Abstract
Purpose :
To rigorously assess the test-retest repeatability, content validity and construct validity of a novel Low Luminance Mobility Test (LLMT) and scoring algorithm to determine whether this new performance task is a suitable endpoint to include in clinical trials of retinitis pigmentosa (RP).
Methods :
A prospective, observational study, included both subjects with RP (n=20), mean best corrected visual acuity (BCVA) of 1.05 logMAR (range 0.55-1.56 logMAR), and without RP (NoRP, n=16, normal BCVA <0.05 logMAR). The LLMT used 13 light levels from 0.12 lux in increasing steps of 0.3 log units up to 500 lux. A novel scoring algorithm utilized MNREAD principles and R programming to develop a Critical Illumination Level (CIL) and Maximum Step Speed (MSS) score. The CIL is the lowest light level prior to a significant drop in adjusted step speed. MSS is the mean speed for the fastest trials on the plateau of the curve. Subjects completed the LLMT monocularly twice over two weeks, and video recordings of the trials, which were masked to visit and light levels, were sent to trained raters for scoring. For inter-rater and intra-rater reliability, grading of video trials of the LLMT across two studies using three raters allowed for the largest sample. In addition, BCVA, contrast sensitivity (CS), kinetic visual fields (KVF), and the VA LV VFQ-48 were compared to CIL scores to establish content validity.
Results :
The LLMT differentiated between NoRP and RP since NoRP subjects had a median CIL of 0.12 lux and mean MSS of 61.0 ± 11.01 steps per minute (spm), whereas RP subjects varied widely in performance with a median CIL of 32 lux (range 0.12-1000 lux) and mean MSS of 33.6 ± 14.93 spm. There was no change in the CIL between test sessions for 75% of RP subjects (n=15), and the five with variability all had constricted KVF diameter <12°. All visual function measures were significantly related to the CIL in a multiple regression model, R2=0.75 (p=0.004) but CS was the greatest contributor. Bland Altman plots for inter-rater and intra-rater video grading indicated biases close to zero, and there were no significant differences between graders (p>0.05).
Conclusions :
The LLMT is a novel low luminance clinical trial endpoint validated for RP with strong test-retest reliability, good content and construct validity, and an objective scoring system. The LLMT is a useful and reliable endpoint for studying low luminance mobility function in RP.
This is a 2021 ARVO Annual Meeting abstract.