Voxel-Guided Morphometry (VGM) for Intraindividual Assessment of Delayed Differential Alterations in Human Brain Volume after Cerebral Infarction

In our study, we examined delayed volume alterations in brain areas remote from the lesion proper in 6 stroke patients with an incomplete MCA infarction. MRI scans were carried out 7 days and between 3 and 16 months after stroke onset. Patients were examined by a 3D FLASH MRI scan. The brains from the same patient at two different times were matched to each other using an extended principal axes theory, a cross-correlation based procedure and a fast automated multiresolution full-multigrid movement model [2]. Brains were matched with a 3-step procedure: (i) ePAT and (ii) cross-correlation procedure for linear coarse (i) and fine (ii) alignment and (iii), a non-linear transformation achieving sub-voxel accuracy between source and reference brain. This enabled exact reproducibility due to the completely automated procedure and the ability to determine the unique movement to the reference volume using the total gray-value information. The resulting deformation fields could be visualized encoded in gray-values. The investigated brains exhibited a volume reduction in widespread cortical areas even remote from the lesion site and in the underlying white matter. Atrophy was also detected in the ipsilateral thalamus and caudate nucleus that were both not affected by the initial ischemia. These remote atrophic brain areas demonstrate that a secondary long-term atrophy develops after acute cerebral ischemia, probably due to degeneration of white matter projections. The secondary brain atrophy was accompanied by an improvement of the clinical state, marked by an improved European Stroke Scale in the follow-up examination. Obviously, secondary brain atrophy does not necessarily correlate with the clinical outcome. In summary, voxel-guided morphometry demonstrates a widespread and locally differential brain atrophy in brain areas remote from the stroke lesion site. These volume changes do not necessarily correlate with the improvement of clinical outcome. Supported by SFB 194. References [1] Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of ischaemic stroke: an integrated view. Trends Neurosci 1999; 22:) 391–397. [2] Schormann T, Zilles K. Three-dimensional linear and nonlinear transformations: an integration of light microscopical and MRI data. Hum Brain Mapp 1998; 6: 339–347.