White matter T1 relaxation time histograms and cerebral atrophy in multiple sclerosis

https://doi.org/10.1016/S0022-510X(02)00044-8Get rights and content

Abstract

T1 relaxation time (T1) provides a quantitative magnetic resonance imaging (MRI) parameter for evaluating tissue damage in the brain. We aimed to measure T1 in the white matter of patients with multiple sclerosis (MS) and study relationships with cerebral atrophy, T2 lesion load and clinical parameters. Twenty-six patients with relapsing–remitting MS and sixteen healthy controls were scanned with dual-echo T2-weighted, 3-dimensional (3-D) magnetization-prepared rapid acquisition gradient echo and whole brain, multi-slice inversion recovery (IR) sequences. White matter masks were defined on axial T1 map slices using semi-automated seed growing and normalized ‘total white matter’ T1 histograms generated. Atrophy data was obtained using the Cavalieri method of modern design stereology. T2 lesion volume was also determined using seed growing.

T1 histogram-derived measures (median, peak height, peak position and standard deviation) in MS patients were significantly different (p<0.0001) from controls. Median T1 correlated significantly with supratentorial (r=0.42, p=0.036), lateral ventricle (r=0.55, p=0.004), and T2 lesion volumes (r=0.84, p<0.0001), but not with clinical parameters.

Total white matter T1 provides a robust, quantitative measure of global disease burden in MS, and also correlates significantly with cerebral atrophy. Serial studies are required to determine its potential role as a surrogate marker of disease progression.

Introduction

Conventional T2-weighted magnetic resonance imaging (MRI) provides a sensitive measure of lesion load in multiple sclerosis (MS), but lacks histopathological specificity for the presumed substrates of irreversible disability, severe demyelination and axonal loss [1]. T1-hypointense lesions have been shown to reflect axonal loss and demyelination [2], [3], and to correlate with disability [4]. However, the qualitative nature of T1-weighted images makes the assessment of T1-hypointense lesion load both subjective and sequence dependent [5]. By contrast, T1 relaxation time (T1) provides a quantitative MRI parameter that can be used to assess tissue damage in discrete lesions [6], [7], [8] and in the normal-appearing white matter (NAWM) [9], [10].

T1 increases as tissue water becomes more mobile and less organized [11]. Experimental models and clinical studies have demonstrated that prolongation of T1 reflects extracellular space expansion [12], [13]. Ultrastructural analyses of chronic MS plaques and the surrounding tissue have shown variable expansion of the extracellular space relating to differing degrees of axonal loss [6], [14]. The wide range of T1 values observed in MS plaques and in NAWM may therefore be an indication of this pathological heterogeneity [6], [7], [8], [10].

Previous quantitative studies of T1 in MS have been performed with varying accuracy and precision using different imaging techniques, the majority in a single-slice mode and requiring protracted acquisition times [15]. T1 analyses have therefore been limited to sampling small regions of NAWM or discrete lesions. We have acquired T1 values for the whole brain using a novel, rapid, multi-slice inversion recovery sequence and segmented the white matter to generate total white matter T1 histograms. Our objectives were: (i) to compare total white matter T1 values in MS patients and controls using histogram analysis, (ii) to understand the significance of the histogram parameters in relation to NAWM T1, and (iii) determine the relationships between total white matter T1 and more established MRI parameters, including cerebral atrophy, T2 lesion load and clinical parameters.

Section snippets

Subjects

Twenty-six patients with clinically definite relapsing–remitting MS were recruited. All subjects had at least two documented relapses as defined by Poser et al. [16] in the 2 years prior to study entry, and had been exacerbation free for at least 1 month. No patient had received previous treatment with immunosuppressive or immunomodulatory drugs, or corticosteroids within the last 1 month. None of the subjects had any significant medical condition, past or present, which could confound the

Total white matter T1 histograms

All histogram-derived measures in the MS patients were significantly different from controls (Table 1, Fig. 2), with the median and peak position shifted to higher T1 values. The patients' histograms showed marked variability in both peak height and ‘spread’ (represented by the SD).

The peak position of the total white matter T1 histograms showed a linear correlation with the NAWM T1 histograms' peak position (r=0.996, p<0.0001) (Fig. 3).

T2 lesion load

The median T2 lesion volume in patients was 3.4 cm3

Discussion

Our study has demonstrated that total white matter T1 values in MS patients are significantly different from controls, and correlate with measures of brain atrophy and the visible lesion burden. We have also shown that the peak position (mode) of total white matter T1 histograms predominantly reflects pathology in the NAWM.

In MS patients, the median and peak of the histograms were significantly shifted to higher T1, the peak height was lower and the spread of values greater than in controls.

References (43)

  • P.A. Brex et al.

    Lesion heterogeneity in multiple sclerosis: a study of the relations between appearances on T1 weighted images, T1 relaxation times, and metabolite concentrations

    J. Neurol., Neurosurg. Psychiatry

    (2000)
  • C.M. Griffin et al.

    MTR and T1 provide complementary information in MS NAWM, but not in lesions

    Mult. Scler.

    (2000)
  • M.A. van Walderveen et al.

    Neuronal damage in T1-hypointense multiple sclerosis lesions demonstrated in vivo using proton magnetic resonance spectroscopy

    Ann. Neurol.

    (1999)
  • H.B. Larsson et al.

    Nuclear magnetic resonance relaxation in multiple sclerosis

    J. Neurol., Neurosurg. Psychiatry

    (1998)
  • S. Naruse et al.

    Proton nuclear magnetic resonance studies on brain edema

    J. Neurosurg.

    (1982)
  • D. Barnes et al.

    Quantitative nuclear magnetic resonance imaging: characterisation of experimental cerebral oedema

    J. Neurol., Neurosurg. Psychiatry

    (1987)
  • H. Lassmann et al.

    Histopathology and the blood–cerebrospinal fluid barrier in multiple sclerosis

    Ann. Neurol.

    (1994)
  • P.S. Tofts et al.

    Towards quantitative measurements of relaxation times and other parameters in the brain

    Neuroradiology

    (1990)
  • C.M. Poser et al.

    New diagnostic criteria for multiple sclerosis: guidelines for research protocols

    Ann. Neurol.

    (1983)
  • J.F. Kurtzke

    Disability rating scales in multiple sclerosis

    Ann. N. Y. Acad. Sci.

    (1984)
  • P. Margosian et al.

    Imaging: imaging with half the data

    Health Care Instrum.

    (1986)
  • Cited by (0)

    View full text