Is magnetic resonance spectroscopy capable of detecting metabolic abnormalities in neurofibromatosis type 1 that are not revealed in brain parenchyma of normal appearance?

BACKGROUND: Results of magnetic resonance spectroscopy studies in normal-appearing brain and in non-neoplastic brain lesions in individuals with neurofibromatosis type 1 (NF1) have been discrepant. OBJECTIVE: We used magnetic resonance spectroscopy to analyze the metabolic patterns in the basal ganglia of patients with NF1 and examine their correlation with focal hyperintense lesions in T2-weighted images (T2-weighted hyperintensities). METHODS: We used magnetic resonance spectroscopy data of 42 individuals with NF1 (18 with and 24 without T2- weighted hyperintensities) and 25 controls matched for gender and age. A single-voxel technique was employed by manually placing a region of interest with a uniform size over a predetermined anatomical region including the globus pallidum and putamen (capsulolenticular region). We further analyzed the ratios of choline/creatine, N-acetyl aspartate (NAA)/creatine, and myoinositol/creatine metabolites and the occurrence of T2-weighted hyperintensities in these regions in individuals with NF1. RESULTS: There was a significant difference between the NF1 and control groups with regard to the mean values of myoinositol/creatine and choline/creatine, with higher metabolite values observed in the NF1 group (P < 0.001). Only the myoinositol/creatine ratio was able to discriminate between NF1 subgroups with and without T2-weighted hyperintensities. For the NAA/creatine ratio, there was no significant difference between the NF1 and the control groups. CONCLUSION: Magnetic resonance spectroscopy allows the characterization of tissue abnormalities not demonstrable in the structural images of individuals with NF1 through choline and myoinositol metabolite analysis. Yet the preserved NAA values argue against demyelination and axonal degeneration occurring in the region, suggesting instead a functional neuronal stability. Taken in association with the findings of lack of clinical manifestations and the known transient nature of T2-weighted hyperintensities in NF1 as demonstrated by other studies, our results support the current histopathologically driven hypothesis that such T2-weighted hyperintensities may be related to intramyelinic edema.