Insights into brain microstructure from the T2 distribution

Magnetic Resonance Im

Insights into brain microstructure from the T2 distribution

Alex MacKaya,b,4, Cornelia Laulea, Irene Vavasoura, Thorarin Bjarnasonb,1,

Shannon Kolindb, Burkhard M7dlercaDepartment of Radiology, University of British Columbia, Vancouver, BC, Canada

bDepartment of Physics and Astronomy, University of British Columbia, Vancouver, BC, CanadacPhilips Medical Systems, Vancouver, British Columbia, BC, Canada

Received 2 December 2005; accepted 2 December 2005

Abstract

T2 weighting is particularly sensitive, but notoriously unspecific, to a wide range of brain pathologies. However, careful measurement and

analysis of the T2 decay curve from brain tissue promise to provide much improved pathological specificity. In vivo T2 measurement requires

accurate 180 pulses and appropriate manipulation of stimulated echoes; the most common approach is to acquire multiple echoes from a

single slice. The T2 distribution, a plot of component amplitude as a function of T2, can be estimated using an algorithm capable of fitting a

multi-exponential T2 decay with no a priori assumptions about the number of exponential components. T2 distributions from normal brain

show peaks from myelin water, intra/extracellular water and cerebral spinal fluid; they can be used to provide estimates of total water content

(total area under the T2 distribution) and myelin water fraction (MWF, fractional area under the myelin water peak), a measure believed to be

related to myelin content. Experiments on bovine brain suggest that magnetization exchange between water pools plays a minor role in the T2

distribution.

Different white matter structures have different MWFs. In normal white matter (NWM), MWF is not correlated with the magnetization

transfer ratio (MTR) or the diffusion tensor fractional anisotropy (FA); hence it provides unique information about brain microstructure.

Normal-appearing white matter (NAWM) in multiple sclerosis (MS) brain possesses a higher water content and lower MWF than controls,

consistent with histopathological findings. Multiple sclerosis lesions demonstrate great heterogeneity in MWF, presumably due to varying

myelin contents of these focal regions of pathology. Subjects with schizophrenia were found to have significantly reduced MWF in the minor

forceps and genu of the corpus callosum when compared to controls, suggesting that reduced frontal lobe myelination plays a role in

schizophrenia. In normal controls, frontal lobe myelination was positively correlated with both age and education; this result was not

observed in subjects with schizophrenia.

A strong correlation between MWF and the optical density from the luxol fast blue histological stain for myelin was observed in formalin-

fixed brain, supporting the use of the MWF as an in vivo myelin marker.

D 2006 Elsevier Inc. All rights reserved.

Keywords: Brain microstructure; Myelin water fraction; T2 Distribution

1. Introduction

T2 plays a role in almost every aspect of medical

magnetic resonance. From the early days, it was clear that

clinical T2-weighted images were exquisitely sensitive to

brain pathology, giving rise to much optimism that T2

0730-725X/$ – see front matter D 2006 Elsevier Inc. All rights reserved.

doi:10.1016/j.mri.2005.12.037

4 Corresponding author. Department of Physics and Astronomy,

University of British Columbia, Vancouver, Canada V6T 1Z1.

E-mail address: [email protected] (A. MacKay).1 Currently at the Department of Electrical Engineering, University of

Calgary, Calgary, Alberta, Canada.

relaxation would lead to pathological specificity. However,

in spite of a great deal of research, pathological specificity

from T2 was not realised [1] for a number of reasons; the

most important being that conventional MRI techniques like

proton density, T1 and T2 weighting are qualitative in a

clinical setting.

The most specific MR techniques are tuned solely to

study the spin species of interest. For example, in 1H MR

spectroscopy the signals from individual brain metabolites

(b0.1% of the total signal) can be measured by using water

suppression and taking advantage of the chemical shift. In

fMRI, the signal from capillary blood (approximately 4% of

aging 24 (2006) 515–525

A. MacKay et al. / Magnetic Resonance Imaging 24 (2006) 515–525516

the grey matter volume) can be separated from that of other

tissue water by virtue of changes in blood T2* caused by

oxygenation changes.

The total 1H magnetic resonance signal from brain

includes contributions from protons in water as well as

nonaqueous protons in molecules such as lipids, proteins

and nucleic acids. The signal of water in tissue has T2 times

longer than 10 ms, while the nonaqueous proton signal

decays to zero in less than 100 As due to large unaveraged

dipolar couplings between adjacent protons [2–4]. Conse-

quently, it is relatively easy to measure the MR signal from

water in brain with no contamination from the fast decaying

nonaqueous tissue signal.

Pure water has a T2 of about 3 s. Water in brain

undergoes much faster T2 relaxation rates; the degree of T2

shortening depending on interactions between water and

nonaqueous tissue. A homogeneous volume, such as a glass

of pure water, gives rise to monoexponential T2 relaxation.

However, brain tissue is inhomogeneous, not only at the

grey/white matter spatial level of 1 to 100 mm, but also at

the cellular spatial level on the order of 1 to 10 Am.

Furthermore, due to Brownian motion, water moves several

micrometers each millisecond due to self-diffusion.

Pathological changes observed in neurodegenerative

diseases include edema (increased intra- or extracellular

water), blood–brain barrier breakdown (tight junction

leakage), inflammation (proliferation of inflammatory cells),

demyelination (breakdown of the myelin sheath), gliosis

(proliferation of glial cells) and axonal loss (breakdown of

the axon). The extent to which these pathologies can be

distinguished by MRI depends upon whether they have a

unique impact on the proton NMR signal; if these

pathological changes affect the organisation of nonaqueous

molecules in cellular structures, water T2 relaxation should

also be affected. It is therefore very challenging to relate T2

relaxation directly to brain microstructure. A model for T2

relaxation in brain must take into account both tissue

architecture and the extent of water diffusion over the

relevant timescales.

This review discusses how in vivo T2 relaxation can

provide specific information about brain anatomy and

pathology. There is particular emphasis on T2 relaxation

pulse sequences and analysis techniques, on the interpreta-

tion of T2 components as water reservoirs, as well as

comparisons with results from other nonconventional MR

techniques. Results are presented from normal volunteers

and people with multiple sclerosis (MS) and schizophrenia.

Fig. 1. T2 decay curves for various white (black line) and grey (grey line)

matter structures, as well as CSF (dashed line).

2. In vivo T2 measurement

The first step in a T2 study of brain is the acquisition of

a high-fidelity T2 decay curve. The most common

approach is to collect multiple echoes in a single MR

sequence. Poon and Henkelman [5] developed a single

slice multi-echo pulse sequence, employing composite

radiofrequency pulses and gradient crusher pulses, which

produces very good decay curves from brain. Key issues

for accurate in vivo T2 decay measurement are maintaining

perfect 180 pulses in the presence of inhomogeneous B1

and B0 fields, and elimination of all contributions from

stimulated echoes accruing from signal excited outside the

selected slice. For quantitative analysis, T2 decay curves

must have high signal-to-noise ratios with the minimum

acceptable noise standard deviation being about 1% of the

signal strength at the shortest echo time [6]. The echo

spacing should ideally be as short as feasible and the echo

train length should be such that the last echoes report only

noise. For in vivo human brain studies, the echo spacing

should be 10 ms or less, and the echo train should exceed

1 s to measure the shortest T2 components and to be

sensitive to T2 times on the 400-ms timescale, respectively.

Unfortunately, the number of echoes acquired is often

limited by considerations of power deposition and MR

scanner pulse programmer restrictions. In the presence of

microscopic magnetic field gradients, such as those caused

by ferromagnetic or paramagnetic constituents, the shape

of the T2 decay curve depends upon the echo spacing [7].

Most of our in vivo T2 decay curve results have been

acquired from 32 echoes with a 10-ms echo spacing [8–

10]. More recently, we have collected 32 echoes at 10-ms

spacing plus a further 16 echoes with an echo spacing of

50 ms [11].

3. T2 decay curve analysis

Fig. 1 shows several T2 decay curves from normal

human brain. Different white and grey matter structures

decay at dissimilar rates. Furthermore, different brain

pathologies have been observed to give rise to unique

decay curves. The semi-logarithmic T2 plots are not well fit

Fig. 3. A schematic representation of the four-pool model.

A. MacKay et al. / Magnetic Resonance Imaging 24 (2006) 515–525 517

by a single straight line indicating that the volumes of

interest contain more than one distinguishable water

reservoir, each with its own T2 time. Fitting such curves

to extract meaningful information in the form of decay

constants and corresponding signal amplitudes becomes

increasingly difficult in the presence of multiple compart-

ments and noise. Since the number of distinguishable water

environments is generally not known a priori, the fitting

algorithm must be capable of estimating the number of

exponentials. Furthermore, many believe that brain T2

decay curves are best characterised not by a fixed number

of discrete T2 components, but rather by a continuous T2

distribution [12].

The regularised nonnegative least squares (NNLS)

algorithm produces robust analyses of multi-component T2

decay curves in terms of a smooth T2 distribution [13,14].

Nonnegativity is a reasonable physical constraint since there

cannot be a negative number of protons contributing to the

MR signal. One can use the following equation to describe

the T2 decay curve, y(ti):

y tið Þ ¼XM

j¼1

s T2j

� �e�ti=T2j þ Ei; i ¼ 1; 2 ; : : : ; N ð1Þ

where t i are the N measurement times, T2j are M

logarithmically spaced T2 times constituting the T2 partition,

s(T2j) is the T2 distribution and Ei is the noise contributing

to the ith decay curve point. The regularised NNLS

algorithm fits a function defined by Eq. (1) to the measured

data by minimizing N:

N ¼ v2 þ lXM

j¼1

s T2j

� �2; lz0: ð2Þ

The first term in N, v2, is chi-squared, and the second

term is the regulariser which, in this case, is the energy of

the T2 distribution. The larger the l parameter the more

Fig. 2. An example of a T2 distribution of human white matter in vivo

with labeled regions corresponding to compartments of white matter. v2

was kept within 2–2.5% of its nominal value.

the algorithm smoothes the T2 distribution at the cost of

v2 misfit: the case of l=0 provides the vmin2 result [14] for

which s(T2j) consists of discrete T2 spikes. One can

regularise the fitting process by placing a constraint on

the v2 value in accordance to vmin2. Regularisation results

in more robust fits in the presence of noise [14] and

provides smooth solutions that better represent the distri-

bution of relaxation times expected from tissue microstruc-

ture [12]. An example of a T2 distribution from normal

human brain in vivo, using the regularising constraint of

1.02vmin2 Vv2V1.025vmin

2 , is shown in Fig. 2.

4. Interpretation of brain T2 distribution

Fig. 2 shows three T2 components in normal human

brain. Several studies have dealt with their assignments

[3,4,8]. The largest component, centred on a T2 of

approximately 70–90 ms, has been assigned to intra/

extracellular water; the smaller component found between

10 and 50 ms is assigned to water trapped between the

bilayers of the myelin sheath; and the component at

T2N2 s is associated with cerebrospinal fluid (CSF).

T2 decay curves from peripheral nerve have been found

to be composed of three T2 components at times of

Fig. 4. Simulated fits to the cross-relaxation data collected at 248C.


roughly 20, 80 and 250 ms [15–18]; these peaks have

been assigned to myelin water, extracellular water and

axonal water, respectively. Although the microscopic

structure of peripheral nerve differs from that of white

matter, it could be anticipated that these three peaks,

corresponding to myelin water, extracellular water and

axonal water, should also be distinguishable in white

matter. Current limitations on the signal-to-noise ratio

available for in vivo T2 relaxation studies are likely

responsible for the difficulty involved in separating the

intra/extracellular water peak. There have been indications

of a prolonged T2 component in heavily myelinated white

matter structures [9] and in MS lesions [19]. By taking

advantage of the above assignments of the T2 peaks, we

can learn about brain microstructure from the shape of the

T2 distribution.

The T2 distribution provides two specific measures

related to central nervous system (CNS) structure: water

content and myelin water fraction (MWF). Since MR is

sensitive to all of the water in CNS tissue, a summation

over all T2 components provides a relative water content

measurement. This measurement can be made absolute,

in gH2O/ml brain, if a water standard is available and

appropriate corrections are made for T1 weighting and B1

inhomogeneity [20]. Local brain water content is increased

in several pathologies, in particular in the presence of

edema. If water trapped in the myelin bilayer is

proportional to myelin content, the shorter T2 peak is a

potential candidate for in vivo measurement of myelin.

The MWF is defined as the amount of signal between 0

and 50 ms relative to the total signal in the T2

distribution. Myelin plays a key role in conduction of

nerve pulses and many disease processes involve loss of

myelin, resulting in a clinical deficit. Therefore an in vivo

myelin marker would be a very useful tool for under-

standing disease mechanisms.

Table 1

Cross-relaxation simulation parameters

Parameter 248C 378C

T1M 343 (83) 830 (52)

T1MW 343 (83) 830 (52)

T1IE 943 (70) 830 (52)

T1NW 943 (70) 830 (52)

Pf (%) 82.4 (3.3) 81.3 (1.9)

MWF (%) 10.9 (3.2) 11.8 (4.6)

TcrD 2064�852

+3285 1021�253+465

TcrMW 227�54

+76 158�50+75

TcrIE 402�88

+119 170�42+62

Standard deviations are shown in parentheses, and error estimates follow the

+ and � signs. Error estimates were determined by allowing the variable of

interest to vary during the simulation in the b+Q or b�Q direction, whilekeeping the other parameters constant, so that the change in the sum of

squares was 5%. All units are in milliseconds unless otherwise stated.

5. Magnetization exchange in white matter

5.1. A four-pool model for bovine brain

To better understand magnetization exchange in white

matter, our group [21] examined fresh bovine brain in

vitro using an NMR spectrometer capable of measuring

the proton NMR signals from both the nonaqueous tissue

and the water. Assuming the nonaqueous tissue is made

up of two pools, each associated with a distinguishable

water reservoir, a biologically viable mathematical de-

scription of white matter requires four proton pools

[22,23] as depicted in Fig. 3. This four-pool model is

made up of nonaqueous protons in myelin (M), non-

myelin (NM), and water protons in myelin (MW) and

intra/extracellular (IE) spaces. One can describe the time

dependence of magnetization using the Bloch equations

assuming that each proton reservoir achieves internal

equilibrium of magnetization rapidly compared to the

rates of magnetization exchange between proton reser-

voirs [24,25]:

ddt

MM ¼ � k12MM � MM �MM lð ÞTM1

þ k21MMW

ddt

MMW ¼ � k21MMW � MMW �MMW lð ÞTMW1

� k23MMW þ k12MM þ k32MIE

ddt

MIE ¼ � k32MIE � MIE �MIE lð ÞTIE1

� k34MIE

þ k23MMW þ k43MNM

ddt

MNM ¼ � k43MNM � MNM �MNM lð ÞTNM1

þ k34MIE

ð3Þ

where the M’s and M(l)’s denote the time-dependent and

equilibrium magnetizations of the four proton pools and the

k’s denote the exchange rates between the pools. The ex-

change rates can be characterized by three cross-relaxation

times [25]:

1TMWcr

¼ PM

PM þ PMW k12 ¼ PMW

PM þ PMW k21

1TDcr

¼ PMW

PMW þ PIE k23 ¼ PIE

PMW þ PIE k32

1T IEcr

¼ PIE

PIE þ PNM k34 ¼ PNM

PIE þ PNM k43

ð4Þ

The cross-relaxation time between the myelin and myelin

water pools is characterized by TcrMW, Tcr

D is the cross-

relaxation time between myelin water and IE water due to

diffusion, and TcrIE is the cross-relaxation time between the IE

water and nonmyelin pool. The P’s are the probabilities of a

proton being in one of the four pools. The probability of a

proton being found in the various pools can be calculated as:

PMW¼MWF � Pf

PIE ¼ Pf � PMW

PM ¼ PNM ¼ 0:5 1� Pf�� ð5Þ

The mobile fraction, P f, was determined using free

induction decay experiments [21].

ig. 6. Relation between diffusion fractional anisotropy and MWF in

ormal white matter. A significant correlation is seen (R2= .6), although

ecific brain regions systematically fall outside the linear relationship.

ig. 7. Relation between MTR and MWF in normal white matter. The poor

orrelation is quantified by a low R2 (.05).

Fig. 5. (A) Geometric mean T2, (B) water content and (C) MWF from 12

normal volunteers for six grey matter (grey circles) and five white matter

(white triangles) structures.


The change in magnetization for aqueous and nonaque-

ous protons was observed using two separate cross-

relaxation experiments based on the Goldman–Shen pulse

sequence (90x–400 As–90�x–scr–measure) [26]. In the

preparation phase of the sequence (the first two 90 pulses),

the signal from nonaqueous protons was effectively nulled

with very little effect on the signal from water. After a

variable delay, scr, the three signal components were

measured. The nonaqueous signal was measured from the

free induction decay after the two 908 pulses, and in a

separate experiment, the water signal was measured with a

Carr–Purcell–Meiboom–Gill sequence. An additional T1

measurement was carried out. Data were acquired at 248Cand 378C. The evolution of the signal components as a

function of cross-relaxation time, scr, was fit to Eq. (3), and

the results are presented in Fig. 4 and Table 1. The key result

F

n

sp

of this study was that the cross-relaxation times between the

various spin reservoirs were sufficiently long such that their

effects on T2 component amplitudes should be small. The

results were in approximate agreement with a previous in

vivo measurement of cross relaxation between myelin water

and intra/extracellular water [27].

5.2. Comparison of the bovine brain four-pool model

with literature

The magnetization transfer technique has been used

extensively in studies of brain tissue integrity [28–30] and

many investigators have modeled magnetization transfer in

white matter [23,31–33]. Most, but not all [23], magneti-

zation transfer investigations employed a two-pool (free

water and nonaqueous protons) model for white matter.

Generally speaking, two-pool model magnetization transfer

parameters derived from the four-pool model results (data

not shown) are in agreement with two pool parameters

derived from experiment. This is encouraging since it

suggests that the two-pool model has roots in the four-pool

model. However, the four-pool model is more useful since,

for example, the two-pool model does not provide

information on myelin or myelin water. The nonaqueous

F

c

Fig. 8. A combined MT and T2 experiment which separates the MT effect

from myelin water and IE water. The data show that the short T2 component

(solid square) has a greater MT effect than the longer T2 component (open

circle). This difference disappears for delays longer than 200 ms.

Fig. 9. (A) Histograms of water content relative to putamen from 85 lesions

(grey) compared to contralateral NAWM (black) from 18 MS patients. (B)


proton fraction parameter F, commonly reported in two-

pool model studies, relates to all nonaqueous protons and its

use as a myelin measure is inappropriate. We believe that

future work on modeling magnetization transfer results

should involve at least four pools.

6. Results from in vivo T2 decay curve analysis in

normal subjects

Much of the MR literature suggests that all white matter,

and all grey matter, structures have similar T2 behaviour.

This is far from true; T2 decay curve measurements

demonstrate that individual brain structures have markedly

different MWFs and different geometric mean T2’s for the

intra/extracellular T2 peak [9]. It must be emphasized that in

central nervous system tissue, T2 behaviour cannot be

described by a single time!

Average IE water geometric mean T2 (mean on a

logarithmic scale), water content and MWF for different

white and grey matter structures in normal subjects are

shown in Fig. 5 [9]. The geometric mean T2 times for the

intra/extracellular water pool varied between 70 and 90 ms

with no separation between grey and white matter struc-

tures. Water content is larger for grey matter than for white

matter, in agreement with previous literature [20]. Myelin

water fraction varies by more than a factor of 2 between

different brain white matter structures and is quite low in all

grey matter areas. We believe that the variations in these

three parameters between different brain regions arise from

differing tissue microstructures.
Histograms of MWF from 189 lesions (grey) compared to contralateral
NAWM (black) from 33 MS patients. X-axis scale refers to upper limit of

each histogram bin. Relative water content for lesions and contralateral

NAWM was 0.95F0.04 and 0.90F0.03, respectively (meanFS.D.).

Relative water content for lesions was significantly different from

contralateral NAWM ( P b.0001). Mean MWF for lesions and contralateral

NAWM was 0.046F0.004 and 0.089F0.002, respectively. Myelin water

fraction for lesions was significantly different from contralateral NAWM

( P b.0001).

7. Comparison to other techniques commonly used to

explore brain pathology

7.1. Diffusion

In white matter, the diffusion coefficient, D, has been

found to be larger along neurons than perpendicular to

them [34–36], and therefore D must be handled as a

tensor, D. Many useful measures can be derived from this

tensor including fractional anisotropy, FA, which is

proportional to the ratio of the standard deviation of the

diffusion tensor to its mean principal value. When water

diffusion is isotropic, i.e., the same in all directions as in

pure water, the principal components of D are equal and

FA is zero. For water associated with myelinated nerve

fibres, diffusion is faster along the axon and FA is large.

The restriction of diffusion in the perpendicular direction

is attributed to myelinated axons, and therefore diffusion

FA should be correlated with myelin content and hence

MWF. Fig. 6 shows a significant correlation (R2=.6)

between FA and MWF in NWM, although specific brain

structures fall systematically outside the linear fit [37].

The MWF/FA relationship is no longer significant when

only white matter structures are considered. It has been

shown that FA is modulated by myelin, but is primarily

dominated by intact axonal membranes [38]. Therefore,

ig. 10. Mean (A) water content (relative to putamen) and (B) MWF of five

ifferent white matter structures from 18 MS patients (open squares)

ompared to 18 age- and gender-matched controls (closed squares). Error

ars are standard errors. For MS patients, all five structures had

gnificantly higher water content than controls ( P b.0006). Four of five

ructures had significantly less myelin water in MS patients: splenium,

posterior internal capsules, minor forceps and major forceps ( P b.05).


diffusion FA provides similar, yet complementary, infor-

mation to MWF, with each measure originating from

different microscopic structures — MWF from myelin

sheaths and diffusion FA from axons and the extraneuro-

nal spaces.

7.2. Magnetization transfer

Magnetization transfer measures the exchange of

magnetization between water and nonaqueous tissue after

the application of an off-resonance pulse selective for

nonaqueous protons and can be characterised by calculat-

ing a MTR=(Mo�Ms)/Mo, where Ms and Mo are the MR

signals with and without the off-resonance pulse. Magne-

tization transfer imaging has become very popular as the

data are relatively easy to acquire and analysis is

straightforward with little postprocessing required. The

magnetization transfer ratio has been found in many

studies to be a robust and sensitive indicator of white

matter pathology [29,39–44], but it is greatly lacking in

specificity. A decrease in MTR, which is often observed in

neurodegenerative diseases, can arise from either an

increase in water content or a decrease in nonaqueous

tissue which includes macrophages, astrocytes, glial cells,

myelin, axons and oligodendrocytes [45–47]. In diseases

that are characterized by both increases in water content

arising from edema or inflammation, and decreases in

nonaqueous tissue due to demyelination and/or axonal loss,

it is impossible to determine which pathological processes

are responsible for a decrease in MTR. Some groups

hypothesise that because myelin contributes to the non-

aqueous tissue pool, MTR is a specific marker for myelin;

however, myelin makes up only approximately 50% of

CNS white matter [48], leaving half of the nonaqueous

pool made of nonmyelin-associated tissue. If MTR were a

specific measure of myelin, a correlation should exist

between MTR and MWF, the previously described T2-

based marker for myelin. A comparison of MWF and MTR

for a variety of white matter structures in normal subjects

showed no correlation between these 2 measures (R2=.05),

as shown in Fig. 7 [49]. Similar results were demonstrated

in MS lesions [50], suggesting that MT and T2 relaxation

provide different, and therefore complementary, informa-

tion about brain tissue microstructure.

7.3. Combined MT and T2

If an off-resonance MT pulse is applied before the

multi-echo T2 relaxation measurement, the magnetization

transfer effect can be separately measured for the short T2

myelin water pool and the intermediate T2 intra/extracel-

lular water pool. The results from a combined MT/T2

experiment are plotted in Fig. 8 [27]. If the relaxation data

are collected immediately after the MT pulse, then the

myelin water MTR is approximately twice that of the intra/

extracellular water MTR. This further supports the

assignment of the short T2 component to myelin water

because close association between water and tissue, i.e.,

myelin water and myelin bilayers, should cause a large

magnetization exchange.

As the delay time between the MT pulse and the multi-

echo T2 relaxation experiment was increased, the MTRs for

the two water pools approached the same value. The likely

mechanism for the convergence of the myelin water MTR

and intra/extracellular water MTR is diffusion of water

between the two compartments during the delay time,

resulting in equalisation of the magnetization in both pools.

With the MT/T2 experiment, an in vivo analogue of the

bovine brain study discussed earlier, we are able to probe

the exchange rate of water across the myelin sheath and

estimate the lifetime of a water molecule within myelin to be

on the order of several hundred milliseconds. This has

important implications for the use of the MWF as an in vivo

myelin marker.

8. Application to diseases

8.1. Multiple sclerosis

Multiple sclerosis is an autoimmune disease character-

ised by edema, inflammation, demyelination and axonal loss

within the central nervous system [51]. People with MS

suffer from a plethora of symptoms including vision

problems, loss of balance, cognitive deficits, fatigue,

F

d

c

b

si

st


tremors and paralysis. The unpredictable nature and

drastically varying disease course from person to person

with MS make this disease particularly difficult to deal with,

both from an emotional and physical standpoint. Although

MS was first described over 100 years ago [52], the

mechanisms governing clinical evolution and disease

progression and their relationship to pathological character-

istics are still poorly understood.

Over the past two decades, magnetic resonance imaging

has become a powerful diagnostic tool for MS. Clinical MR

images of MS brain and spinal cord can easily visualise focal

areas of the disease known as lesions. Lesions can appear

and disappear, although they typically increase in number

and volume as the disease progresses. Many pharmaceutical

companies use changes in MRI measures such as burden of

disease (lesion load) as primary outcome measures, and

while conventional MRI provides some information about

the disease process of MS, little correlation has been found

between total lesion volume and measures of clinical

disability [53–56]. This dissociation suggests that there is

ongoing active disease within areas that appear either

Fig. 11. Plots of frontal lobe MWF in relation to (A) age and (B) years of education

relations between frontal lobe MWF and both age ( P= .01) and education ( P= .0

were seen in patients with schizophrenia.

unchanged from month to month, such as visually stable

lesions, or normal, such as normal-appearing white matter

(NAWM), that is not seen on conventional MRI, but is

contributing to disease progression.

The search for further insights into the pathophysiology

of both MS lesions and NAWM has led to the develop-

ment of new MR techniques that are more sensitive and

specific than conventional MRI. Using T2 relaxation, one

can measure both total water content and MWF, properties

that few have investigated in vivo. Previous work

examining 33 subjects with MS compared to 18 controls

has shown that, on average, MS lesions had 7.7% higher

total water content and 50% less MWF than NAWM [57].

Wide heterogeneity in lesion MWF was observed, includ-

ing some lesions that exhibited a total loss of myelin

water, suggesting complete destruction of myelin [57].

Fig. 9A and B shows histograms of water content and

MWF for MS lesions. Previous work has also shown

NAWM total water content was, on average, 1.8% greater

than that of NWM and that the MWF was 15% lower in

NAWM than in NWM [57]. Fig. 10A and B shows water

in healthy controls and schizophrenia patients. While statistically significant

06) were observed in healthy subjects, no statistically significant relations

Fig. 12. Example of a strong correlation between MWF and blue optical

density of luxol fast blue stain for various regions of interest in a formalin-

fixed MS brain.


content and MWF for various white matter structures from

subjects with MS compared to controls. Therefore, MS

affects more of the brain than just the bright lesions visible

on conventional images. A simple model for water in brain

demonstrates that edema alone cannot account for the

observed changes [57], suggesting NAWM has undergone

some loss of myelin, a finding which is supported by

pathological studies [58,59].

8.2. Schizophrenia

Schizophrenia has a lifetime risk of approximately 1%,

usually begins in early adult life and often results in decades

of severe disability. It involves disorganised thinking,

delusions, hallucinations, changes in emotions and changes

in behaviour. Despite effective treatments for psychotic

symptoms in some people, the mechanisms that account for

disability are poorly understood. One type of chronic

disability experienced by people with schizophrenia is

impaired cognition, a symptom for which treatment has

been less successful.

A key factor in schizophrenia is a lack of connectivity

between regions of functioning brain. A possible mecha-

nism for this connectivity deficit is reduced myelination in

the neurons that connect the frontal lobe to other areas of the

brain. To test this hypothesis, a study [60] was carried out to

compare brain MWF measurements in people with schizo-

phrenia with normal controls that were matched for age,

gender and parental education.

While conventional brain MR images of subjects with

schizophrenia look normal, the T2 relaxation results

demonstrated a statistically significant reduction in MWF

in the frontal lobe from 30 people with schizophrenia as

compared to 27 matched healthy controls. Furthermore,

frontal lobe MWF increased with age, and with years of

education, in healthy controls but not in subjects with

schizophrenia (see Fig. 11). In summary, this study

suggested that impaired myelination could contribute to

abnormalities of neuronal connectivity and persistent

functional impairment in schizophrenia.

9. Pathological validation

T2 relaxation has the potential to quantitatively define

the role of myelin-specific pathology, thereby furthering

our understanding of demyelinating disease pathogenesis.

However, as for any MRI method claiming to be an in vivo

marker for myelin, pathological validation is crucial: if

myelin water is indeed a marker for myelin it must

correlate quantitatively with histopathological measures of

myelin. Studies on guinea pig [45,46] and rat [61] models

showed white matter T2 distributions to contain multiple

components and found that the short T2 component was

correlated with histological measures of myelin. Fortu-

itously, the T2 distribution from formalin-fixed human

brain is shaped similarly to that from brain in vivo,

although the T2 times are shifted to lower times. This is

perhaps not surprising since water T2 times are strongly

influenced by cellular structure [62] which is relatively

unchanged by fixation. Histopathological studies have

shown that the MWF qualitatively corresponds to the

anatomic distribution of myelin as demarked by the luxol

fast blue stain [63]. Fig. 12 [10] shows a strong correlation

between MWF and blue optical density in fixed brain slices

stained by the luxol fast blue stain for myelin. These

studies provide strong evidence that the MWF is, indeed, a

marker for myelin in white matter.

References

[1] Bottomley PA, Hardy CJ, Argersinger RE, Allen-Moore G. A review

of 1H nuclear magnetic resonance relaxation in pathology: are T1 and

T2 diagnostic? Med Phys 1987;14(1):1–37.

[2] Fischer HW, Rinck PA, Van Haverbeke Y, Muller RN. Nuclear

relaxation of human brain gray and white matter: analysis of field

dependence and implications for MRI. Magn Reson Med 1990;

16(2):317–34.

[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(2):214–27.

[4] Stewart WA, MacKay AL, Whittall KP, Moore GR, Paty DW. Spin–

spin relaxation in experimental allergic encephalomyelitis. Analysis of

CPMG data using a non-linear least squares method and linear inverse

theory. Magn Reson Med 1993;29(6):767–75.

[5] Poon CS, Henkelman RM. Practical T2 quantitation for clinical

applications. J Magn Reson Imaging 1992;2(5):541–53.

[6] Graham SJ, Stanchev PL, Bronskill MJ. Criteria for analysis of

multicomponent tissue T2 relaxation data. Magn Reson Med 1996;

35(3):370–8.

[7] Ye FQ, Martin WR, Allen PS. Estimation of brain iron in vivo by

means of the interecho time dependence of image contrast. Magn

Reson Med 1996;36(1):153–8.

[8] 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(6):673–7.


[9] 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(1):34–43.

[10] Laule C, Leung E, Li DKB, Traboulsee T, Oger J, Paty DW, et al.

Myelin water imaging in multiple sclerosis: quantitative correlations

with histopathology. Mult Scler 2003;(Suppl 1):S35.

[11] Skinner M, Whittall KP, MacKay AL. Effect of variable echo spacing

in multi-echo sequences for resolving long T2 components in multiple

sclerosis. Proceedings of the annual meeting of the International

Society for Magnetic Resonance in Medicine; 2001; Glasgow,

Scotland, 2001. p. 904.

[12] Kroeker RM, Henkelman RM. Analysis of biological NMR relaxation

data with continuous distributions of relaxation times. J Magn Reson

1986;69:218–35.

[13] Lawson CL, Hanson RJ. Solving least squares problems. Englewood

Cliffs, NJ7 Prentice-Hall; 1974.

[14] Whittall KP, MacKay AL. Quantitative interpretation of NMR

relaxation data. J Magn Reson 1989;84:64–71.

[15] Vasilescu V, Katona E, Simplaceanu V, Demco D. Water compart-

ments in the myelinated nerve: III. Pulsed NMR results. Experientia

1978;34(11):1443–4.

[16] Peled S, Cory DG, Raymond SA, Kirschner DA, Jolesz FA. Water

diffusion, T(2), and compartmentation in frog sciatic nerve. Magn

Reson Med 1999;42(5):911–8.

[17] Does MD, Snyder RE. T2 relaxation of peripheral nerve measured in

vivo. Magn Reson Imaging 1995;13(4):575–80.

[18] Wachowicz K, Snyder RE. Assignment of the T(2) components of

amphibian peripheral nerve to their microanatomical compartments.

Magn Reson Med 2002;47(2):239–45.

[19] Helms G. Volume correction for edema in single-volume proton MR

spectroscopy of contrast-enhancing multiple sclerosis lesions. Magn

Reson Med 2001;46(2):256–63.

[20] Tofts PS. Quantitative magnetic resonance methods. London7 Elsev-

ier; 2004.

[21] Bjarnason T, Vavasour I, Chia C, MacKay A. Characterization of the

NMR behaviour of white matter in bovine brain. Magn Reson Med

2005;54(5):1072–81.

[22] Vavasour IM, Magnetic resonance of human and bovine brain

[Doctoral]. Vancouver: University of British Columbia, 1998

[23] Stanisz GJ, Kecojevic A, Bronskill MJ, Henkelman RM. Character-

izing white matter with magnetization transfer and T(2). Magn Reson

Med 1999;42(6):1128–36.

[24] Zimmerman JR, Brittin WE. Nuclear magnetic resonance studies in

multiple phase systems: lifetime of a water molecule in an adsorbing

phase of silica gel. J Chem Phys 1957;61:1328–33.

[25] Edzes HT, Samulski ET. The measurement of cross relaxation effects

in the proton NMR spin-lattice relaxation of water in biolo-

gical system: hydrated collagen and muscle. J Magn Reson 1978;31:

207–209.

[26] Goldman M, Shen L. Spin–spin relaxation in LaF3. Phys Rev 1966;

(144):321–31.

[27] Vavasour IM, Whittall KP, Li DK, MacKay AL. Different magneti-

zation transfer effects exhibited by the short and long T(2)

components in human brain. Magn Reson Med 2000;44(6):860–6.

[28] Wolff SD, Balaban RS. Magnetization transfer contrast (MTC) and

tissue water proton relaxation in vivo. Magn Reson Med 1989;10(1):

135–44.

[29] Dousset V, Gayou A, Brochet B, Caille JM. Early structural changes

in acute MS lesions assessed by serial magnetization transfer studies.

Neurology 1998;51(4):1150–5.

[30] Filippi M, Dousset V, McFarland HF, Miller DH, Grossman RI. Role

of magnetic resonance imaging in the diagnosis and monitoring of

multiple sclerosis: consensus report of the White Matter Study Group.

J Magn Reson Imaging 2002;15(5):499–504.

[31] Henkelman RM, Huang X, Xiang QS, Stanisz GJ, Swanson SD,

Bronskill MJ. Quantitative interpretation of magnetization transfer.

Magn Reson Med 1993;29(6):759–66.

[32] Morrison C, Henkelman RM. A model for magnetization transfer in

tissues. Magn Reson Med 1995;33(4):475–82.

[33] Sled JG, Pike GB. Quantitative imaging of magnetization transfer

exchange and relaxation properties in vivo using MRI. Magn Reson

Med 2001;46(5):923–31.

[34] Moseley ME, Cohen Y, Kucharczyk J, Mintorovitch J, Asgari HS,

Wendland MF, et al. Diffusion-weighted MR imaging of anisotropic

water diffusion in cat central nervous system. Radiology 1990;176(2):

439–45.

[35] Moseley ME, Kucharczyk J, Asgari HS, Norman D. Anisotropy in

diffusion-weighted MRI. Magn Reson Med 1991;19(2):321–6.

[36] Chenevert TL, Brunberg JA, Pipe JG. Anisotropic diffusion in human

white matter: demonstration with MR techniques in vivo [comment]

Radiology 1990;177(2):401–5.

[37] Maedler B, Whittall KP, MacKay AL. Correlation of multicomponent

T2-relaxation data with diffusion tensor anisotropy measures in human

brain. Proceedings of the annual meeting of the International Society

of Magnetic Resonance in Medicine; 2002; Honolulu, Hawaii; 2002.

p. 1188.

[38] Beaulieu C. The basis of anisotropic water diffusion in the nervous

system — a technical review. NMR Biomed 2002;15(7–8):435–55.

[39] Brex PA, Leary SM, Plant GT, Thompson AJ, Miller DH.

Magnetization transfer imaging in patients with clinically isolated

syndromes suggestive of multiple sclerosis. Am J Neuroradiol

2001;22(5):947–51.

[40] FilippiM, CampiA, 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(3 Pt 1):478–82.

[41] Catalaa I, Grossman RI, Kolson DL, Udupa JK, Nyul LG, Wei L,

et al. Multiple sclerosis: magnetization transfer histogram analysis

of segmented normal-appearing white matter. Radiology 2000;

216(2):351–5.

[42] Cercignani M, Bozzali M, Iannucci G, Comi G, Filippi M. Magnet-

isation transfer ratio and mean diffusivity of normal appearing white

and grey matter from patients with multiple sclerosis. J Neurol

Neurosurg Psychiatry 2001;70(3):311–7.

[43] Laule C, Vavasour IM, Whittall KP, Oger J, Paty DW, Li DK, et al.

Evolution of focal and diffuse magnetisation transfer abnormalities in

multiple sclerosis. J Neurol 2003;250(8):924–31.

[44] Leary SM, Silver NC, Stevenson VL, Barker GJ, Miller DH,

Thompson AJ. Magnetisation transfer of normal appearing white

matter in primary progressive multiple sclerosis. Mult Scler 1999;

5(5):313–316.

[45] 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(9):1319–25.

[46] 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(6):586–95.

[47] Dousset V, Brochet B, Vital A, Gross C, Benazzouz A, Boullerne A,

et al. Lysolecithin-induced demyelination in primates: preliminary in

vivo study with MR and magnetization transfer. Am J Neuroradiol

1995;16(2):225–31.

[48] Norton W, Cammer W. Isolation and characterization of myelin. In:

Morell P, editor. Myelin, 2nd ed. New York7 Plenum; 1984. p. 147–95.

[49] Laule C, Vavasour IM, Paty D, Li DKB, Arnold DL, MacKay AL.

Correlation between magnetization transfer ratio and myelin water

content in normal white matter and MS lesions. Proceedings of the

annual meeting of the International Society for Magnetic Resonance

in Medicine. 2002; Honolulu, Hawaii; 2002. p. 182.

[50] 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(5):763–8.

[51] Keegan BM, Noseworthy JH. Multiple sclerosis. Annu Rev Med

2002;53:285–302.


[52] Charcot JM. Histologie de la sclerose en plaques. Gaz Hopit Civils

Milit 1868;41:554, 557, 566.

[53] Paty DW, Li DK. Interferon beta-1b is effective in relapsing-remitting

multiple sclerosis: II. MRI analysis results of a multicenter,

randomized, double-blind, placebo-controlled trial. UBC MS/MRI

Study Group and the IFNB Multiple Sclerosis Study Group.

Neurology 1993;43(4):662–7.

[54] Anonymous. Interferon beta-1b in the treatment of multiple sclerosis:

final outcome of the randomized controlled trial. The IFNB Multiple

Sclerosis Study Group and The University of British Columbia MS/

MRI Analysis Group. Neurology 1995;45(7):1277–85.

[55] Gasperini C, Horsfield MA, Thorpe JW, Kidd D, Barker GJ, Tofts PS,

et al. Macroscopic and microscopic assessments of disease burden by

MRI in multiple sclerosis: relationship to clinical parameters. J Magn

Reson Imaging 1996;6(4):580–4.

[56] Thompson AJ, Miller D, Youl B, MacManus D, Moore S, Kingsley D,

et al. Serial gadolinium-enhanced MRI in relapsing/remitting multiple

sclerosis of varying disease duration. Neurology 1992;42(1):60–3.

[57] 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(3):284–93.

[58] Allen IV, Glover G, Anderson R. Abnormalities in the macroscopi-

cally normal white matter in cases of mild or spinal multiple sclerosis

(MS). Acta Neuropathol Suppl 1981;7:176–8.

[59] Itoyama Y, Sternberger NH, Webster HD, Quarles RH, Cohen SR,

Richardson Jr EP. Immunocytochemical observations on the distribu-

tion of myelin-associated glycoprotein and myelin basic protein in

multiple sclerosis lesions. Ann Neurol 1980;7(2):167–77.

[60] Flynn SW, Lang DJ, Mackay AL, Goghari V, Vavasour IM, Whittall

KP, et al. Abnormalities of myelination in schizophrenia detected in

vivo with MRI, and post-mortem with analysis of oligodendrocyte

proteins. Mol Psychiatry 2003;8(9):811–20.

[61] 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(4):638–45.

[62] 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.

[63] 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(10):

1506–10.

Insights into brain microstructure from the T2 distribution

Documents

Transcript of Insights into brain microstructure from the T2 distribution