Examining Whether AOSLO-Based Foveal Cone Metrics in Achromatopsia and Albinism Are Representative of Foveal Cone Structure

Abstract

Purpose: Adaptive optics scanning light ophthalmoscopy (AOSLO) imaging in patients with achromatopsia (ACHM) and albinism is not always successful. Here, we tested whether optical coherence tomography (OCT) measures of foveal structure differed between patients for whom AOSLO images were either quantifiable or unquantifiable.

Methods: The study included 166 subjects (84 with ACHM; 82 with albinism) with previously acquired OCT scans, AOSLO images, and best-corrected visual acuity (BCVA, if available). Foveal OCT scans were assessed for outer retinal structure, outer nuclear layer thickness, and hypoplasia. AOSLO images were graded as quantifiable if a peak cone density could be measured and/or usable if the location of peak density could be identified and the parafoveal mosaic was quantifiable.

Results: Forty-nine percent of subjects with ACHM and 57% of subjects with albinism had quantifiable AOSLO images. Older age and better BCVA were found in subjects with quantifiable AOSLO images for both ACHM (P = 0.0214 and P = 0.0276, respectively) and albinism (P = 0.0073 and P < 0.0004, respectively). There was a significant trend between ellipsoid zone appearance and ability to quantify AOSLO (P = 0.0028). In albinism, OCT metrics of cone structure did not differ between groups.

Conclusions: Previously reported AOSLO-based cone density measures in ACHM may not necessarily reflect the degree of remnant cone structure in these patients.

Translational relevance: Until AOSLO is successful in all patients with ACHM and albinism, the possibility of the reported data from a particular cohort not being representative of the entire population remains an important issue to consider when interpreting results from AOSLO studies.

Figure 1.

OCT-based photoreceptor metrics. For subjects with achromatopsia, foveal ONL thickness at the lowest part of the foveal pit and OCT EZ grade were assessed between the ELM and inner limiting membrane (ILM) using LRPs (example shown to right of OCT scan) in cases of complete foveal excavation as shown here. For subjects with albinism, ONL thickness was measured between the ELM and outer plexiform layer (OPL) at the location of maximum OS length. Although maximum OS length was assessed between the EZ and IZ, it is important to note for albinism that the ONL thickness and OS length measurements were not taken from the same LRP (as described in the Methods). All foveal ONL measurements included any Henle fiber layer that might have been present. Scale bar: 200 µm.

Figure 2.

Examples of variable AOSLO image quality in subjects with achromatopsia. Shown are split-detector images demonstrating variability in images that are quantifiable or usable (JC_0605 and JC_11401) versus images that are unquantifiable (JC_10257 and JC_10213). Although the peak cone density in JC_11401 cannot be accurately quantified due to packing of cones and resolution of split-detector imaging, parafoveal cones are easily identifiable and allow estimation of the location of peak cone density. Scale bar: 50 µm.

Figure 3.

Examples of variable AOSLO image quality in subjects with albinism. Shown are confocal images demonstrating variability in images that are quantifiable (JC_0103 and GS_10979) versus images that are unquantifiable (JC_10043 and JC_11934). Scale bar: 50 µm.

Figure 4.

Subjects with quantifiable AOSLO images have significantly better BCVA than subjects with unquantifiable AOSLO images. This trend was present in both (A) subjects with achromatopsia (P = 0.0276, Mann–Whitney test) and (B) subjects with albinism (P < 0.0004, unpaired t-test). The ends of the boxes are the 25th and 75th percentiles, the dashed line is the median, and the whiskers span the range of data.

Figure 5.

Representation of foveal outer retinal structure in subjects with quantifiable AOSLO images versus subjects with unquantifiable AOSLO images. (A) For achromatopsia, the EZ was graded as grade 1 if the EZ was intact and continuous, grade 2 if the EZ was disrupted, grade 3 if the EZ was absent and the external limiting membrane was collapsed, and grade 4 if a hyporeflective zone was present. There was a significant trend between the appearance of the EZ and the ability to quantify AOSLO images (P = 0.0028, χ² test for trend). (B) For albinism, there was no significant difference in OS length between those subjects with albinism for whom AOSLO images were quantifiable (mean ± SD = 32.32 ± 3.82 µm) and those for whom AOSLO images were unquantifiable (mean ± SD = 31.41 ± 3.71 µm; P = 0.9900, Mann–Whitney test). The ends of the boxes are the 25th and 75th percentiles, the dashed line is the median, and the whiskers span the range of data.

Figure 6.

Foveal ONL thickness was not significantly different in subjects with quantifiable AOSLO images compared to subjects with unquantifiable AOSLO images. This was the case in both (A) subjects with achromatopsia (P = 0.1320, Mann–Whitney test) and (B) subjects with albinism (P = 0.9900, unpaired t-test). The ends of the boxes are the 25th and 75th percentiles, the dashed line is the median, and the whiskers span the range of data.

Figure 7.

Foveal hypoplasia assessment was not significantly different in subjects with quantifiable AOSLO images compared with subjects with unquantifiable AOSLO images. This trend was present in both (A) subjects with achromatopsia for the presence or absence of foveal hypoplasia (P > 0.9999, Fisher's exact test) and (B) subjects with albinism for foveal hypoplasia grade according to the Leicester system (P = 1, χ² test for trend). For albinism, foveal hypoplasia grade 1a is marked by the absence of the extrusion of plexiform layers and presence of a nearly normal foveal pit, OS lengthening, and ONL widening; grade 1b is absence of the extrusion of plexiform layers and presence of a shallow foveal pit, OS lengthening, and ONL widening; grade 2 is absence of the extrusion of plexiform layers and foveal pit and presence of OS lengthening and ONL widening; grade 3 is absence of the extrusion of plexiform layers, foveal pit, and OS lengthening and presence of ONL widening; and grade 4 is absence of the extrusion of plexiform layers, foveal pit, OS lengthening, and ONL widening.