Do we need to consider head-on-body position, starting roll position and presence of visuospatial neglect when assessing perception of verticality after stroke?

ABSTRACT

Background and Objective

Considering various factors that influence the accuracy of the Subjective Visual Vertical (SVV) and Subjective Postural Vertical (SPV), standardization of assessment methods is needed. This retrospective study examined the contribution of Head-on-Body (HOB) position, starting roll position (SRP) and visuospatial neglect (VSN) to SVV and SPV constant errors (i.e. deviation from true vertical). Also, the contribution of HOB position and VSN presence to SVV and SPV variability (i.e. intra-individual consistency between trials) was assessed.

Methods

First-ever unilateral hemispheric stroke survivors (<85 years; <100 days post-stroke) were assessed with three HOB positions (neutral, contralesional, and ipsilesional) and seven starting positions (20°Contralesional to 20° ipsilesional) of the laser bar and tilt chair. Linear mixed models were selected to evaluate the contribution of HOB, SRP, and VSN to SVV/SPV constant errors and variability.

Results

Thirty-four subjects (24 VSN-/10 VSN+) were assessed. A tilted HOB position led to significantly higher constant errors for the SVV and SPV (the latter only in the VSN- group), and an increased SVV variability. SRP only significantly contributed to the SVV constant errors and only in the VSN- group. Furthermore, the presence of VSN resulted in a significantly higher SVV and SPV variability.

Conclusions

HOB position and the presence of SRP and VSN are important factors to consider during SVV and SPV measurements. Assessment with a neutral HOB position leads to more accurate results. HOB position and SRP influence the results of SVV and SPV differently in individuals with and without VSN, which highlights the relevance of VSN assessment.

KEYWORDS:

• Stroke
• verticality perception
• visual vertical
• postural vertical
• spatial cognition
• visuospatial neglect

Acknowledgments

The authors wish to express their gratitude to Erik Fransen (StatUa, University of Antwerp) for statistical assistance.

Disclosure statement

The author(s) report there are no competing interests to declare.

Additional information

Funding

The work was supported by the University of Antwerp (research fellow CvdW) and the Special Research Fund of the University of Antwerp (DOCPRO no. 40180, research fellow EE).