Anatomy- versus Sensitivity-Based Loci Preselection in Detecting USH2A-Retinopathy Microperimetric Progression

TL;DR

This study compares different methods for detecting vision loss progression in a specific retinal disease, finding that sensitivity-based and anatomical approaches perform similarly but better than others.

Contribution

The study introduces and evaluates a new sensitivity-based method for detecting progression in USH2A-retinopathy.

Findings

mFTP progression rate was significantly faster than MMS and ESS, but similar to HRS.

ESS had lower baseline sensitivity and was more prone to floor effects.

mFTP and HRS showed comparable performance in both trend- and event-based analyses.

Abstract

To compare microperimetry progression rate in USH2A-retinopathy using prespecified points based on fundus autofluorescence coregistration with loci preselected based on retinal sensitivity profile.

Cohort longitudinal study.

Seventeen eyes from 17 patients with biallelic pathogenic variants in USH2A gene.

Microperimetry was recorded using 10-2 grid. The grid was partitioned into 68 2° × 2° nonoverlapping squares, representing the retinal coverage of each locus. Four metrics were defined at baseline: (1) mean macular sensitivity (MMS): average sensitivity of all loci; (2) edge of scotoma sensitivity (ESS): average sensitivity of all loci adjacent to a scotomatous loci at baseline; (3) modified Rate of Progression in USH2A-related Retinal Degeneration study-defined functional transitional point (mFTP): selection based on ranking of the proportion peripheral adjacent loci that showed ≥7 decibel (dB) decrease; and (4) hyperautofluorescent ring sensitivity (HRS): average sensitivity of stimulus squares which the hyperautofluorescent ring boundary transects into. Trend-based progression rates (gradient from linear regression) were compared between these metrics, and event-based analysis of the US Food and Drug Administration-defined clinically significant change in visual field (mean change of ≥7 dB across ≥5 prespecified loci).

Trend- and event-based measures in MMS, ESS, mFTP, and HRS.

Seventeen patients (median age 37.0 years) had mean baseline values of 9.7 dB, 9.2 dB, 17.9 dB, and 13.1 dB for MMS, ESS, mFTP, and HRS, respectively. Using all longitudinal data (mean follow-up 4.0 years), trend analysis showed mFTP progression rate (–1.53 ± 1.37 dB/year) was significantly faster than MMS (–0.51 ± 0.63 dB/year) and ESS (–1.11 ± 1.23 dB/year) but similar to HRS (–1.29 ± 1.41 dB/year). Edge of scotoma sensitivity was more prone to floor effect and had lower baseline sensitivity than mFTP and HRS. In event-based analysis, the proportion of eyes that demonstrated clinically significant mean change was similar between ESS (2-year 36.4%, overall 45.5%), mFTP (2-year 33.3%, overall 43.8%), and HRS (2-year 28.5%, overall 42.9%) but noticeable less in MMS (2-year 13.3%, overall 12.5%).

Hyperautofluorescent ring sensitivity and mFTP showed comparable performance in both trend- and event-based analyses, superior to that of MMS and ESS. Additional advantage of mFTP is inclusion of patients without the autofluorescent ring.

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Figures