Abstract
Purpose :
A typical procedure used to quantify cone density (CD) is to consider the Preferred Retinal Locus (PRL) to manage the locus of measurement and to measure CD on a Region Of Interest (ROI) of 50x50 µm. The aim of this study is to test the relevance of these procedures.
Methods :
We used the rtx1TM Adaptive Optics fundus camera (Imagine Eyes) to acquire, along the horizontal meridian, 3 series of 7 images (at 0°, +/-2°, +/-4° and +/-6°), each subtended 4x4° on the retina of 30 non pathologic retinae of 15 subjects (20 to 32 years old). Series of images were montaged using the i2k Retina software (Dual Algn Inc.). Repeatability of the montage procedure was evaluated. Error of fixation was defined and calculated as the difference between the PRL at 2, 4 and 6° of eccentricities and the corresponding locus obtained by measuring the distance from the foveal PRL on the montage. Using the right eye’s retina of the subjects, CD and spacing, and proportion of hexagonal packing arrangement were analyzed, each degree from 2 to 6° (only on the temporal meridian), using various procedures. First, we tested 3 sizes of ROI (50x50 µm ; 100x100 µm ; 50(h)x150(v) µm). Then, we measured the criteria on 3 ROI of 50x50 µm along a vertical line (3 ROI procedure), each square being distant of 50 µm. The last procedure consisted in measuring the criteria on the surface occupied by 150 cones.
Results :
The average difference between two montages was 23±20 µm. The average error of fixation, mainly horizontally in the direction of the fovea, was 62±35 µm, and increased with eccentricity up to 87±35 µm. Moreover, according to the literature, eye motion even for the best fixators can be up to 0.5 deg (≈150 µm). Whatever the criteria (e.g. CD/spacing or hexagonal packing arrangement), the 50x50 µm ROI shows the worst repeatability whereas all other procedures are equivalent. However, no difference in the average values was observed for all criteria. Variation of criteria with eccentricities is in accordance with the literature.
Conclusions :
The PRL should not be considered to measure the density at a precise locus, a montage should be preferred. Cone criteria should be measured on ROI larger than 50x50 µm to increase repeatability.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.