Development of a real-life mobility test for the performance evaluation of advanced retinitis pigmentosa subjects

Abstract

Purpose : The mobility tasks proposed in the literature are generally non-standardized and too complex therefore inappropriate for the evaluation of advanced visually impaired subjects. Their mobility performances do not sufficiently distinguish between the different levels of low vision (i.e. on-chart versus off-chart, e.g. counting fingers, hand motion, and light perception). This study proposes an exportable controlled simplified test to assess mobility performance of retinitis pigmentosa (RP) subjects with "very low vision".

Methods : Twenty-five subjects with RP (visual acuity less than 20/100) were recruited for four visits: two visual exams and two mobility tests (S1 and S2) each separated by one month in order to evaluate the repeatability and reliability of the test. The subjects performed the mobility test at Streetlab® on a controlled platform, integrating objective measurement tools. The mobility test consisted of following a line of different widths (10 cm and 1 cm) and contrasts (100%, 70%, 50% and 25%) under the same lighting environment (300 lux and 4500 kelvin) in both binocular and monocular conditions. The subjects were equipped with biometric sensors to precisely record mobility parameters (e.g. percentage of PWS (PPWS), traveled distance ratio, trajectory errors, etc.). A performance score [0 - 100] was then calculated based on those parameters.

Results : The mobility test is highly reproducible (test-retest intra-class correlations = 0.97) and reliable (correlation S1/S2 = 0.97). The median performance score of the population varies with the width and the contrast of the line from 0 to 83. The performance score is correlated with visual acuity (r² = 0.56) and contrast sensitivity (r² = 0.55), meaning that the most affected patients perform worse (e.g. fewer trials completed, lower PPWS, higher trajectory errors and traveled distance ratio) than the less affected patients of our population.

Conclusions : The performance score of our mobility test discriminates between "low vision" levels to categorize patients with very reduced visual capacities and so, can be a marker of the natural progression of the disease at advanced stages. Our mobility test could therefore be used to assess the effect of a therapeutic, curative or preventive, intervention targeting this group of patients.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.