Parafoveal cone loss in inherited retinal dystrophy patients using the rtx1 adaptive optics retinal camera

Abstract

Purpose : Retinal imaging using adaptive optics (AO) enables “in vivo” analysis of cone metrics which cannot be attained with other image modalities. In this study, we report our results on parafoveal cone loss in Inherited retinal dystrophy (IRD) patients compared to healthy controls using the rtx1 adaptive optics retinal camera.

Methods : Twenty IRD patients (retinitis pigmentosa n=14, cone dystrophy n=1, Stargardt’s disease (STGD) n=2, Best disease n=1, achromatopsia n=1, central areolar choroidal dystrophy n=1; mean age 51 ± 13 years) and 31 controls (mean age 38 ± 11 years) were enrolled in the study. Our imaging protocol consisted of 2 rows (1.8° and -1.8° degrees vertically) of 7 overlapping images (4°x 4°, 1.5° overlap) from 2° nasal to 14° temporal from the fovea. Cone density and cone spacing were measured using the build-in automated algorithm at 2° and 10.5° temporal eccentricity for both rows. Outer retinal layer (ORL) thickness from OCT was correlated with cone density.

Results : Cone metrics could be obtained in 11 IRD patients and in 18 controls. Measurements were not possible in 9 IRD patients with late-stage disease, due to absent cones and in 13 controls due to low image quality. Between the rows cone metrics did significantly differ at 2° eccentricity for controls (p=0.002). At 2° eccentricity, the mean density and mean spacing were 17253 ± 3293 mm² and 8.48 ± 0.89 microns in IRD patients, compared to 21757 ± 1739 mm² and 7.51 ± 0.29 microns respectively in controls (p<0.001, p<0.001). At 10.5° eccentricity, the mean density and mean spacing were 11263 ± 4618 mm² and 10.85 ± 2.33 microns in IRD patients, compared to 14704 ± 1817 mm² and 9.48 ± 1.21 microns respectively in controls (p=0.054, p=0.041). Mean ORL thickness at 2° and 10.5° eccentricity were 86,95 ± 12.59 µm and 78.90 ± 6.67 µm in IRD patients and 84,50 ± 13,40 µm and 80,97 ± 2,94 µm in controls. A significant correlation was found between ORL thickness and cone density (R=0.591, p<0.001).

Conclusions : Cone metrics were significantly reduced in IRD patients at 2° eccentricity compared to controls. Parafoveal cone loss could be accurately measured using our AO imaging protocol in early-stage patients, while this could not be demonstrated with ORL thickness.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.