Premium
Quantitative sodium imaging using ultra‐high field magnetic resonance imaging in patients with Alzheimer's disease
Author(s) -
Haeger Alexa,
Coste Arthur,
LermanRabrait Cécile,
Lagarde Julien,
Schulz Jörg B.,
Vignaud Alexandre,
Sarazin Marie,
Bottlaender Michel,
Reetz Kathrin,
Romanzetti Sandro,
Boumezbeur Fawzi
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.042107
Subject(s) - magnetic resonance imaging , nuclear medicine , medicine , flip angle , nuclear magnetic resonance , physics , radiology
Background In Alzheimer’s disease (AD), energy deficit is suspected to occur before clinical symptoms arise. Therefore, neuroimaging methods sensitive to metabolic alterations are attractive to improve diagnostics, therapeutics and follow‐ups of AD patients. An indirect measure of deficient Na + /K + ‐ATPase‐activity during neurodegeneration is provided by the tissue sodium concentration (TSC) measured by Sodium‐( 23 Na)‐MRI. Method 16 patients (7 Male, 71.7±8 years) diagnosed with AD (MoCA 15.1±7) were recruited from Ste. Anne Hospital in Paris as well as 16 cognitively healthy age‐matched control subjects (7 Male, 69±3.1 years; MoCA 28.1±1.8). Each subject underwent 23 Na and 1 H MRI examinations at the 7T MRI scanner (Siemens) at Neurospin, CEA. A dual‐resonance 1 H/ 23 Na volume coil (Rapid Biomedical) was used for a 23 Na MRI protocol acquiring ultra‐short echo‐time images (3mm isotropic resolution) at two flip‐angles (a1=25º/a2=55º) (Coste et al ., 2019). Briefly, processing steps comprised image reconstruction, denoising, motion‐ and B 1 ‐inhomogeneities‐corrections, and a calibration step using 4 external references. Individual TSC maps were co‐registered to their T 1 ‐weighted anatomical reference (MP2RAGE, 750 µm isotropic resolution) and a partial‐volume‐effect correction (Thomas et al ., 2016) was implemented to limit the influence of sodium from cerebrospinal fluid. Anatomical and TSC images were normalized to the OASIS template using ANTs. A parameter free permutation analysis was performed voxel‐wise on normalized TSC images and on the LogJacobian of the deformation fields. After, region‐of‐interest‐analyses on hippocampus from VolBrain‐Atlas followed. Result The voxel‐based analysis showed significant widespread increases in TSC of AD patients, particularly in parietal, temporal and frontal regions. Interestingly, the observed TSC changes were located in areas not directly associated to significant structural changes (Fig 1). Furthermore, TSC in the hippocampus was found significantly higher in AD than in controls (42.9±4.3mmol/l; 35.8±3.9mmol/l; Wilcoxon‐rank‐sum; p =0.00015, z=3.8) and was correlated, after controlling for age, with both the MoCA‐scores ( r =‐0.62; p =0.0002) and hippocampal volumes ( r =‐0.39; p =0.032). Conclusion High‐field quantitative 23 Na MRI in AD shows increased TSC values across a wide range of brain regions. These results may provide a deeper insight into metabolic alterations occurring during neurodegeneration. 23 Na MRI could therefore serve as a useful tool to characterize disease progression or the impact of therapeutic intervention.