A systematic literature review & meta-analysis of the use of MR spectroscopy for the study, diagnosis, subtype classification & prognosis of multiple sclerosis. Abstract: Background: Multiple sclerosis (MS) is a primary demyelinating disorder of the central nervous system (CNS), where the myelin sheath covering neuronal axons becomes progressively damaged, either in a multifocal or a diffuse manner across the brain and spinal cord. Magnetic resonance imaging (MRI) techniques introduced an extremely helpful tool in diagnosing the disease, observing its natural history and classifying its various subtypes, in correlation with the clinical presentation of the patients. Moreover, they offered a reliable way for evaluating the effectiveness of anti-MS pharmacological interventions. The aim of this work is to perform an integrative evaluation on the literature related to the use of magnetic resonance spectroscopy (MRS) in the study and clinical management of MS. Methods: The literature search looked for any publications investigating both clinical MS & MRS without limits as to publication date. Of 260 publications which met the initial search criteria, 157 were excluded leaving 103 for analysis. Results: Evolution of MRS as a tool for investigating MS is almost exclusively based on proton spectroscopy, with NAA, choline, lipids & lactate the primary metabolites investigated. Generally phosphocreatine is elevated in MS brains; the broad phosphorous component is reduced in plaques; & NAA is reduced in those positive for the human leukocyte antigen DRB1*1501. Clinical application of MRS is challenging: technical factors generate variability between subjects which is not related to the underlying disease; SNR is poor; suppression of water & fat signals can be inconsistent; and multiple & different processing steps, often manufacturer dependent, cannot always be controlled for. Technical & statistical adjustments have been tried to improve quality of data e.g. oedema volume correction, metabolite-nulling, & absolute quantification techniques, which have been proposed to control for age. Very few studies have looked at spinal cord lesions. MRS data across multiple publications was looked at in the following ways: global metabolite levels; grey matter levels; acute lesion levels; chronic lesion levels; and disease-free white matter levels. Conclusion: Based on the cumulative results, the pattern of changes in the spectroscopic values of the basic metabolites in disease-free white matter of MS patients are similar to those of healthy controls, & so as yet the technique cannot predict who will develop MS or where a new MS lesion might occur in a person already with MS. Characterizing biochemical markers in the plaques themselves is difficult because of the different ages & activities of plaques at the times of scanning. Project type: Systematic review Imaging keywords: Magnetic resonance (MR) Magnetic resonance spectroscopy (MRS) Application / disease keywords: Multiple sclerosis (MS) Supervisor(s): Dr Peter Connick Prof Robin Sellar Programme: Neuroimaging for research MSc Year: 15-16 This article was published on 2024-08-22