Incidence of Visuospatial Neglect in Acute Stroke: Assessment and Stroke Characteristics in an Unselected 1-Year Cohort.

Background: The true incidence of visuospatial neglect, impaired attention toward contralesional space, remains unclear. Common variability sources are sensitivity differences of conventional assessments and the exclusion of patients with language, motor, and other cognitive impairments. We aimed to determine the incidence of visuospatial neglect in an unselected cohort of patients with acute stroke using video oculography during free visual exploration, a newly established assessment overcoming the aforementioned biases.

Methods: Single-center, prospective, observational cohort study. We screened every patient admitted to a representative Swiss stroke center over 1 year (N=626). Two hundred eighty-five patients were eligible (first-ever stroke within 72 hours), and 221 were included. The incidence of visuospatial neglect was determined with conventional paper-pencil assessments and video oculography during free visual exploration. Demographic, risk, and stroke-related factors, as well as stroke localization, were also considered. Feasibility and ability to detect visuospatial neglect of the assessments were evaluated.

Results: The overall incidence of visuospatial neglect was ≈38%: widely varying location-specifically: ≈61% and ≈22% for stroke in the right and left cerebral hemispheres, respectively, and ≈14% to ≈37% for some less commonly affected infratentorial areas or multifocal stroke. In hemispheric stroke, visuospatial neglect was most common when the middle (≈64% right and ≈21% left) and posterior (≈53% right and ≈25% left) cerebral artery territories were affected. Neglect patients had higher National Institutes of Health Stroke Scale scores, more commonly atrial fibrillation and thrombectomy, and less commonly an undetermined stroke cause. They were older, with ≈4% yearly increase in the odds of having visuospatial neglect. Video oculography during free visual exploration was administrable and detected visuospatial neglect more often than conventional paper-pencil assessments.

Conclusions: The incidence of visuospatial neglect in an unselected cohort, using a highly sensitive assessment, is considerably higher than previously assumed and can also occur after less typically localized strokes. These results can enhance the awareness of visuospatial neglect in the acute setting, potentially facilitating earlier identification and therapy of this disabling disorder.