Nuclear magnetic resonance imaging and spectroscopy in experimental brain edema in a rat model.

Many aspects of the use of high-resolution nuclear magnetic resonance (NMR) imaging in the examination of brain edema have not been fully explored. These include the quantitation of edema fluid, the ability to distinguish between various types of edema, and the extent to which tissue changes other than a change in water content can affect NMR relaxation times. The authors have compared NMR relaxation times obtained by both in vivo magnetic resonance imaging (MRI) and in vitro NMR spectroscopy of brain-tissue samples from young adult rats with cold lesions, fluid-percussion injury, hypoxic-ischemic injury, bacterial cerebritis, and cerebral tumor. Changes in relaxation times were compared with changes in brain water content, cerebral blood volume, and the results of histological examination. In general, both in vivo and in vitro longitudinal relaxation times (T1) and transverse relaxation times (T2) were prolonged in the injured hemispheres of all experimental groups. Water content of tissue from the injured hemispheres was increased in all groups. A linear correlation between T2 (but not T1) and water content was found. Changes in the values of T1 and T2 could be used to distinguish tumor from cold-injured tissue. Cerebral blood volume was reduced in the injured hemispheres and correlated inversely with prolongation of T1 and T2. The results of this study suggest that, in a clinical setting, prolongation of T2 is a better indicator of increased water content than prolongation of T1, yet quantitation of cerebral edema based solely upon prolongation of in vivo or in vitro T1 and T2 should be undertaken with caution.