A telerehabilitation program to improve visual perception in children and adolescents with hemianopia consecutive to a brain tumor: a single-arm feasibility and proof-of-concept trial

Abstract

Background: Brain tumor in children can induce hemianopia, a loss of conscious vision, profoundly impacting their development and quality of life, yet no effective intervention exists for this pediatric population. This study aimed to explore the feasibility, safety, and potential effectiveness of a home-based audiovisual stimulation in immersive virtual-reality (3D-MOT-IVR) to improve visual function and functional vision.

Methods: In a phase 2a, open-labeled, nonrandomized, single-arm study, conducted from July 2022 to October 2023 (NCT05065268), 10 children and adolescents with stable hemianopia were enrolled to perform 20-min sessions of 3D-MOT-IVR every other day for six weeks from home. We assessed feasibility by monitoring adoption, adherence and completion rates, remote data transfer and qualitative feedback. Safety was evaluated using validated cybersickness questionnaires. Comprehensive vision assessments following standardized low-vision evaluation procedures were conducted pre- and post-intervention, with follow-ups at 1- and 6 months.

Findings: The home-based 3D-MOT-IVR intervention proved both feasible and safe, with no reported adverse events. All participants completed the prescribed stimulations and the pre- and post-intervention assessment points, 90% completed the follow-ups. Nine out of ten participants showed clinically meaningful enhancement in visual function and/or functional vision, namely binocular visual field restoration and increased reading speed, but two showed concomitant deterioration in monocular visual field. These positive effects were sustained at the 6-month follow-up. Exploratory outcomes revealed a significant positive correlation between the performance at the 3D-MOT-IVR intervention and the visual perception at the binocular visual field test.

Interpretation: Our findings underscore the feasibility and safety of home-based audiovisual stimulation in immersive virtual-reality as a potential intervention for improving visual perception in children/adolescents with hemianopia consecutive to a pediatric brain tumor. These promising results lay a strong foundation for a larger randomized controlled trial, offering hope for a meaningful breakthrough in visual rehabilitation for this vulnerable population.

Funding: Meagan Bebenek Foundation and University Health Network Foundation.

Keywords: Audiovisual stimulation; Hemianopia; Low grade gliomas; Supportive care; Virtual-reality.

Fig. 1

CONSORT flow diagram.

Fig. 2

Dynamic audiovisual 3D-MOT-IVR stimulation. a. 8 orange spheres appear in the headset. One sphere is arbitrary cued (blue) and returns to orange after 5 s. b. All spheres start moving randomly for 20 s. c. After 20 s, the spheres stop moving and the participants must select, using laser pointer (light blue) the sphere matching the one previously cued (mark all procedure). The outcome is then displayed as either a hit (indicating a correct selection) or a miss (reflecting an incorrect selection), accompanied by feedback sound. d. Sequence returns to start configuration for the next trial.

Fig. 3

Contrast sensitivity. Graphs showing contrast sensitivity measures pre/post-intervention and at 1 and 6-month follow-up. 1M FU, 1-month follow-up; 6M FU, 6-month follow-up; COR, coefficient of repeatability.

Fig. 4

3D-MOT-IVR usage. Distribution and number of sessions (n = 180) performed during the day (06:00–00:00 divided in 3 h bins) by the participants (n = 10). The size of the circle indicates the performance of the session (average target speed in °/s for a correct choice).

Fig. 5

3D-MOT-IVR Overt and covert tracking performance. Graph representing the target speed for correct choice in the participants (n = 10) during the first and last session of overt and covert tracking (∗: 0·02 < p < 0·05, ∗∗: 0·005 < p < 0·01).

Fig. 6

Correlation head/eye and target trajectories between overt and covert tracking paradigms. Representative example of head/eye/target trajectory recordings during overt tracking in P9 (left homonymous hemianopia). a-a′. Graphs represent the position (in degree) as a function of time (s) of the target (orange (a) and blue (a′) dashed lines), the head (red (a), dark blue (a′) solid lines) and the eyes (orange (a) and blue (a′) solid lines) following the Y (orange, a) and X (blue, a′) axes during OVERT tracking. The gray shaded area represents the blind field. Head/target and Eye/target correlation coefficients for each axis are represented. b-b′. Graphs represent the position (in degree) as a function of time (s) of the target (orange (b) and blue (b′) dashed lines), the head (red (b), dark blue (b′) solid lines) and the eyes (orange (b) and blue (b′) solid lines) following the Y (orange, b) and X (blue, b′) axes during COVERT tracking. The gray shaded area represents the blind field. Head/target and Eye/target correlation coefficients for each axis are represented.

Fig. 7

Correlation between 3D-MOT-IVR performance and binocular visual field restoration. Graphs represent the correlation between performance (speed of the target - in °/s - for a correct choice) and the number of points seen at the Esterman binocular visual field test during overt (left) and covert tracking (right). Dashed lines represent the 95% CI. Pearson's correlation coefficient r values: overt tracking, r = 0·052 [95% CI: −0·46, 1·00], p = 0·44; covert tracking, r = 0·82 [95% CI: 0·54, 1·00], p = 5·8E-4.