The Preferred Retinal Locus in Macular Degeneration: Relating Structure and Function

Purpose: Central field loss (CFL) that ensues from macular degeneration can impact many activities of daily living, including reading, in both younger (as in Stargardt disease, STGD) and older (age-related macular degeneration, AMD) subjects. Subjects with CFL typically choose a non-central retinal location, called the preferred retinal locus (PRL) for fixation. This dissertation aims to understand and relate functional and structural changes within the PRL. Methods: Preliminary studies determined the effectiveness of the MP-1 microperimeter (a) to compensate for excessively unstable fixational eye movements (FEMs), such as occur in subjects with CFL, and (b) to accurately register the retinal test locations on baseline and subsequent automated follow-up testing. Subsequently, the following functional measures were obtained for 29 subjects with bilateral CFL: (a) reading performance using hand-held MNRead charts and LCD-displayed MP1 Read charts, (b) contrast-detection thresholds using the Freiburg acuity test, (c) fixation stability on 3-letter words, measured as bivariate contour ellipse areas (BCEAs) with the MP-1, (d) sensitivity in the central visual field determined with a standard 10-2 threshold grid, and (e) fine-grained sensitivity within the word-fixation PRL for supra-threshold 13x13 arc min spots. Spectral-domain optical coherence tomography (SD-OCT) was used to assess structural characteristics of the PRL, specifically, thickness ratios for the retinal pigment epithelium - Bruch’s membrane complex (RPE-BM), the photoreceptor and outer nuclear layer (PL), and the total retina layers (TRL) between PRL locations where test spots were and were not consistently detected. Finally, 8 younger (< 35 years) and 8 older (>50 years) naive subjects with normal vision read high and low contrast sentences presented one word at a time at the fovea and 5 and 10° in the inferior field. Random 13x13 arc min blocks corresponding to 0-78% of the text area were set to the background luminance to simulate retinal micro-scotomas (MSs) and a staircase algorithm estimated the threshold reading rate. Results: The MP-1 compensated ~90% of the experimentally induced increase in FEMs and the average registration error was ~8 arc min. The maximum reading speed of subjects with CFL correlated poorly with contrast thresholds, BCEA, PRL eccentricity, median sensitivity around the PRL and all retinal thickness ratios. Twenty-two of 29 subjects with CFL (AMD: 8/10 subjects; STGD: 10/12 subjects) exhibited one or more MSs, defined as local regions of insensitivity for supra-threshold targets within the PRL. Although the average percentage of MSs was similar in the cohorts with AMD (25.4%) and STGD (20.3%), reading speed was significantly faster in STGD than AMD subjects. Average thickness ratios for RPE-BM, PL and TRL were 0.97, 0.84 and 0.86 respectively in the AMD cohort and 0.97, 0.77 and 0.89 respectively in the STGD cohort. Only TRL in subjects with AMD differed significantly from 1. In normally-sighted subjects, log reading rate decreased significantly with decreasing contrast and increasing age, eccentricity, and density of element-deletions. For a given eccentricity and contrast, a higher density of element-deletions maximally affected the older subjects. Conclusion: The compensation of the MP-1 for excessive FEMs and the registration between retinal test locations during baseline and follow-up testing are sufficient to assess functional changes within local retinal regions in subjects with CFL. MSs exist within the PRL of a high proportion of subjects with CFL, but are not strongly associated with structural changes determined using SD-OCT. Based on a simulation in normally-sighted subjects, we expect impact of MSs on reading to be greater for older than younger subjects with CFL.