Proton MRS in neurological disorders

https://doi.org/10.1016/S0720-048X(99)00051-0Get rights and content

Abstract

Proton magnetic resonance spectroscopy (1H MRS) permits the acquisition of the signal arising from several brain metabolites. At long echo-time (TE) 1H MRS can detect N-acetyl-aspartate containing compounds, choline containing compounds, creatine+phosphocreatine and lactate. At short TE, lipids, tryglicerides, alanine, glutamate, glutamine, GABA, scyllo-inositol, glucose, myo-inositol, carnosine and histydine are visible. 1H MRS can be performed with single-voxel, multivoxel, single slice and multislice techniques. With single voxel 1H MRS it is possible to measure metabolites relaxation time, which allows the measurement of metabolite concentrations. This technique can be useful in the study of focal lesions in the central nervous system (CNS) such as epilepsy (pre-surgical identification of epileptic focus), brain tumors (evaluation of recurrence and radiation necrosis), stroke, multiple sclerosis, etc. Single slice and multislice 1H MRS imaging (1H MRSI) can be performed only at long TE and permits the mapping of the brain metabolites distribution which makes them particularly useful in studying diffuse diseases and heterogeneous lesions of the CNS. 1H MRS can also be useful in the evaluation of ‘ischemic penumbra’ of stroke; developmental (myelin and neuronal dysgenesis); head trauma (evaluation of cerebral damage not visible with MRI); degenerative disorders (identification of microscopic pathology not visible with MRI); and metabolic diseases (metabolic disturbances with specific metabolic patterns).

Introduction

Magnetic resonance spectroscopy (MRS) represents a major advancement toward an in vivo and noninvasive assessment of brain metabolism. The available MRS methods include localized 1H MRS at long and short echo time (TE), localized 31P MRS, 13C MRS, 15N MRS, 19F MRS and 23Na MRS.

Long TE 1H MRS detects the signal arising from four metabolites: N-acetyl-aspartate containing compounds (NA), choline-containing compounds (Cho), creatine+phosphocreatine (Cre) and lactate (Lac). Short TE 1H MRS identifies peaks from mobile lipids, Lac, alanine, NA, glutamate, glutamine and γ-aminobutyric acid, Cre, Cho, taurine, scyllo-inositol, glucose, myo-inositol (mI), carnosine and histidine.

Due to the low proton concentration, virtually all neurotransmitters, including acetylcholine, norepinephrine, dopamine, serotonine are beyond the detection limits (0.5–1.0 mM) of 1H MRS. Another limitation is the inaccessibility of most macromolecules (myelin, proteins, nucleosides, nucleotides, RNA and DNA) because of their limited mobility.

1H MRS has been applied to the study of many brain disorders. In recent years, this technique has evolved from single voxel, to multivoxels, and then to single and multislice 1H MRS imaging (1H MRSI). Single voxel 1H MRS is particularly suited for studying focal central nervous system (CNS) lesions. It can be easily performed at long and short TE, thus increasing the number of detectable brain chemicals. Single voxel 1H MRS has a relatively short acquisition time, which allows T1 and T2 measurement, and makes possible metabolites quantitation. The major disadvantage of single voxel 1H MRS is the possibility of evaluating only a small part of the brain without giving an overall view of the brain metabolism.

Single slice, and even more so multislice, 1H MRSI, are more suitable to assess diffuse CNS disorders and large and heterogeneous CNS lesions. The latter technique permits the simultaneous acquisition of metabolite signal intensities from four 15-mm slices divided into 0.84 ml single-volume elements [1]. The acquired data can be displayed in a tomographic format, thus enabling the mapping of the spatial extent of metabolic abnormalities in brain disorders.

Spectroscopic abnormalities could be: nonspecific and related to MRI visible brain abnormalities; nonspecific and nonassociated with MRI visible brain abnormalities; specific and directly related to the chemistry of the metabolic abnormality. Spectroscopic abnormalities may precede MRI changes. Finally, 1H MRS may be a valuable, objective tool to follow-up the course of several brain disorders and eventually to evaluate the effect of treatments.

In the mature brain NA is found only in neurons and axons [2], [3], [4]. NA is reduced in many brain disorders, in presence of neuronal and/or axonal loss, such as infarcts [5], [6], brain tumors [7], epilepsy [8], multiple sclerosis [9] and neurodegenerative diseases [10]. NA is specifically increased in Canavan’s disease [11] (see leukodystrophies, Section 9).

The Cre peak is generated by the sum of creatine and phosphocreatine and indirectly reflects energy metabolites. Since this sum is relatively constant under a variety of pathological processes, it has been used very often as a ‘reference’ peak to normalize metabolite signal intensities.

The Cho signal is generated by glycerophosphocholine, phosphocholine and free choline, which participate in membrane synthesis and degradation [2], [4], [12]. The Cho signal is increased in demyelinating diseases [9], in brain tumors [7], while a reduced Cho signal is found in hypomyelinating diseases [13].

In normal conditions the Lac signal is not detectable because of its low concentration. In pathologic conditions, when energy metabolism is deranged (ischemia, brain tumors, mitochondrial disturbances, etc.) Lac becomes detectable.

mI is thought to be located only in glial cells and is therefore considered a glial marker [14]. mI is increased in demyelinating diseases and in dementia [15].

Section snippets

Normative data

mI dominates the spectrum at birth, while Cho is responsible for the strongest peak in older infants. NA and Cre are less represented in neonates than in adults. NA and Cre increase, while Cho and mI decrease, during the first weeks of life. A stable pattern is reached with adult age, then NA decreases with aging.

Due to these age related changes it is crucial to have normative data for infants, children, adults and the elderly. The first comprehensive description of metabolite signal intensity

Alzheimer’s disease

Alzheimer’s disease (AD) is the most common cause of dementia in elderly people as it represents 50–60% of all the dementias. Vascular dementia accounts for 10–15%, another 10–15% is represented by mixed forms and the remaining 15–25% include dementias secondary to organic diseases. In AD, the diagnosis during life may be difficult and relies mainly on clinical criteria. MRI may help by depicting the preferred and early pattern of atrophy which involves structures like the hippocampus and

Brain tumors

MRI is quite sensitive in detecting tumor lesions but it is not very specific in differentiating and grading different types of neoplastic lesions. 1H MRS of gliomas [21] showed that the NA signal intensity was generally decreased and the Cho signal intensity was increased. Lac was detected only in some patients with either hypermetabolic or hypometabolic lesions, thus showing a lack of correlation between the Lac signal and human gliomas metabolism. 1H MRSI showed that, in high grade gliomas,

Epilepsy

Temporal lobe epilepsy (TLE) is the most common form of partial epilepsy, in many cases surgical intervention may be necessary if the seizures are not controlled by antiepileptic drugs. The most common lesion associated with TLE is hippocampal sclerosis (HS), which has been found in up to 65% of patients with TLE. With MRI HS looks like an atrophic area with increased T2-weighted signal intensity. To increase the sensitivity of MRI in detecting HS, quantitative techniques, like hippocampal

Motor neuron disease

Pioro et al. [25] studied patients with different forms of motor neuron disease (MND) by single slice spectroscopy. They differentiated patients according to the predominant symptoms and neurological signs and subdivided them into three groups. The first with the involvement of both the first and second motor neuron, were defined as having amyotrophic lateral sclerosis (ALS); the second, with signs of involvement of the primary upper motor neuron (PUMND); and the third with progressive spinal

AIDS

In the brain of AIDS patients 1H MRS showed a decrease of NA and Cre, associated with an increase of Cho and mI in both lesions and areas appearing normal at MRI [27]. The CNS involvement in patients with AIDS could be due either to a primary effect of the virus, or to an infectious or neoplastic process secondary to the immunodeficiency condition.

Multiple sclerosis

Multiple sclerosis (MS) is an autoimmune disease with variable course and clinical manifestations, characterized by focal demyelinatying lesions (plaques) in the CNS white matter.

Acute plaques are characterized by edema and some demyelination, while chronic plaques are characterized by gliosis and, to a minor extent, neuronal loss. Acute plaques can show an increase of Cho, Lac, mobile lipids peaks (products of myelin breakdown) [28] and the so called ‘marker peaks’ in the 2.1 to 2.6 ppm region

Leukodystrophies

In Canavan’s disease, the impairment of the NA breakdown is responsible for the specific increase in the NA signal intensity.

Tedeschi et al. [13] have recently described a new white matter syndrome characterized by a dramatic decrease of NA, Cho and Cre and by the increase of Lac. These 1H MRSI findings were in agreement with the pathologic data and suggested the hypothesis that the disorder was due to a metabolic defect causing hypomyelination and secondary axonal dysfunction.

Hereditary

Mitochondrial disorders

The brain derives its required energy mainly through oxidative metabolism. It follows that disorders in which there is an impairment of oxidative phosphorilation are characterized by preeminent CNS involvement. The impairment of oxidative metabolism is responsible for the accumulation of Lac which then becomes detectable with 1H MRS. Moreover the secondary neuronal dysfunction or loss can be identified by the decrease in the NA signal. 1H MRSI proved capable of differentiating between different

Stroke

1H MRS and 1H MRSI may help to evaluate the heterogeneous metabolic impairment occurring in the early stage of the disease and eventually to formulate a more precise prognostic evaluation. Barker et al. [33] found a reduction of NA in the stroke core and increased Lac in the peripheral region, where the NA signal was normal, thus depicting an ischemic penumbra and eventually allowing a better prognostic judgement. They also found an increased Cho signal, which was thought to be due to membrane

Parkinsonian syndromes

In clinical practice the differential diagnosis between parkinsonian syndromes such as cortico-basal degeneration (CBD), progressive supranuclear palsy (PSP) and Parkinson’ s disease (PD) may be a puzzling issue. Neuropathologically confirmed cases of PSP have been misdiagnosed with PD, CBD, Alzheimer’ s disease and multiple system atrophy. In a recent 1H MRSI study, Tedeschi et al. [34], found a different regional pattern of NA/Cre and NA/Cho abnormalities in all groups of patients (PSP, PD

Conclusions

At the present time, 1H MRS and 1H MRSI have several limitations. To obtain a good signal to noise ratio, the exam duration is still long and is exposed to the risk of patient movement. Even with the smallest possible volume of resolution (0.5 cc), we are still far from the desired optimal resolution. Moreover, some anatomical regions, such as the cerebellum and temporal lobes, are difficult to be explored owing to their closeness to bones and sinuses that are responsible for magnetic field

References (35)

  • B.L. Miller et al.

    In vivo 1H MRS choline: correlation with in vitro chemistry/histology

    Life Sci

    (1996)
  • J.H. Duyn et al.

    Multisection proton MR spectroscopic imaging of the brain

    Radiology

    (1993)
  • B.L. Miller

    A review of chemical issues in 1H NMR spectroscopy: N-acetylaspartate, creatine and choline

    NMR Biomed

    (1991)
  • J.R. Moffett et al.

    Immunohistochemical localization of N-acetylaspartate in rat brain

    NeuroReport

    (1991)
  • J. Urenjak et al.

    Proton nuclear magnetic resonance spectroscopy unambiguously identifies different neural cell types

    J Neurosci

    (1993)
  • Graham GD, Kalvach P, Blamire AM et al. Clinical correlates of proton magnetic resonance spectroscopy findings after...
  • H. Lanfermann et al.

    Metabolic changes in a cute and subacute cerebral infarctions: findings at proton MR spectroscopic imaging

    Radiology

    (1995)
  • G. Tedeschi et al.

    Increase of choline signal coincides with malignant degeneration of cerebral gliomas: a serial proton magnetic resonance spectroscopic imaging study

    J Neurosurg

    (1997)
  • A. Connelly et al.

    Proton magnetic resonance spectroscopy in MRI-negative temporal lobe epilepsy

    Neurology

    (1998)
  • S. Narayanan et al.

    Imaging of axonal damage in multiple sclerosis: spatial distribution of magnetic resonance imaging lesions

    Ann Neurol

    (1997)
  • L.L. Cheng et al.

    Quantitative neuropathology by high resolution magic angle spinning proton magnetic resonance spectroscopy

    Proc Natl Acad Sci USA

    (1997)
  • S.J. Austin et al.

    Localized 1H NMR spectroscopy in Canavan’s disease: a report of two cases

    Magn Reson Med

    (1991)
  • G. Tedeschi et al.

    Proton magnetic resonance spectroscopic imaging in childhood ataxia with diffuse white matter hypomyelination

    Neurology

    (1995)
  • B.D. Ross

    Biochemical considerations in 1H-spectroscopy. Glutamate and glutamine; myo-inositol and related metabolites

    NMR Biomed

    (1991)
  • T.K. Shonk et al.

    Probable Alzheimer disease: diagnosis with proton MR spectroscopy

    Radiology

    (1995)
  • G. Tedeschi et al.

    Brain regional distribution pattern of metabolite signal intensities in young adults by proton magnetic resonance spectroscopic imaging

    Neurology

    (1995)
  • Lundbom N, Barnett A, Bonavita S et al. MR image segmentation and tissue metabolite contrast in 1H spectroscopic...
  • Cited by (98)

    • Huperzine aggravated neurochemical and volumetric changes induced by D-galactose in the model of neurodegeneration in rats

      2022, Neurochemistry International
      Citation Excerpt :

      Also, in the D-gal + H group, the decrease in NAA/tCr at the eight week point was more significant than in the D-gal group (see Figs. 5 and 6). Typically, neurodegeneration is accompanied by a disrupted function and apoptosis of the neurons; thus NAA concentration (a marker of neuron number and function) decreases if the decline in the number of neurons or their dysfunction is severe enough (Giroud et al., 1996; Clark, 1998; Bonavita et al., 1999; Grachev et al., 2001; Rosso et al., 2017). A decrease in NAA is a physiologic event in normal ageing (Cleeland et al., 2019).

    • Intracranial metastases

      2022, Handbook of Neuro-Oncology Neuroimaging
    • Multi-vendor standardized sequence for edited magnetic resonance spectroscopy

      2019, NeuroImage
      Citation Excerpt :

      Proton (1H) magnetic resonance spectroscopy (MRS) is a non-invasive tool for measuring endogenous metabolite levels in the human brain to investigate healthy and pathological physiology (Bonavita et al., 1999).

    • Metabolic changes in de novo Parkinson's disease after dopaminergic therapy: A proton magnetic resonance spectroscopy study

      2015, Neuroscience Letters
      Citation Excerpt :

      Single-voxel 1H-MRS and multi-voxel proton MR spectroscopic imaging (1H-MRSI) are non-invasive techniques which provide in vivo informations about the distribution and concentration of a number of cerebral metabolites. The metabolites detectable with 1H-MRS include (i) the prominent resonances of N-acetylaspartate (NAA), interpreted as a neuronal marker; (ii) choline-containing compounds (Cho), viewed as a marker of membrane turnover or inflammation; (iii) creatine + phosphocreatine (Cr), which reflects energy metabolites; (iiii) myo-inositol (mI), suggested as a glial marker [24]. The aim of this study was to investigate metabolic changes in the motor cortex in a group of de novo drug-naïve PD patients before and 10 months after ropinirole treatment by using 1H-MRSI.

    • Low-grade gliomas: The challenges of imaging

      2014, Diagnostic and Interventional Imaging
    • Inflammatory Demyelinating Diseases

      2014, Quantitative MRI of the Spinal Cord
    View all citing articles on Scopus
    View full text