Multiple Sclerosis - Semantic Scholar

Multiple Sclerosis http://msj.sagepub.com

Myelin water imaging in multiple sclerosis: quantitative correlations with histopathology C Laule, E Leung, D KB Li, A L Traboulsee, D W Paty, A L MacKay and G RW Moore Mult Scler 2006; 12; 747 DOI: 10.1177/1352458506070928 The online version of this article can be found at: http://msj.sagepub.com/cgi/content/abstract/12/6/747

Published by: http://www.sagepublications.com

Additional services and information for Multiple Sclerosis can be found at: Email Alerts: http://msj.sagepub.com/cgi/alerts Subscriptions: http://msj.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations (this article cites 25 articles hosted on the SAGE Journals Online and HighWire Press platforms): http://msj.sagepub.com/cgi/content/abstract/12/6/747#BIBL

Downloaded from http://msj.sagepub.com at TORONTO UNIVERSITY on February 9, 2007 © 2006 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

ARTICLE

Multiple Sclerosis 2006; 12: 747753

Myelin water imaging in multiple sclerosis: quantitative correlations with histopathology C Laule1,2,5, E Leung3,6, DKB Li2,5, AL Traboulsee4,5, DW Patya,4,5, AL MacKay1,2,5,6 and GRW Moore3,6 Various magnetic resonance (MR) techniques are used to study the pathological evolution of demyelinating diseases, such as multiple sclerosis (MS). However, few studies have validated MR derived measurements with histopathology. Here, we determine the correlation of myelin water imaging, an MR measure of myelin content, with quantitative histopathologic measures of myelin density. The multi-component T2 distribution of water was determined from 25 formalin-fixed MS brain samples using a multi-echo T2 relaxation MR experiment. The myelin water fraction (MWF), defined as T2 signal below 30 milliseconds divided by the total signal, was determined for various regions of interest and compared to Luxol fast blue (myelin stain) mean optical density (OD) for each sample. MWF had a strong correlation with myelin stain [mean (range) R2 /0.67 (0.45 0.92)], validating MWF as a measure of myelin density. This quantitative technique has many practical applications for the in vivo monitoring of demyelination and remyelination in a variety of disorders of myelin. Multiple Sclerosis 2006; 12: 747 753. http://msj.sagepub.com Key words: brain; histopathology; MRI; multiple sclerosis; myelin; T2 relaxation

Introduction The ability to study myelin in vivo has tremendous consequences for the understanding of neurological diseases, such as multiple sclerosis (MS). Myelin is composed of non-aqueous constituents (lipids and proteins) and water, both of which contain protons. Magnetic resonance imaging (MRI) creates images based on differences in the physical properties of protons in different tissue environments, such as those found in the central nervous system (CNS). Due to current technological limitations, conventional MRI cannot directly assess myelin because the proton signal from non-aqueous myelin decays to zero in approximately 50 microseconds [1]. Ultrashort TE MRI techniques [2], can access signals with T2s in this range, but they are not specific for myelin. However, the MR signal

from water in brain tissue with decay times longer than 10 milliseconds is accessible by MRI. Therefore, myelin structure and pathology can be probed by studying the MRI-visible water associated with myelin. Multi-echo T2 relaxation is an MRI technique that can probe specific water environments, whereby the signal from mobile protons is separated into different water pools based on their T2 relaxation time. Previous in vivo work [1,3,4], has shown that mobile protons in healthy human brain can be separated into three compartments: (1) a very long T2 component of approximately 2 seconds assigned to cerebrospinal fluid; (2) an intermediate component of the order of 80 milliseconds from intra- and extra-cellular water; and (3) a short T2 component of approximately 20 milliseconds arising from myelin water trapped tightly between the myelin bilayers. The latter component has been

a

Deceased. Department of Physics and Astronomy, University of British Columbia, Canada 2 Department of Radiology, University of British Columbia, Canada 3 Department of Pathology and Laboratory Medicine (Neuropathology), University of British Columbia, Canada 4 Department of Medicine, University of British Columbia, Canada 5 University of British Columbia Hospital, Canada 6 Vancouver General Hospital, Vancouver, BC, Canada Address for correspondence: C Laule, Magnetic Resonance Imaging, University of British Columbia Hospital, 2211 Wesbrook Mall, Vancouver, BC, V6T 2B5, Canada. E-mail: [email protected] Received 16 May 2005; accepted 27 February 2006 1

– 2006 SAGE Publications

10.1177/1352458506070928 Downloaded from http://msj.sagepub.com at TORONTO UNIVERSITY on February 9, 2007 © 2006 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

748

C Laule et al.

shown to be variably decreased in MS lesions [4 6], as well as diffusely reduced in the normal appearing white matter (NAWM) when compared to healthy controls [6]. Myelin water imaging has the potential to quantitatively define the role of myelin specific pathology and, thereby, to further our understanding of MS pathogenesis and the impact of therapeutic interventions on remyelination. However, for any MR technique claiming to be an in vivo marker for a specific tissue component, pathological validation is crucial. Only a handful of studies have quantitatively investigated the correlation between MR derived measures and histopathology [7 13]. T2 measurements from animal models have shown the T2 distribution to be multi-component and pathological comparison suggested that the short T2 component is specific for myelin content in brain [7,8]. Studies of injured peripheral nerve from the rat have also provided evidence for the multi-compartmental nature of the T2 decay curve in nerve tissue and have shown that the size of the short T2 component correlated well with a quantitative histological assessment for myelin, in particular, the processes of myelin loss through nerve injury and remyelination [10]. Studies in human brain have shown that postmortem formalin-fixed MRI is comparable to in vivo imaging [14]. Fortuitously, there is little change of the myelin water signal post-mortem, both shortly after death in situ and upon tissue fixation with formalin [15]. The T2 distribution from the formalin-fixed brain is qualitatively shaped, similar to that from brain in vivo , although quantitatively, the T2 times are shifted to shorter times. A good qualitative correspondence was observed between myelin water in the formalin-fixed brain and the anatomical distribution of myelin [15], as indicated by Luxol fast blue (LFB), a stain originally introduced by Klu ¨ ver and Barrera in 1953 [16], widely believed to stain phospholipid components of myelin [17 19], which is widely used for the histological demonstration of myelin. To provide further evidence that the myelin water signal is an accurate marker of myelin content, we conducted a post-mortem validation study. The goal of this study was to determine if the short T2 signal quantitatively correlated with a histopathologic measure of myelin density, and thus, could serve as an MRI measure of myelin content.

female, five male; mean age /66 years (range: 41 80 years); mean disease duration/23 years (range: 1 49 years), 12 secondary progressive (SP) MS and one acute MS). One of the 12 SP patients had also undergone MR examination of the brain in situ immediately post-mortem and, additionally, following formalin fixation. All brains were fixed for a minimum of two months in 10% formalin. The brains of six patients were hemisected along the midsagittal plane and one half of each brain was placed in a gel (30% egg albumin and 0.5% gelatin cross-linked with 1.5% glutaraldehyde [20]) in a rectangular semi-transparent polyethylene container, which allowed direct visualization of the medial surface of the specimen. Conventional spin echo dual echo images were acquired (TE /20/60 milliseconds, TR/2000 milliseconds) to determine locations of interest. Ten planes from the six hemispheres were selected for further examination. The brains of the remaining seven patients were sectioned into approximately 1-cm thick slices along the coronal or transverse planes and placed into plastic containers filled with formalin. Conventional spin echo dual echo images were acquired (TE /20/60 milliseconds, TR/2000 milliseconds) to determine locations of interest and 15 1-cm thick samples were selected for further examination. These 15 samples were then individually repackaged into formalin-filled plastic containers with the surface of interest placed against the bottom of the container.

MRI experiments MR images were acquired in a transmit/receive head coil using a 1.5 Tesla scanner (GE Signa EchoSpeed operating at the 5.7 software level). MRI experiments initially included a T1-weighted localizer (TE /8 milliseconds, TR /300 milliseconds), followed by a conventional spin echo dual echo (TE /20/60 milliseconds, TR/2000 milliseconds). Single slice T2 relaxation data were acquired using a 32 echo sequence, consisting of a 908 slice selective pulse, followed by 32 rectangular composite 1808 pulses, flanked by slice-selective crusher gradient pulses for the elimination of signal from outside of the selected slice (TR /3000 milliseconds, TE /10 milliseconds, BW /31 kHz, thickness/3 mm, matrix /256 /128 or 256/256, averages/4 or 8) [4]. The field of view was adjusted according to the size of the sample.

Methods Tissue selection and preparation

Tissue embedding and staining

Twenty-five brain samples were examined from 13 patients with pathologically proven MS (eight

Brain samples were prepared for tissue sectioning in two ways. For the 10 planes from the six brains

Multiple Sclerosis 2006; 12: 747 753 Downloaded from http://msj.sagepub.com at TORONTO UNIVERSITY on February 9, 2007 © 2006 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

http://msj.sagepub.com

Pathological validation of Myelin Water Imaging in MS embedded in gel, the transverse imaging planes were displayed on a 1:1 scale sagittal image of the mesial surface of the hemisphere. These planes were the central plane of the T2 relaxation slice scanned. Selected imaging planes and the outline of the mesial surface of the brain were traced from the sagittal image onto a transparency. This was superimposed on the mesial surface of the specimen that was cut along these planes. The embedding gel was then removed and the resulting brain slices were 1 cm thick, one surface of which correlated with the T2 relaxation experiment. For the remaining 15 samples, the surface which correlated with the MRI data was simply that which was placed against the bottom of the container. Gross brain slices were photographed, dehydrated, and paraffin-embedded. Sections (10-mm thick) were cut from the surface of the paraffin block using a sledge microtome and stained with LFB for myelin. All slides were then scanned using a back-lit scanner and stored electronically as uncompressed tiff images. We investigated the histological variation across the 3 mm of five samples by comparing the average regions of interest (ROI) LFB OD from five 10 mm thick histology slices spaced evenly across the 3 mm to the LFB OD obtained when ROIs were only measured on the middle histology slice. There was negligible difference; therefore, the pathology slides corresponding to the center of the MRI slice were used for subsequent comparisons.

Calculations and statistical analysis ROIs in white matter, grey matter and MS lesions were outlined on the TE /10 milliseconds MR image. The decay curves were assumed to be multi-exponential in nature. The following is a general equation, which can be used to describe the measured multi-exponential relaxation signal, yi [21]:

yi

M X

sj exp(ti =T2j ) i1; 2; . . . n

j1

where ti are the measured times, M /120 logarithmically spaced T2 times within the range of 15 milliseconds to 2 seconds, n is the total number of data points and sj is the relative amplitude for each partitioned T2 time. A non-negative least squares (NNLS) algorithm was used to minimize both x2 and an energy constraint that smoothes the T2 distribution, sj (T2j ), providing better, consistent fits in the presence of noise [21 23]. The following expression was minimized: http://msj.sagepub.com

x2 m

M X

s2j ;

749

m]0

j1

The larger the m parameter, the more the routine smoothes the T2 distribution at the cost of misfit. For the case of m/0, x2min would result. Regularized smooth T2 distributions were created by minimizing the above expression with the energy constraint of 1:02x2min 5x2 51:025x2min : Myelin water fraction (MWF) was defined as the signal with T2 relaxation times below 30 milliseconds relative to the total signal in the T2 distribution. The short T2 component range of 0 30 milliseconds was smaller than that of in vivo studies (0 50 milliseconds) as formalin fixation caused T2 relaxation times to become shorter. Myelin water maps were created by displaying the MWF at each pixel in the image. The digital histopathology images (LFB) were then registered to the TE /10 milliseconds MR image using a 10-point registration algorithm available in Image Pro Plus 4.5 (Media Cybernetics, Silver Spring, MD). The registration process involved manually placing 10 anchor points on a reference image (TE /10 milliseconds) and 10 location-matched anchor points on the object image (LFB histopathology). The object image (LFB histopathology) was then spatially transformed by linear translation, rotation and scaling to match the TE/ 10 milliseconds MR image. ROIs from the T2 relaxation experiment were then mapped onto the histopathology images. The mean optical density (OD) of the LFB stain in the ROI was determined using Image Pro Plus 4.5 (Media Cybernetics). Correlations were investigated between MWF and LFB OD for each tissue sample. All statistical analysis was carried out using SPSS 12.0.

Results Preservation of MWF post-mortem and upon formalin fixation The T2 distribution of a NAWM ROI from an MS patient sampled immediately post-mortem in situ is shown in Figure 1, in comparison to the sample after fixation with formalin, as well as to an in vivo location-matched ROI from another MS patient. Each distribution contained both a short (myelin water) and an intermediate component. The in vivo plot was consistent with previously published work, where the myelin associated water appeared at /20 milliseconds and the intra/extra-cellular peak at /80 milliseconds [24]. The intermediate component from the post-mortem in situ sample was wider and its mean T2 increased to 112 milliseconds, when compared to in vivo , while the Multiple Sclerosis 2006; 12: 747 753

Downloaded from http://msj.sagepub.com at TORONTO UNIVERSITY on February 9, 2007 © 2006 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

750

C Laule et al. Correlation of MWF and myelin staining

Figure 1 In vivo, in situ and formalin-fixed brain T2 distributions. T2 distribution of a normal appearing white matter ROI from the in situ sample (dashed line) is shown in comparison to the sample after fixation with formalin (grey line), as well as to in vivo location-matched ROI from an MS patient (solid line). All three distributions have a similar shape. However, the intermediate component is shifted to longer times for the in situ sample and shorter times for the in-formalin sample when compared to in vivo.

short T2 component remained relatively unchanged. Data from the formalin-fixed sample showed the intermediate peak narrowing and mean relaxation time shortened to 63 milliseconds, while the short T2 component again remained relatively unchanged compared to both in vivo and in situ results. Maps of the MWF (myelin water maps) from the post-mortem in situ sample demonstrated a good qualitative correlation with myelin water maps of the same brain after formalin fixation, as shown in Figure 2.

All 25 myelin water maps exhibited a strong qualitative resemblance to the LFB staining for myelin, as demonstrated in Figure 3. The total number of ROIs varied for different samples, ranging from 23 to 45 per slice. MWF correlated strongly with LFB OD for each sample [mean R2 / 0.67 (range: 0.45 0.92)], P B/0.0001 for all samples. Examining white matter ROIs only, a correlation was still found between MWF and LFB OD (mean R2 /0.29, P B/0.0001). Figure 4 shows strong correlations between MWF and OD of LFB for various ROIs from white matter, grey matter and lesions in three different samples. Table 1 summarizes the correlation coefficients between MWF and LFB OD for the 25 samples.

Discussion Myelin water imaging promises to offer new insight into demyelinating diseases, such as MS. Validation of techniques claiming to measure myelin content is crucial if we are to further understand the pathophysiology of MS based on in vivo MR measures. The current study provides strong and direct evidence that myelin water imaging based on multi-echo T2 relaxation is indeed a valid marker for CNS myelin density. Staining uniformity A potential technical concern when comparing to a quantitative analysis of histopathological staining

Figure 2 In situ and formalin-fixed myelin water maps. In situ (a) proton density MRI and (b) myelin water map of an MS patient, (c) corresponding formalin-fixed brain myelin water map for the right hemisphere.



Pathological validation of Myelin Water Imaging in MS

751

may have limited the quantitative correlation between MWF and OD. We minimized this effect by choosing a histology slice from the middle of the MRI slice. The effect of magnetization exchange on MWF

is controlling for staining uniformity and consistency, both across a single slide and between different samples. By visual inspection, staining intensity appeared uniform within a slide in NAWM areas. To avoid error due to different staining intensities between samples, we chose to limit the comparison of OD and MWF to ROIs within individual slices only.

Our in vitro T2 measurements were carried out in the magnet at the ambient temperature of approximately 208, while in vivo measurements are acquired at 378C. Physical exchange of water molecules between the myelin and intra/extracellular water compartments could influence the relative intensities of the two T2 peaks. If the exchange between myelin water and intra/extracellular water were much faster at 378C than at 248C, then the measured MWFs would be smaller at the higher temperature. This scenario is unlikely since the in vivo exchange rate between myelin water and intra/extracellular water has been measured [25] to be about 200 milliseconds much longer than the observed T2 times of 20 and 80 milliseconds for the two water pools. Thus, the post-mortem MWF values are a close approximation of what would be seen in living tissues.

The effect of partial volume on MWF

T2 relaxation from in vivo, post-mortem in situ and formalin-fixed brain

We compared 3 mm thick MRI slices with 10 mm thick histology slides. Thus, partial volume effects

Examination of post-mortem in situ data (Figure 1) showed both a short and an intermediate T2

Figure 3 Qualitative correlation between MRI and histopathology. Good qualitative correlation between (a) TE/10 milliseconds, (b) myelin water image, (c) LFB.

Figure 4 Quantitative correlation between MWF and LFB OD. Three samples which had strong correlations between MWF and OD of LFB (a) R2 /0.79, (b) R2 /0.69, (c) R2 /0.78, for various regions of interest made up of white matter (black circles), grey matter (grey circles) and lesion (open circles).



752

C Laule et al.

Table 1 Compilation of R2 values for the correlation between myelin water fraction and LFB optical density for the 25 fixed brains (PB/0.0001 for all cases) Sample

Myelin correlation

Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 Case 7 Case 8 Case 9 Case 10 Case 11 Case 12 Case 13 Case 14 Case 15 Case 16 Case 17 Case 18 Case 19 Case 20 Case 21 Case 22 Case 23 Case 24 Case 25 Mean (SD)

0.85 0.72 0.48 0.88 0.78 0.68 0.57 0.63 0.77 0.92 0.64 0.76 0.72 0.56 0.79 0.49 0.45 0.66 0.49 0.61 0.66 0.62 0.57 0.65 0.69 0.67(0.13)

component, in agreement with in vivo studies [4,24]. However, an increase in mean relaxation time of the intermediate signal (assigned to intra/ extra-cellular water) was observed when compared to in vivo measurements. A lengthening of the intermediate T2 relaxation time upon death has also been reported in several other studies involving multi-echo T2 measurements of animal models [1,3]. The increase in relaxation time of the intra/ extra-cellular water suggests an increase in the mobility of the water molecules in this environment, possibly due to autolysis of cells, which begins at the time of death. This is supported by one study which found autolytic changes to cause an increase in both T1 and T2 relaxation times in human brain tissue [26]. After tissue fixation with formalin, T2 relaxation measurements continued to exhibit both a short and an intermediate component. MWF values were comparable to in vivo measurements [24]. However, a decrease in the mean relaxation time of the intermediate signal was observed when compared to in vivo measurements. As water T2 times have been shown to be strongly influenced by compartmentalization of cellular structure [27], it is expected that the relaxation time of formalin-fixed tissue may shorten, since formalin acts by establishing bonds between protein groups to increase cellular stability, thereby ‘fixing’ the sample. This decrease of T2

in formalin-fixed tissue has been observed by several other groups [14,28,29]. Correlation of MWF with histology Comparison of MWF and OD of LFB staining of myelin for individual samples demonstrated a very strong average correlation, indicating that myelin water imaging is indeed a valid MRI measure of myelin content in CNS tissue. This correlation remained when examining white matter ROIs only, suggesting the observed relationship is not a bimodal effect due to the separation of white and grey matter values. MWF is a measure of total myelin content. While it is sensitive to myelin loss, the technique is not yet able to distinguish between myelin decreases accompanying axonal loss versus primary demyelination or remyelination.

Conclusion Myelin water imaging provides in vivo tissue specific information that promises to have important consequences for understanding demyelinating disorders, such as MS. This study validates MWF as a measure of myelin density and supports the use of myelin water imaging to study myelin pathology and the role of demyelination and remyelination in MS.

Acknowledgements The authors would like to thank Thor Bjarnason, Dr Irene Vavasour and Dr Kevin Rowan for helpful discussions. In addition, we would like to sincerely thank all of the MS patients and their families, as well as the MS Society of Canada.

References 1. Stewart WA, MacKay AL, Whittall KP, Moore GR, Paty DW. Spin-spin relaxation in experimental allergic encephalomyelitis. Analysis of CPMG data using a nonlinear least squares method and linear inverse theory. Magn Reson Med 1993; 29: 767 75. 2. Robson MD, Gatehouse PD, Bydder M, Bydder GM. Magnetic resonance: an introduction to ultrashort TE (UTE) imaging. J Comp Assist Tomogr 2003; 27: 825 46. 3. Menon RS, Allen PS. Application of continuous relaxation time distributions to the fitting of data from model systems and excised tissue. Magn Reson Med 1991; 20: 214 27. 4. 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 77.



Pathological validation of Myelin Water Imaging in MS 5. Vavasour IM, Whittall KP, MacKay AL, Li DK, Vorobeychik G, Paty DW. A comparison between magnetization transfer ratios and myelin water percentages in normals and multiple sclerosis patients. Magn Reson Med 1998; 40: 763 68. 6. Laule C, Vavasour IM, Moore GRW, Oger J, Li DKB, Paty DW et al . Water content and myelin water fraction in multiple sclerosis: a T2 relaxation study. J Neurol 2004; 251: 284 93. 7. Gareau PJ, Rutt BK, Bowen CV, Karlik SJ, Mitchell JR. In vivo measurements of multi-component T2 relaxation behaviour in guinea pig brain. Magn Reson Imaging 1999; 17: 1319 25. 8. Gareau PJ, Rutt BK, Karlik SJ, Mitchell JR. Magnetization transfer and multicomponent T2 relaxation measurements with histopathologic correlation in an experimental model of MS. J Magn Reson Imaging 2000; 11: 586 95. 9. van Waesberghe JH, Kamphorst W, De Groot CJ, van Walderveen MA, Castelijns JA, Ravid R et al . Axonal loss in multiple sclerosis lesions: magnetic resonance imaging insights into substrates of disability. Ann Neurol 1999; 46: 747 54. 10. Webb S, Munro CA, Midha R, Stanisz GJ. Is multicomponent T2 a good measure of myelin content in peripheral nerve? Magn Reson Med 2003; 49: 638 45. 11. Stanisz GJ, Webb S, Munro CA, Pun T, Midha R. MR properties of excised neural tissue following experimentally induced neuroinflammation. Magn Reson Med 2004; 51: 473 79. 12. Mottershead JP, Schmierer K, Clemence M, Thornton JS, Scaravilli F, Barker GJ et al . High field MRI correlates of myelin content and axonal density in multiple sclerosis a post-mortem study of the spinal cord. J Neurol 2003; 250: 1293 301. 13. Moore GRW. MRI clinical correlations: more than inflammation alone what can MRI contribute to improve the understanding of pathological processes in MS. J Neurol Sci 2003; 206: 175 79. 14. 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. 15. Moore GRW, Leung E, MacKay AL, Vavasour IM, Whittall KP, Cover KS et al . A pathology-MRI study of


16. 17. 18. 19. 20. 21. 22. 23. 24.

25.

26. 27.

28.

29.

753

the short-T2 component in formalin-fixed multiple sclerosis brain. Neurology 2000; 55: 1506 10. ¨ ver H, Barrera E. A method for the combined Klu staining of cells and fibres in the nervous system. J Neuropathol Exp Neurol 1953; 12: 400 403. Segarra J. Histological and histochemical staining methods. In Teduchi C ed. Neuropathology methods and diagnosis . Little, Brown and Company, 1970: 233 69. Salthouse T. Luxol fast blue ARN: a new solvent azo dye with improved staining qualities for myelin and phospholipids. Stain Technol 1962; 37: 313 16. Lycette R, Danforth J, Koppel J, Olwin J. The binding of Luxol fast blue ARN by various biological lipids. Stain Technol 1970; 45: 155 60. Goodman LA, Model PG. Superinnervation enhances the dendritic branching pattern of the Mauthner cell in the developing axolotl. J Neurosci 1988; 8: 776 91. Whittall KP, MacKay AL. Quantitative interpretation of NMR relaxation data. J Magn Reson 1989; 84: 64 71. Lawson CL, Hanson RJ. Solving least squares problems . Prentice-Hall, 1974. Fenrich F, Beaulieu C, Allen PS. Relaxation times and microstructures. NMR Biomed 2001; 14: 133 39. Whittall KP, MacKay AL, Graeb DA, Nugent RA, Li DK, Paty DW. In vivo measurement of T2 distributions and water contents in normal human brain. Magn Reson Med 1997; 37: 34 43. Vavasour IM, Whittall KP, Li DK, MacKay AL. Different magnetization transfer effects exhibited by the short and long T(2) components in human brain. Magn Reson Med 2000; 44: 860 66. Grodd W, Schmitt WG. Proton relaxation characteristics of human and animal tissue in vitro . Changes during autolysis and fixation. ROFO 1983; 139: 233 40. Araujo C, MacKay AL, Whittall KP, Hailey JRT. A diffusion model for spin spin relaxation of compartmentalized water in wood. J Magn Reson 1993; 101: 248 61. Kamman RL, Go KG, Stomp GP, Hulstaert CE, Berendsen HJ. Changes of relaxation times T1 and T2 in rat tissues after biopsy and fixation. Magn Reson Imaging 1985; 3: 245 50. Shuter B, Tofts PS, Pope JM. The effect of sacrifice on image signal, T1, T2, and T*2 in liver, kidney, and brain of the Wistar rat. Magn Reson Imaging 1995; 13: 563 74.
