of the evolution of pathologic changes in the MS lesion and non-lesional white matter as reflected by MRI. Possible future contributions of MRI to unveiling the dynamic pathology of MS will also be discussed.

Keywords:  Multiple sclerosis, Pathology, Magnetic resonance imaging

 

MRI has proven to be a very sensitive technique for the detection of multiple sclerosis (MS) lesions and has provided remarkable insight into the dynamic nature of the disease process [1], [2], [3]. However, our understanding of the relationship of these imaging findings to changes occurring in the tissue is rudimentary, and very few formal studies have addressed this issue. This review will briefly outline the progress in this field to date, and will point out future avenues of research whereby MRI can contribute to improve our understanding of the pathologic process in MS.

Back to Article Outline

1. Technical issues 

It is important to realize that there are technique issues which must be considered in studies of MRI-pathology correlation in MS. This first is tissue integrity. Clearly, the closest match to the in vivo condition is biopsy, which unfortunately is limited by the tissue available and is usually performed under circumstances where MS is usually not the primary clinical diagnosis at the time.

While autopsy provides the investigator with the entire CNS for study, autolysis and the effects of fixatives are important considerations when correlating the pathologic findings with MRI. From the point of view of autolysis, interestingly T1 and T2 relaxation times are relatively stable up to 24 h post-mortem [4], [5] suggesting that CNS harvested up to that time would be appropriate material for MRI-pathology correlation studies. Once the brain is fixed in formalin, there is a shortening of both T1 and T2 [4], [6], [7], [8]. Nevertheless, it has been found that formalin-fixed tissue is useful for MRI-pathology correlation studies [6].

Another important issue is that of volume averaging. The MRI slice usually is several millimeters thick. Its image is actually an “average” of the tissue densities throughout the slice thickness. Thus, the MRI picture represents the averaging of various pathologic changes or normal parenchyma through the thickness of the tissue slice imaged. This is particularly pertinent to lesions with irregular borders and small lesions which may be averaged in with normal appearing white matter above or below the lesion. In contrast, the histologic section representing that MRI slice has a thickness of only a few microns and is thus subject to much less tissue variability through its thickness. One way to control somewhat for this discrepancy is to take several histologic sections through the tissue represented by the MRI slice.

Back to Article Outline

2. Previous MRI-pathology correlation studies of T1 and T2 

Since the introduction of MRI in the diagnosis and management of MS, there has been an effort to correlate changes seen with various MRI techniques with specific histopathologic features. In the case of T2 which shows MS lesions exquisitely, it appears that a variety of histopathologic features, either singly or in combination, can result in the same T2 change. Thus, demyelination [9], the presence of macrophages [10], gliosis [11], edema [12], expansion of the extracellular space [13], and vascular permeability [14] have all been reported to be responsible for the MRI abnormalities in MS. With respect to T1, it has been shown that the irreversible “black holes” seen with this technique correlate with tissue destruction and axonal loss [15].

The enhancement seen in MS plaques after gadolinium-DTPA is thought to be due to blood–brain barrier breakdown secondary to the vascular inflammatory infiltrates. Pathologic studies have supported this notion, in that lesions with enhancement show inflammatory activity, the blood–brain abnormality being attributed to either the macrophage infiltrate [10] or perivascular lymphocytic infiltrates [16]. Recent pathologic studies have shown a considerable degree of vascular neogenesis at the periphery of MS lesions [17] and this may be a significant contribution to the enhancement.

Thus, it would seem that the changes seen on the T2 scan in MS are not histopathologic specific and that a given T2 abnormality may have a variety of histologic features as its basis. On the other hand, the chronic “black holes” seen on T1 images probably represent areas of irreversible tissue destruction.

Back to Article Outline

3. Histopathologic specificity of MRI techniques in MS 

There are some more recently introduced MRI techniques which show promise for histopathologic specificity. These include magnetic resonance spectroscopy (MRS) techniques. The fact that these spectra actually represent signal from specific chemical compounds would imply that detection of chemical markers specific for a certain structure indicates the presence of that structure in the area imaged. One such marker is N-acetyl aspartate (NAA) which is an axonal/neuronal-specific compound which is decreased in chronic MS plaques [18], correlating with the finding of axonal loss in pathologic studies [19]. Abnormal lipid peaks are detected in MS plaques by MRS and these have been attributed to myelin breakdown products [20].

Recent studies have focused on the use of magnetization transfer imaging (MTI) in MS which also shows abnormalities in the MS lesion [21]. Experimental and clinical studies are attempting to determine the basis of these abnormalities [21], [22]. New MTI techniques, such as fast phase acquisition of composite echoes (FastPACE), by analyzing the physical variables contributing to the MTI image, offer promise of being able to distinguish contributions from tissue components to the image [23].

Diffusion-weighted MRI is another new technique which is being applied to MS. Since it measures water molecule diffusion, which can be altered by physical barriers such as normal and pathologic tissue components, it may provide important contributions to the in vivo demonstration of MS pathology, particularly with respect to involvement of fibre tracts as visualized by diffusion tensor imaging [24].

A promising MRI technique from the point of view of monitoring myelin changes in MS is the short-T2 component [25]. The short-T2 component is the shortest of three components of the T2 relaxation distribution [25] and has a duration of 10 to 50 ms. The longest (greater than 1 s) originates from CSF and the intermediate (70–95 ms) is thought to come from extracellular and intracellular fluid [25]. That the short-T2 component emanates from myelin is supported by a MRI-pathology correlative study which found that its distribution corresponds to the distribution of myelin and it is absent in chronic MS plaques which are completely devoid of myelin [26].

Back to Article Outline

4. Histopathologic correlates of patterns within MS lesions on MRI 

The topography of changes in a given MS lesion on MRI can correlate with histopathologic features and provide some insight as to the nature of that lesion. For instance, it is thought that the peripheral rim of abnormality sometimes evident in MS plaques on the T2 image is due to the presence of numerous macrophages at the border of the lesion [10], [27]. Similarly, the concentric rings of demyelination and myelin preservation seen in Baló's concentric sclerosis can also be detected by MRI [28]. Further attention to topographic patterns in MS lesions on MRI may provide important clues in the pathogenesis of these lesions and the stage of maturation of a given lesion.

Back to Article Outline

5. Non-lesional white matter in MS 

Examples of comments on changes in the normal appearing white matter, or non-lesional white matter, in MS [29] have been recorded in the pathology literature for some time. Foci of vascular fibrosis, perivascular inflammation and demyelination had been described at the light microscopic level [30], [31] and increased lysosomes, particularly in astrocytes, at the untrastructural level [32]. Recent studies of the normal appearing white matter have shown axonal loss in these regions [33], and that this appears to be secondary to destruction of axons within plaques [34].

From the MRI point of view, while abnormalities have been documented in normal appearing white matter [35], correlation with histopathologic changes has not been carried out. Thus, when high-field strength magnets [36] or thin-slice imaging [37] is employed, abnormalities in the form of very small focal lesions [35] are detected in white matter which appears normal on routine MRI. In addition, diffuse abnormalities are detected in non-lesional white matter in as much as there is reduction of NAA by MRS [38], reduction of magnetization transfer ratio (MTR) on MTI [39], and wide spread abnormalities in diffusion-weighted imaging [24]. These focal and diffuse abnormalities have aroused new interest in the non-lesional white matter in MS, but the tissue changes which underlie them are not characterized.

Back to Article Outline

6. The initial lesion in MS 

Arising within the normal appearing white matter are the small initial lesions in MS, many of which eventually expand and become well-established chronic MS plaques. The initial pathologic change and pathogenetic event in MS is unknown. However, MRI studies are providing some insight into this and eventual pathologic correlation will be very informative. In this regard, it has been shown that the very earliest change at the site of a future MS plaque is a change in MTR which can antedate gadolinium enhancement by months [40]. Clearly, pathologic correlation studies of these lesions, which have MTR abnormality but no other MRI abnormality, will be important in determining the early events in the pathogenesis of the focal lesion in MS. Similarly, the small transient lesions seen on routine MRI [41] will provide important clues to understanding the initiation of the MS plaque.

Back to Article Outline

7. The progression of MS 

From the point of view of MRI contribution to understanding the pathogenesis of progression in MS, this will be discussed in terms of what is known of the MRI correlates of histopathology with respect to the focal lesion and with respect to the more widespread changes in normal appearing white matter. The following approach is one interpretation of the current data available.

At the present time, it appears the initial MRI abnormality at the location of a future MS lesion is a focal change in MTR, which may persist for some time before being detected by other MRI techniques. The pathologic basis of this early abnormality, while clearly very important, is unknown. The subsequent gadolinium enhancement at this site is indicative of neovascularization and sufficient vascular inflammatory infiltrates to cause breakdown of the blood–brain barrier to the point where gadolinium leakage from vessels is detectable. Once the lesion is established as a focal hyperintensity on the T2 scan, a variety of histopathologic reactions are present—including demyelination, gliosis, varying degrees of edema, etc. However, specific histopathologic features may be detected by certain techniques. For example, the finding of loss of the short-T2 component would indicate demyelination. Lipid peaks on MRS would indicate ongoing myelin breakdown into neutral lipid as part of the demyelinating process. Reduction of NAA would be indicative of axonal loss in the lesion. In this manner, by applying a variety of MRI techniques to analyze a given lesion, a picture of the histopathologic features can be surmised.

As the lesion, with its peripheral rim of active demyelination and gadolinium enhancement, expands into the adjacent white matter, a temporal topographic gradient results such that the center of lesion shows older changes and progressively more recent changes are evident more peripherally up to the border of on-going demyelination. It is possible future MRI technology will be able to detect such gradients more effectively than current techniques. Indeed, it may be possible to detect remyelination at the border and within active MS plaques and in shadow plaques, utilizing techniques such as the short-T2 distribution to image myelin. For instance, the return of the short-T2 component to a region where it was previously absent may indicate remyelination.

It has recently become clear that routine T2 scans can show ill-defined slight T2 abnormalities which have been referred to as “dirty white matter” [42], which also shows abnormalities on MTI [43]. The histologic correlate for this is unknown but may well be secondary change which will prove useful in following the progression or regression of lesions.

As indicated previously those lesions which evolve into permanent “black holes” on T1 imaging are characterized by significant parenchymal destruction that is irreversible and represent the end-stage of lesion progression.

With respect to the MRI abnormality in normal appearing white matter, future pathology correlative studies should utilize thin slices and high field strength magnets to determine the nature of the very small focal lesions seen in normal appearing white matter by these techniques. Similarly, the diffuse changes in normal appearing white matter detected by MTI and diffusion-weighted imaging should also be subject to pathologic correlation. It would be anticipated that axonal loss from Wallerian degeneration of axons transected in focal lesions would be an important component of the diffuse white matter abnormality, but this must await pathologic correlation.

Back to Article Outline

8. Future MRI-pathology correlation studies in MS 

Since the initial MRI-pathology study in MS [44], it has been established that MRI detects MS lesions and such studies are continuing to dissect out the changes in tissue responsible for the MRI abnormalities. These changes in the MS lesion and in normal appearing white matter are highly dynamic. However, histology images are static snap-shots in time. The translation of findings from this static image into the dynamic process in MS is often difficult. Comparison of previous MRI scans done on patients with that done on the post-mortem tissue sampled for pathologic correlation may prove to be one way to bridge this gap. In the case of biopsy studies, repeated post-biopsy follow-up with respect to changes on the MRI and clinical behavior may also contribute significantly to our understanding of lesion evolution in MS [45].

Back to Article Outline

9. Summary 

While conventional MRI techniques, particularly T2 relaxation, display the lesions of MS exquisitely, these techniques are not histopathologic specific. There are, however, a number of techniques which offer promise of histopathologic specificity. In the future, the combination of various techniques, some of which have histopathologic specificity, may provide a composite picture of the various pathologic features of MS lesions. This is particularly important in the case of very early lesions. Even though histopathology may not be predictable by certain techniques, patterns seen within lesions might provide important information concerning their pathology or age, and degree of maturation. The evolution of these changes over time, both in lesions and in non-lesional white matter, may become apparent in MRI-pathology correlation studies which correlate not only the scan taken at the time of the tissue sample but antecedent and subsequent scans. Thus, much work needs to be done in this relatively young field of research which is attempting to translate the dynamic changes seen in the imaging of MS into biological reality.

Back to Article Outline

Acknowledgements 

The author would like to thank his collaborators in the Department of Medicine (Neurology), Radiology (Neuroradiology), Pathology (Neuropathology), Physics and Astronomy, the MS Clinic and the UBC-MRI Research Group at the University of British Columbia (UBC) and Vancouver Hospital and Health Sciences Centre for helpful advice; and Esther Leung for technical support. MRI-Pathology correlation research in the author's laboratory is supported by research grants from the Multiple Sclerosis Society of Canada and Berlex.

Back to Article Outline

References 

  1. Paty DW, Oger JJF, Kastrukoff LF, Hashimoto SA, Hooge JP, Eisen AA, et al.  MRI in the diagnosis of MS: a prospective study with comparison of clinical evaluation, evoked potentials, oligoclonal banding, and CT. Neurology. 1988;38:180–185
  2. McDonald WI, Miller DH, Barnes D. The pathological evolution of multiple sclerosis. Neuropathol. Appl. Neurobiol. 1992;18:319–334
  3. Fazekas F, Barkhof F, Filippi M, Grossman RI, Li DKB, McDonald WI, et al.  The contribution of magnetic resonance imaging to the diagnosis of multiple sclerosis. Neurology. 1999;53:448–456
  4. Kamman RL, Go KG, Stomp GP, Hulstaert CE, Berendsen HJC. Changes of relaxation times T1 and T2 in rat tissues after biopsy and fixation. Magn. Reson. Imaging. 1985;3:245–250
  5. Györffy-Wagner Z, Englund E, Larsson EM, Brun A, Cronqvist S, Persson B. Proton magnetic resonance relaxation times T1 and T2 related to post-mortem interval. An investigation on porcine brain tissue. Acta Radiol. Diagn. (Stockholm). 1986;27:115–118
  6. Nagara H, Inoue T, Koga T, Kitaguchi T, Tateishi J, Goto I. Formalin fixed brains are useful for magnetic resonance imaging (MRI) study. J. Neurol. Sci. 1987;81:67–77
  7. Grafe MR, Press GA, Berthoty DP, Hesselink JR, Wiley CA. Abnormalities of the brain in AIDS patients: correlation of postmortem MR findings with neuropathology. Am. J. Neuroradiol. 1990;11:905–911
  8. Unger EC, Gado MH, Fulling KF, Littlefield JL. Acute cerebral infarction in monkeys: an experimental study using MR imaging. Radiology. 1987;162:789–795
  9. Estes ML, Rudick RA, Barnett GH, Ransohoff RM. Stereotactic biopsy of an active multiple sclerosis lesion. Immunocytochemical analysis and neuropathologic correlation with magnetic resonance imaging. Arch Neurol. 1990;47:1299–1303
  10. Nesbit GM, Forbes GS, Scheithauer BW, Okazaki H, Rodriguez M. Multiple sclerosis: histopathologic and MR and/or CT correlation in 37 cases at biopsy and three cases at autopsy. Radiology. 1991;180:467–474
  11. Stewart WA, Hall LD, Berry K, Churg A, Oger J, Hashimoto SA, et al.  Magnetic resonance imaging (MRI) in multiple sclerosis (MS): pathological correlation studies in eight cases. Neurology. 1986;36(suppl. 1):320; [abstract]
  12. Ormerod IEC, Miller DH, McDonald WI, du Boulay EPGH, Rudge P, Kendall BE, et al.  The role of NMR imaging in the assessment of multiple sclerosis and isolated neurological lesions. A quantitative study. Brain. 1987;110:1579–1616
  13. Barnes D, Munro PMG, Youl BD, Prineas JW, McDonald WI. The longstanding MS lesion: a quantitative MRI and electron microscopic study. Brain. 1991;114:1271–1280
  14. Newcombe J, Hawkins CP, Henderson CL, Patel HA, Woodroofe MN, Hayes GM, et al.  Histopathology of multiple sclerosis lesions detected by magnetic resonance imaging in unfixed central nervous system tissue. Brain. 1991;114:1013–1023
  15. van Walderveen MAA, Kamphorst W, Scheltens P, van Waesberghe JHTM, Ravid R, Valk J, et al.  Histopathologic correlate of hypointense lesions on T1-weighted spin-echo MRI in multiple sclerosis. Neurology. 1998;50:1282–1288
  16. Katz D, Taubenberger JK, Cannella B, McFarlin DE, Raine CS, McFarland HF. Correlation between magnetic resonance imaging findings and lesion development in chronic, active multiple sclerosis. Ann. Neurol. 1993;34:661–669
  17. Ludwin SK, Henry JM, McFarland H. Vascular proliferation and angiogenesis in multiple sclerosis: clinical and pathogenetic considerations. In: MS 2000. Symposium on conquering multiple sclerosis: research in Canada in the new millennium. Multiple Sclerosis Society of Canada; 2000; Abstract
  18. Arnold DL, Matthews PM, Francis GS, O'Connor J, Antel JP. Proton magnetic resonance spectroscopic imaging for metabolic characterization of demyelinating plaques. Ann. Neurol. 1992;31:235–241
  19. Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mörk S, Bö L. Axonal transection in the lesions of multiple sclerosis. N. Eng. J. Med. 1998;338:278–285
  20. Wolinsky JS, Narayana PA, Fenstermacher MJ. Proton magnetic resonance spectroscopy in multiple sclerosis. Neurology. 1990;40:1764–1769
  21. Filippi M. Magnetization transfer imaging to monitor the evolution of individual multiple sclerosis lesions. Neurology. 1999;53(suppl. 3):S18–S22
  22. Brochet B, Dousset V. Pathological correlates of magnetization transfer imaging abnormalities in animal models and humans with multiple sclerosis. Neurology. 1999;53(suppl. 3):S12–S17
  23. Ropele S, Strasser-Fuchs S, Augustin M, Stollberger R, Enzinger C, Hartung H-P, et al.  A comparison of magnetization transfer ratio, magnetization transfer rate, and the native relaxation time of water protons related to relapsing-remitting multiple sclerosis. Am. J. Neuroradiol. 2000;21:1885–1891
  24. Filippi M, Inglese M. Overview of diffusion-weighted magnetic resonance studies in multiple sclerosis. J. Neurol. Sci. 2001;186(suppl. 1):S37–S43
  25. MacKay A, Whittall K, Adler J, Li D, Paty D, Graeb D. In vivo visualization of myelin water in brain by magnetic resonance. Magn. Reson. Med. 1994;31:673–677
  26. Moore GRW, Leung E, MacKay AL, Vavasour IM, Whittall KP, Cover KS, et al.  A pathology-MRI study of the short-T2 component in formalin-fixed multiple sclerosis brain. Neurology. 2000;55:1506–1510
  27. Brück W, Bitsch A, Kolenda H, Brück Y, Stiefel M, Lassmann H. Inflammatory central nervous system demyelination: correlation of magnetic resonance imaging findings with lesion pathology. Ann. Neurol. 1997;42:783–793
  28. Gharagozloo AM, Poe LB, Collins GH. Antemortem diagnosis of Baló concentric sclerosis: correlative MR imaging and pathologic features. Radiology. 1994;191:817–819
  29. Seitelberger F. Histochemistry of demyelinating diseases proper including allergic encephalomyelitis and Pelizaeus–Merzbacher's disease. In:  Cuming JN editors. Modern scientific aspects of neurology. London: Edward Arnold; 1960;p. 337–344
  30. Allen IV, McKeown SR. A histological, histochemical and biochemical study of the macroscopically normal white matter in multiple sclerosis. J. Neurol. Sci. 1979;41:81–91
  31. Allen IV, Glover G, Anderson R. Abnormalities in the macroscopically normal white matter in cases of mild or spinal multiple sclerosis (MS). Acta Neuropathol. Suppl. (Berlin). 1981;7:176–178
  32. Arstila AU, Riekkinen P, Rinne UK, Laitinen L. Studies on the pathogenesis of multiple sclerosis: participation of lysosomes on demyelination in the central nervous system white matter outside plaques. Eur. Neurol. 1973;9:1–20
  33. Evangelou N, Esiri MM, Smith S, Palace J, Mathews PM. Quantitative pathological evidence for axonal loss in normal appearing white matter in multiple sclerosis. Ann. Neurol. 2000;47:391–395
  34. Evangelou N, Konz D, Esiri MM, Smith S, Palace J, Mathews PM. Regional axonal loss in the corpus callosum correlates with cerebral white matter lesion volume and distribution in multiple sclerosis. Brain. 2000;123:1845–1849
  35. Filippi M, Tortorella C, Bozzali M. Normal-appearing white matter changes in multiple sclerosis: the contribution of magnetic resonance technique. Mult. Scler. 1999;5:273–282
  36. Karlik SJ, Erskine M, Lee D, Rice G. Comparison of 1.5 and 4.0 Tesla magnetic resonance imaging examinations of patients with multiple sclerosis. In: 123rd annual meeting of the American Neurological Society. 1998;p. 69; Abstract
  37. Barbosa S, Blumhardt LD, Roberts N, Lock T, Edwards RH. Magnetic resonance relaxation time mapping in multiple sclerosis: normal appearing white matter and the “invisible” lesion load. Magn. Reson. Imaging. 1994;12:33–42
  38. De Stefano N, Matthews PM, Antel JP, Preul M, Francis G, Arnold DL. Chemical pathology of acute demyelinating lesions and its correlation with disability. Ann. Neurol. 1995;38:901–909
  39. Filippi M, Campi A, Dousset V, Baratti C, Martinelli V, Canal N, et al.  A magnetization transfer imaging study of normal-appearing white matter in multiple sclerosis. Neurology. 1995;45:478–482
  40. Filippi M, Rocca MA, Martino G, Horsfield MA, Comi G. Magnetization transfer changes in the normal appearing white matter precede the appearance of enhancing lesions in patients with multiple sclerosis. Ann. Neurol. 1998;43:809–814
  41. Willoughby EW, Grochowski E, Li DKB, Oger J, Kastrukoff LF, Paty DW. Serial magnetic resonance scanning in multiple sclerosis: a second prospective study in relapsing patients. Ann. Neurol. 1989;25:43–49
  42. Zhao G, Li DKB, Wang X, Cheng Y, Paty DW. MRI dirty-appearing white matter in MS. Neurology. 2000;54(suppl. 3):A121; [abstract]
  43. Laule C, Moore GRW, Leung E, Michelin E, Whittall KP, MacKay AL, et al.  A serial MR study of dirty white matter in MS: T1, T2, MT, water content and myelin water content. Mult. Scler. 2001;7(suppl. 1):S90; [abstract]
  44. Stewart WA, Hall LD, Berry K, Paty DW. Correlation between NMR scan and brain slice data in multiple sclerosis. Lancet. 1984;2:412
  45. Bitsch A, Kuhlmann T, Stadelmann C, Lassmann H, Lucchinetti C, Brück W. A longitudinal MRI study of histopathologically defined hypointense multiple sclerosis lesions. Ann. Neurol. 2001;49:793–796

PII: S0022-510X(02)00347-7

Journal of the Neurological Sciences
Volume 206, Issue 2 , Pages 175-179, 15 February 2003

Article Tools