Article Text

MR spectroscopy, a new in vivo biomarker for dementia disorders?
  1. Lars-Olof Wahlund
  1. Correspondence to Professor Lars-Olof Wahlund, Section for Clinical Geriatrics, Department of NVS, Karolinska Institutet, Hälsovägen 7, Novum floor 5, Stockholm S-14186, Sweden; lars-olof.wahlund{at}

Statistics from

In their JNNP publication, Laakso et al present a potential novel diagnostic tool for assessing dementia disorders. The technique they use, MR spectroscopy (MRS),1 is not new, but the application to detect compounds in the cerebrospinal fluid (CSF) to aid the differential diagnosis of dementia, is. MRS exploits the fact that hydrogen nuclei in matter perceive different magnetic fields depending on their local chemical environment. This slight shift in the intramolecular magnetic field gives rise to differences in the resonance frequency of the protons. The result of an MRS examination is a frequency spectrum, allowing for metabolite quantification or selection of unique features/patterns. Although the hydrogen nucleus (1H) is mostly used in MRS, other nuclei (13C, 19F, etc) have also been studied using this technique.

MRS is a useful technique in the clinic, since it offers the possibility of non-invasive study of certain in vivo brain processes. Usually, MRS of the brain is performed on clinical scanners with fairly low magnetic field strengths, 1.5 and 3 T. Only a limited number of compounds can be detected with signals high enough to be measured or even mapped in the brain. Laakso et al conducted an in vitro study, analysing CSF samples at high field strength (11 T)—an analytical approach, allowing for detection of numerous metabolites at a much lower concentration threshold. This is a method that has been widely used for many decades to determine exact molecular sample composition in analytical chemistry.2

To look for more effective diagnostic methods and to find reliable biomarkers to diagnose, for example, Alzheimer's disease, is obviously of great importance.

The technique Laakso et al present has a potentially high capacity to identify different dementia disorders. The patients in the study were diagnosed according to current diagnostic criteria and the authors report that they could identify several specific features, or “fingerprints”, that were unique for several of the diagnoses. They also reported that they can detect very early cases of Alzheimer’s disease. These findings are highly interesting but have to be validated in further studies; it is also important to have a more detailed description of the different proteins they have identified, not least to increase our knowledge of the background pathophysiology of the different diseases. This technique, which is novel in many ways, is similar to the work performed in many neurochemical laboratories where single proteins such as τ and Aβ have been identified in CSF using other analytical methods.

Currently, the assessment of dementia diagnosis is based on information from the patient's clinical symptoms. The diagnostic classification systems International Classification of Diseases, 10 Edition (ICD-10) and Diagnostic and Statistical Manual of Mental Disorders fourth Edition (DSM-IV) build up the diagnostic criteria on clinical presentations with little or no information from the pathophysiological backgrounds of the diseases. The number of dementia patients is increasing and will continue to increase in the next decades, resulting in a strong impact on healthcare systems and an increased cost for healthcare worldwide. Our struggle to understand dementia, and how dementia starts and develops, has given us increased knowledge about neurodegenerative disorders. The relation between clinical symptoms and disturbed protein metabolism is now well established, at least for Alzheimer's disease.3 Most dementia disorders are probably multifactorial, but to what extent external factors such as lifestyle or cardiovascular and cerebrovascular diseases influence the cognitive function has not yet been fully understood. There are cases with onset of disease at a young age, and in those patients, the influences of genetic factors is probably the main or perhaps even the only cause of the disease.

However, the vast majority of Alzheimer patients are relatively old, above 85 years of age. In those cases, most probably, a number of other factors are of importance, for instance, vascular and lifestyle factors.4 In the diagnostic work up in clinical routine, the patient's history and a physical and psychiatric evaluation are still the golden standard. Additional methods such as advanced imaging with MRI or positron emission tomography are increasingly used in secondary and tertiary evaluation centres. The availability of these modalities is still limited in most of the world. Analysis of CSF has become successful as a diagnostic tool.5 The identification of the key proteins in Alzheimer’s disease, τ and hyperphosphorylated τ as well as Aβ42, has had an impact on our method of diagnosis, and these proteins have now been suggested as an important part of the diagnosis in the revised diagnostic criteria for Alzheimer's disease.6 ,7 The rationale behind this is that CSF reflects relevant pathological changes in the brain more directly than proteins in blood analysis can. However, imaging techniques are currently available that can monitor the key pathological processes, for example, amyloid and τ imaging; but samples for these techniques are not as readily available as CSF taps. A large amount of work is going on to find new biomarkers in the CSF, not only for Alzheimer’s disease but also for frontotemporal dementia and Lewy body dementia, other examples of neurodegenerative disorders.

In vivo MRS of the brain has been used for 20 years to study brains in dementia and Alzheimer's disease. The technique has not yet been successful in terms of finding new biomarkers for dementia disorders. Certain compounds that are able to reflect neuronal activity and glial cell activity can be identified using MRS, but the technique is still far from ready for clinical routine work. Their strengths are non-invasiveness and has the possibility of being further developed to use higher magnetic field levels.

In the future, there will probably be more specific and unique proteins that can be related to not only Alzheimer’s disease but also to other neurodegenerative disorders, as well as to vascular components that might influence cognitive decline either as a single vascular cognitive impairment or as a factor that will increase cognitive decline in other neurodegenerative disorders. It is also obvious that the increased knowledge of the pathophysiological background for Alzheimer's questions the current diagnostic criteria, ICD-10 and DSM-IV in particular, criteria that do not include information on biological markers or any other pathological events in the brain. The method suggested by Laakso et al has the potential to increase our knowledge regarding the pathophysiology in neurodegenerative disorders by the identification of new proteins involved in the pathophysiological disease process. It also has the possibility to be useful as a diagnostic marker if the MRS patterns can be validated in larger studies.


View Abstract


  • Competing interests None declared.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles