Higher visual function deficits are independent of visual acuity measures in children with cerebral visual impairment

Cerebral visual impairment (CVI), the leading cause of bilateral visual impairment in children, is often characterized by visual acuity (VA) loss and higher visual function deficits (HVFDs). However, the relationship between VA loss and HVFDs remains unknown. A previous study using the Higher Visual Function Question Inventory (HVFQI) demonstrated that normal VA did not preclude HVFDs. In this prospective controlled study of children with CVI, we examine the relationship between HVFDs and degrees of VA loss to refine our understanding of this relationship. We introduce two new indices-HVFD spectrum and severity-to provide a comprehensive view of how CVI affects the individual child and the entire cohort. We also performed an analysis to determine the effectiveness of the HVFQI in eliciting HVFDs and present a preliminary analysis of the relationship between HVFDs and age. The study participants included 59 children with CVI (age: 9.87 +/- 3.93 years [mean +/- SD]; binocular VA: 0.35 +/- 0.34 log MAR.) and 120 neurotypical (NT) children with normal visual acuity (age: 8.7 +/- 2.8 years; binocular VA: 0.14 +/- 0.16 logMAR). Clinical history and notes independently confirmed the diagnosis of CVI. Parents were interviewed with the HVFQI, and their responses were recorded using a five-level Likert scale. Mann-Whitney U-test (MWU) determined the ability of HVFQI to distinguish between CVI and NT participants; Fisher's exact test (FET) and d-variable Hilbert-Schmidt independence criteria (dHSIC) assessed the independence between HVFDs and VA. The average spectrum (range 0-1) and severity (range 1-5) indices for CVI (spectrum: 0.65 +/- 0.24, severity: 3.1 +/- 0.77) and NT (spectrum: 0.12 +/- 0.17, severity: 1.42 +/- 0.49) were markedly different. MWU (p-value <0.00001) confirmed the ability of HVFQI to distinguish CVI from NT children for both indices. The FET reported a p-value of 0.202, which indicates that the data does not exhibit any relation between the HVFDs severity and VA. Analysis using dHSIC supports these findings (p-value 0.784). Based on these results, we urge that all children with suspected CVI need to be assessed for HVFDs in addition to VA measures. The HVFQI can potentially increase our understanding of the neural basis of visual perception, cognition, and visually guided action and lead us toward a conceptual model of CVI, translating to clinical practice improvements.