Abstract
Purpose :
To determine, in individuals with age-related macular degeneration (AMD), the location-specific topographical association between differential light sensitivity (DLS), determined by microperimetry, and retinal layer thicknesses, obtained with spectral-domain optical coherence tomography (SD-OCT) and segmented using the Iowa Reference Algorithm.
Methods :
The cohort comprised 52 participants with early to late AMD, median age 80.5 yrs (range 55-93 yrs). Participants with ocular co-morbidity other than mild cataract (>Grade 2, LOCS III) were excluded. All participants underwent, in the study eye, microperimetry (MAIA microperimeter; 32 location 5° stimulus pattern, 1.27cdm-2 background luminance, stimulus size III) and volumetric imaging with the Spectralis SD-OCT (20° x 15° scan, 60-249 B-scans). The volumetric scans were segmented (9 boundaries, 8 retinal layers) using the Iowa Reference Algorithm. Three zones of topographical correspondence were evaluated: out to eccentricities of 1°, 3° and 5°. DLS was compared to a normative database (n=70). The topographical relationships between retinal layer thickness and DLS for each of the 3 zones were determined using standard multiple regression.
Results :
The median (IQR) total retinal thickness, retinal pigment epithelium and outer segment layers were 291µm (276, 310), 21µm (20, 23) and 21µm (16, 26), respectively. The median (IQR) of the Mean Sensitivity (MS) index within 1°, 3° and 5° was 20.8dB (14.3, 24.5), 20.8dB (14.5, 24.7), 21.6dB (14.3, 24.9). The median number (IQR) of locations exhibiting a probability level at ≤2% by Total Deviation (TD) and by Pattern Deviation (PD) probability analysis within 5° were 22 (8, 30) and 7 (4, 12) locations; the median (IQR) difference (TD-PD) was 11 (1, 16) locations. The total retinal thickness was an independent predictor of the mean TD (linear units) and accounted for 38.3% (adjusted R2) of the variance (p<0.001). The individual retinal layer thicknesses were rejected by the model (p>0.05). In secondary analyses, retinal layer thicknesses corresponding to abnormalities at ≤2% TD and to abnormalities at ≤2% PD were each not significantly different from locations exhibiting normality (all p>0.05).
Conclusions :
The limited relationships between retinal layer thicknesses and DLS indicate the need to adopt both techniques in the evaluation of AMD, and the necessity for further refinement of one or both outcomes.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.