Novel mobility test to assess functional vision in patients with inherited retinal dystrophies

Abstract

Importance: This novel endpoint tracks functional vision changes in patients with inherited retinal dystrophies (IRDs) over time.

Background: The aims of the study were to determine whether a multi-luminance mobility test (MLMT) can detect functional vision changes over time in subjects with IRDs and to assess natural history and potential effects of investigational agents.

Design: This is a prospective, observational study.

Participants: Sixty-two subjects were enrolled. Sixty (29 normal sighted and 31 visually impaired) were eligible; 54 (28 visually impaired and 26 normal-sighted) completed all testing visits.

Methods: Subjects navigated MLMT courses three times over 1 year. At each visit, subjects completed testing using individual eyes, and both eyes, at up to nine standardized, increasing luminance levels (range 1 to 400 lux). Accuracy and speed were evaluated and compared with visual acuity (VA), visual field (VF) and a visual function questionnaire.

Main outcome measures: Accuracy and speed of normal and visually impaired subjects on MLMT, and reliability and content validity of MLMT were the main outcome measures.

Results: MLMT distinguished normal-sighted from visually impaired subjects. All control subjects passed all MLMT attempts at all tested light levels. Visually impaired subjects' performance varied widely; some declined over 1 year. Performance declined markedly below certain VA and VF thresholds. Concordance on performance on two baseline visits was high: correlations for accuracy were 94% and 98% for lowest common and highest common lux levels.

Conclusions and relevance: MLMT differentiated visually impaired from control populations and, in visually impaired subjects, identified a range of performances; and tracked performance declines over time, consistent with these progressive conditions.

Keywords: functional vision; inherited retinal dystrophy; luminance; mobility test.

Conflict of interest statement

Competing/conflicts of interest: Spark authors (DCC, KAH and JW) are employees of Spark Therapeutics and hold equity in the company. The following authors are inventors on a submitted patent related to the MLMT: Daniel C. Chung, Jean Bennett, Jennifer Wellman, Sarah McCague and Katherine A. High. Zi-Fan Yu and Satha Thill provided statistical consulting to Children’s Hospital of Philadelphia (CHOP) and Spark Therapeutics through their employer, Statistics Collaborative, Inc. Julie DiStefano-Pappas provided data management services to CHOP and Spark Therapeutics through her employer, Westat, Inc. Jean Bennett and Kathleen Marshall are inventors on a copyright for the visual function questionnaire.

© 2017 Royal Australian and New Zealand College of Ophthalmologists.

Figures

Figure 1

Multi-luminance mobility test sample layout. Obstacles include two raised steps with arrows (2 in. high), two grass tiles, three black tiles representing holes, two elevated foam blocks (4.5 in. high), two Styrofoam (STY) cones, one stop sign (adjustable from 40 to 72 in. high), one step-over obstacle (9 in. high), one waist-high obstacle (represented by the blue tree, 27 in. high) and one waste basket (13 in. high).

Figure 2

Rigor in multi-luminance mobility test (MLMT) assessment. Following 40 min of dark adaptation, subjects underwent randomized eye patching, and then no eye patching, and were instructed to navigate one of 12 randomized MLMT courses. Testing was performed at the estimated sub-sensitivity cut-off level, and the estimated lower light sensitivity level. Videos were graded remotely by two masked, trained, independent graders, adjudicated by a third as necessary. A final score was computed based on accuracy and time needed to complete the course, factoring in any penalties. A passing grade on both components was necessary to pass the MLMT.

Figure 3

Accuracy score by course configuration, all visits. Each colour/symbol combination represents an individual person.

Figure 4

Time and accuracy scores, by pass/fail status, in the (a) normal-sighted and (b) visually impaired groups. All normal-sighted subjects received a ‘pass’ grade on the multi-luminance mobility test (MLMT) at all tested light levels. In contrast, visually impaired subjects showed a range of performance on both time and accuracy across multiple light levels. Each data point represents an individual subject’s measurement at a visit; multiple points per subject are plotted. Subjects passed on the accuracy component of the MLMT if their accuracy score was ≤0.25.

Figure 5

Visual acuity and accuracy scores, by pass/fail status, in the (a) normal-sighted and (b) visually impaired groups.

Figure 6

Goldmann visual fields (III4e) and accuracy score, by pass/fail status, in the (a) normal-sighted and (b) visually impaired groups.

Figure 7

Humphrey visual fields: macular threshold and accuracy, by pass/fail status, in the (a) normal-sighted and (b) visually impaired groups.

Figure 8

Multi-luminance mobility test change score, bilateral assessment, Baseline Visit 2 to 1-Year Visit.