Assessment of whole brain N-acetylaspartate – a new additional MR marker in mild cognitive impairment and Alzheimer's disease

Introduction: Biochemical abnormalities in AD and MCI have been investigated in vivo with localised proton MR spectroscopy (1H-MRS) providing data on small tissue samples of grey and white matter. Instead of using small tissue samples for MRS we used the recently proposed Whole Brain N-acetylaspartate (WBNAA) method to obtain a quantitative read out on a neuronal marker, which should provide useful information in addition to morphological information from MR imaging. We report its use in a cohort of clinically and neuropsychologically well characterised patients with mild cognitive impairment (MCI) and early Alzheimers disease (AD).

Methods: WBNAA spectra from 20 patients (10 AD median age 73; range 55–78/10 MCI median age 70,5; range 50–86) and 7 controls (median age 31; range 29–40 years) were analysed. Based on a detailed clinical/neuropsychological assessment patients were categorised as early AD or MCI. 1H-MRS and a high resolution 3D T1-weighted MPRAGE data set were obtained on a 3T scanner (Siemens ALLEGRA, Erlangen Germany). The brain volume was calculated with SIENA. Four consecutive MRS measurements were obtained in all individuals. The spectra were obtained and analyzed unaware of the clinical data. For each patient, the NAA peak integral was determined for each of the four measurements and the mean integral was calculated. The WBNAA concentration was then estimated as mean NAA integral divided by brain volume.

Results: The WBNAA was markedly reduced in all AD patients (mean 8,12mM±2,54) compared to controls (13,02mM±1,03)/(p<10–4), also there was a significant difference found between AD patients and individuals with MCI (11,30mM±2,23)/(p<0.008). Between controls and MCI patients the difference was not significant, but a strong trend was found (p<0.06). The MMSE score was significant different for AD (median 24.5, range 20–28) compared to MCI (median 28.5, range 25–30); p<0.02 (Wilcoxon rank sum test).

Conclusions: WBNAA is a sensitive quantitative measure of loss of neuronal integrity, which appears sensitive even in stages of relatively mild neuropsychological changes. WBNAA data aquisition is a relatively fast technique. WBNAA appears a useful and promising noninvasive measure providing an additional metabolic perspective on neuronal integrity. Our results may also indicate, that clinical scores appear sensitive to estimate changes in the early phase of neurodegeneration.