Table 3.
Summary of [1H]magnetic resonance spectroscopy longitudinal studies
| Authors | Year | Patients | Follow-up (months) | Converted to AD | MRI | Voxel size (cm3) | Brain region | Results |
|---|---|---|---|---|---|---|---|---|
| Chao and colleagues [70] | 2005 | 17 CIND | 43 | 6 | 1.5 T, Siemens | 0.9 | Medial temporal lobe | CIND converters had less medial temporal lobe NAA than controls. No difference between CIND stable and controls. No significant difference between CIND converters and CIND stable |
| Modrego and colleagues [62] | 2005 | 55 MCI | 36 | 29 | 1.5 T, GE | 8 | Left hippocampus, right parietal cortex, left occipital cortex | Occipital cortex NAA/Cr predicted conversion to dementia. ROC analysis for NAA <1.61 predicted conversion with 100% sensitivity and 75% specificity. Area under the curve was 0.91 with positive predictive value of 83% and a negative predictive value of 100% |
| Metastasio and colleagues [63] | 2006 | 25 MCI | 12 | 5 | 1.5 T, GE | 8 | Left and right paratrigonal white matter | Lower NAA/Cr at baseline for MCI converters compared with MCI stable |
| Rami and colleagues [64] | 2007 | 14 MCI, 28 prodromal AD | 12 | 3 MCI, 16 prodromal AD | 1.5 T, GE | 8 | Posterior cingulate, left temporal pole, left temporoparietal cortex | Posterior cingulate had higher Cho/Cr at baseline in MCI converters compared with MCI stable. Temperoparietal lobe showed lower NAA, Cho, and Cr at baseline in prodromal AD converters compared with nonconverters |
| Kantarci and colleagues [68] | 2007 | 49 MCI | 13 | 18 | 1.5 T, GE | 8 | Bilateral posterior cingulate and inferior precuneus | No baseline 1H MRS differences between MCI stable and MCI converters |
| Feyed and colleagues [65] | 2008 | 119 MCI | 29 | 54 | 1.5 T, GE | 8 | Bilateral posterior cingulate, left occipital cortex | NAA/Cr <1.40 in posterior cingulate predicted conversion of MCI to probable AD with a sensitivity of 82% and specificity of 72%. Area under the curve was 0.82 |
| Pilatus and colleagues [69] | 2009 | 15 MCI | 42 | 6 | 1.5 T, Philips | 8 | Posterior cingulate, parietal white matter | Did not replicate predictive power of NAA at baseline |
| Kantarci and colleagues [66] | 2009 | 151 MCI | 12 month evaluation | 75 | 1.5 T, GE | 8 | Bilateral posterior cingulate and inferior precuneus | Multivariate analysis showed that NAA/Cr added predictive value in addition to hippocampal volume and the presence of cortical infarction |
| Modrego and colleagues [67] | 2011 | 71 MCI | 22 | 27 | 1.5 T, GE | 8 | Bilateral posteromedial cortex, left medial occipital lobe | Decreased NAA/Cr in converters compared with nonconverters in posteromedial cortex and occipital lobe. NAA/Cr ratio ≤ 1.43 in posteromedial parietal cortex predicted conversion to probable AD at 74.1% sensitivity and 83.7% specificity; area under curve was 0.84. In left occipital lobe, 85.2% sensitivity, 61.4% specificity; area under the curve of 0.8 |
Since 2005, there have been nine longitudinal MRS studies performed in subjects with cognitive impairment. Subjects were typically followed for 1 to 3 years to identify a cohort that converted to dementia. AD, Alzheimer's disease; Cho, choline containing compounds; CIND, cognitively impaired not demented; Cr, creatine; GE, GE Healthcare (Waukesha, WI, USA); MCI, mild cognitive impairment; MRI, magnetic resonance imaging; MRS, magnetic resonance spectroscopy; NAA, N-acetylaspartate; Philips, Philips Medical Solutions (Best, The Netherlands); Siemens, Siemens Medical Solutions (Erlangen, Germany).