Patterns of Visual Task-based Functional MRI Activation in Chronic Posterior Cerebral Artery Stroke Patients.

Abstract

Stroke is aprincipal cause of disability worldwide. In motor stroke, the tools for stratification and prognostication are plentiful. Conversely, in stroke causing mainly visual and cognitive problems, there is still no gold standard modality to use. The purpose of this study was to explore the fMRI recruitment pattern in chronic posterior cerebral artery (PCA) stroke patients and to investigate fMRI as abiomarker of disability in these patients. The study included 10chronic PCA stroke patients and another 10age-matched volunteer controls. The clinical presentation, cognitive state, and performance in visual perceptual skills battery (TVPS-3) were determined for both patients and control groups. Task-based fMRI scans were acquired while performing apassive visual task. Individual and group analyses of the fMRI scans as well as correlation analysis with the clinical and behavioral data were done. At the level of behavioral assessment there was non-selective global impairment in all visual skills subtests. On visual task-based fMRI, patients recruited more brain areas than controls. These activations were present in the ipsilesional side distributed in the ipsilesional cerebellum, dorsolateral prefrontal cortex mainly Brodmann area (BA)9, superior parietal lobule (somatosensory associative cortex, BA7), superior temporal gyrus (BA22), supramarginal gyrus (BA40), and contralesional associative visual cortex (BA19). Spearman's rank correlation was computed to assess the relationship between the TVPS scores and the numbers of fMRI neuronal clusters in each patient above the main control activations, there was anegative correlation between the two variables, r(10) = -0.85, p ≤ 0.001. In chronic PCA stroke patients with residual visual impairments, the brain attempts to recruit more neighboring and distant functional areas for executing the impaired visual skill. This intense recruitment pattern in poorly recovering patients appears to be asign of failed compensation. Consequently, fMRI has the potential for clinically relevant prognostic assessment in patients surviving PCA stroke; however, as this study included no longitudinal data, this potential should be further investigated in longitudinal imaging studies, with alarger cohort, and multiple time points.