Quantitative magnetic resonance spectroscopy reveals a potential relationship between radiation-induced changes in rat brain metabolites and cognitive impairment.

To test the efficacy of magnetic resonance spectroscopy (MRS) in identifying radiation-induced brain injury, adult male Fischer 344 rats received fractionated whole-brain irradiation (40 or 45 Gy given in 5-Gy fractions twice a week for 4 or 4.5 weeks, respectively); control rats received sham irradiation. Twelve and 52 weeks after whole-brain irradiation, rats were subjected to high-resolution MRI and proton MRS. No apparent lesions or changes in T(1)- or T(2)-weighted images were noted at either time. This is in agreement with no gross changes being found in histological sections from rats 50 weeks postirradiation. Analysis of the MR spectra obtained 12 weeks after fractionated whole-brain irradiation also failed to show any significant differences (P > 0.1) in the concentration of brain metabolites between the whole-brain-irradiated and sham-irradiated rats. In contrast, analysis of the MR spectra obtained 52 weeks postirradiation revealed significant differences between the irradiated and sham-irradiated rats in the concentrations of several brain metabolites, including increases in the NAA/tCr (P < 0.005) and Glx/tCr (P < 0.001) ratios and a decrease in the mI/tCr ratio (P < 0.01). Although the cognitive function of these rats measured by the object recognition test was not significantly different (P > 0.1) between the irradiated and sham-irradiated rats at 14 weeks postirradiation, it was significantly different (P < 0.02) at 54 weeks postirradiation. These findings suggest that MRS may be a sensitive, noninvasive tool to detect changes in radiation-induced brain metabolites that may be associated with the radiation-induced cognitive impairments observed after prolonged fractionated whole-brain irradiation.