Quantitative proton magnetic resonance spectroscopy of the human cervical spinal cord at 3 tesla

Anna Federica Marliani, Valeria Clementi, Luca Albini-Riccioli, Raffaele Agati, Marco Leonardi

Research output: Contribution to journalArticlepeer-review


Cervical spinal cord spectroscopy has the potential to add metabolic information to spinal cord MRI and improve the clinical evaluation and research investigation of spinal cord diseases, such as multiple sclerosis (MS) and intraspinal tumors. However, in vivo proton MR spectroscopy (1H-MRS) of the spinal cord is difficult to perform due to magnetic field inhomogeneities, physiological movements, and the size of the anatomical region of interest (ROI). For these reasons, few spinal cord 1H-MRS studies have been undertaken and two preliminary studies on a 3T system were only recently presented as abstracts. In this work we demonstrate the feasibility of cervical spinal cord quantitative 1H-MRS on a clinical 3T system, propose a study protocol, and report quantification results obtained from healthy volunteers. The main metabolite concentration ratios obtained in 10 healthy subjects, as provided by LCModel, were as follows: total N-acetyl aspartate/creatine (tNAA/Cr) 1.4 ± 0.3, choline/creatine (Cho/Cr) 0.5 ± 0.1, and myoinositol/creatine (ml/Cr) 1.7 ± 0.2. A significant difference was found between spinal cord tNAA, Cr, Cho, and ml concentration ratios and brainstem concentrations previously acquired on the same system.

Original languageEnglish
Pages (from-to)160-163
Number of pages4
JournalMagnetic Resonance in Medicine
Issue number1
Publication statusPublished - Jan 2007


  • 3 Tesla
  • LCModel
  • Metabolite quantification
  • Proton magnetic resonance spectroscopy
  • Spinal cord

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology


Dive into the research topics of 'Quantitative proton magnetic resonance spectroscopy of the human cervical spinal cord at 3 tesla'. Together they form a unique fingerprint.

Cite this