Aktuelle Neurologie 2009; 36 - V235
DOI: 10.1055/s-0029-1238425

Mapping of visuospatial attention in acute stroke patients with and without neglect

R Umarova 1, D Saur 1, C Kaller 1, MS Vry 1, V Glauche 1, I Mader 1, C Weiller 1
  • 1Freiburg

Introduction: Neglect is considered to be a deficit of visuospatial attention, which typically occurs in acute stroke patients. Chronic neglect patients were shown to have hyperexcitability of the left hemispheric structures that lead to the rightward bias of spatial attention. However, the difference in the functional state of the attention system between acute stroke patients with neglect (NGL) and without neglect (non-NGL) is still unknown. In the present study, we aimed at exploring this issue using functional MRI.

Methods: We examined 26 healthy control subjects (30.3±7.3 years old) and 23 patients (67.23±10.23 years old, 13 with NGL and 10 non-NGL) 50.57±31.22 hours after ischemic stroke in the territory of the right middle cerebral artery. Non-NGL and NGL patients did not significantly differ in age and time after stroke onset. All subjects were examined with fMRI while performing a task for visuospatial attention, which was a mix-designed Posner-like paradigm. During the task blocks, subjects were presented a central valid cue, followed by the right/left targets (valid trials, in 66%) or no target (null event, 33%). In total, 64 right and 64 left targets were presented to each subject. The task blocks were modeled as a main regressor. We analyzed the fMRI activation between 3 groups: healthy subjects (HLTH), non-NGL, and NGL.

Results: HLTH, non-NGL and NGL subjects detected 64, 57.8±10.7 and 21.6±21.8 left targets correspondingly; and 64, 59.2±10.4 and 51.6±8.5 right targets correspondingly. In healthy subjects fMRI paradigm activated visual and key attentional areas in both hemispheres. In non-NGL patients, the pattern of activation was the same as in HLTH, though the size of effect was significantly smaller. In NGL patients, the right visual and attention relevant areas had extremely small size of effect. Contrasting non-NGL vs. NGL showed difference in the activation of all key attentional centers in the left hemisphere and post-Rolandic structures in the right hemisphere.

Discussion: In non-NGL, the pattern of activation did not differ from HLTH that enabled the effective visuospatial processing. In contrast, NGL patients showed reduced activation in most of left and right attentional regions including undamaged right visual areas. Thus, though the structural lesion in non-NGL and NGL patients was similar, NGL patients demonstrated distinct functional state of their enire visuospatial attention system.