Table 1.
Characteristics of 12 Male Subjects with Choroideremia.
Fig 1.
Horizontal SD-OCTs of all 12 subjects with X-linked choroideremia.
Horizontal SD-OCT line scans through the fovea of all 12 subjects (labeled) are presented here in the same order given in Table 1. Images are scaled and cropped to subtend a uniform retinal distance (6 mm). Degree of pathology varies, and older patients generally exhibit greater loss of central retina. Pathologic features can be seen: outer retinal tubulations (ORTs) are visible as hyperreflective ovaloid rings with hyporeflective lumens (filled arrowheads) and interlaminar bridges (ILBs) are visible as wedge-shaped hyporeflective structures, sometimes with a hyperreflective exterior, extending from the outer plexiform layer to Bruch’s membrane (open arrowheads). ILBs frequently coincide with the termination of the central zone of preserved retina, and sometimes flank ORTs. Note the asymmetry of many OCTs, particularly in cases of severe pathology. Longitudinal images are shown for 3 subjects (KS_0044, JC_0778, and JC_0782), showing early IZ attenuation and characteristic peripheral-to-foveal progression of atrophy. Scale bars, axial & lateral: 250 μm.
Fig 2.
Split-detector AOSLO imaging allows unambiguous imaging of photoreceptors and lesions borders.
Regions bordering areas of atrophy are shown here for all 5 subjects who were imaged using both confocal and split-detector AOSLO. Confocal images of 400 x 400 μm regions of interest (ROIs) are shown beside split-detector images of the same areas. Subjects and modalities are labeled. Confocal and split-detector images from a normal male, JC_0616, are shown for comparison. The ROIs are located at the following eccentricities: KS_0044, 3780 μm; JC_0699, 1432 μm; JC_0778, 2772 μm; JC_0782, 3050 μm; DW_10173, 1883 μm; JC_0616, 1899 μm. Note that confocal images often present confounding ambiguities in pathology: lesion borders are poorly defined, and abnormal cone reflectivity cannot be clearly distinguished from the reflectivity arising from rods, debris, and/or atrophic retina. In contrast, split-detector images offer superior delineation of lesion borders and unambiguous imaging of cone inner segments. Rods can be seen between cones in 4 of 5 cases (all but JC_0699). Scale bar: 100 μm.
Fig 3.
Abrupt termination of photoreceptor mosaic at lesion borders in choroideremia is contrasted to gradual loss of photoreceptors in retinitis pigmentosa.
Images from a subject with X-linked choroideremia (JC_0778, top) are compared with images from a previously-published[32] subject with retinitis pigmentosa (DH_10161, bottom). To facilitate comparison, images of the same modality share the same scaling. Labeled arrowheads in OCTs correspond to respectively labeled AOSLO images, and imaged regions are selected to correspond to pathologic features. Panels A and E show 0.25 x 0.25° (approximately 74 x 74 μm) areas of retina with intact IZ and EZ bands on SD-OCT. In both pathologies, cone inner segments appear grossly normal on split-detector (A2, E2), while only RP displays notably dim or dark cones on confocal (E1). Panels B and F show areas of retina with IZ band dropout but intact EZ. Confocal AOSLO in both (E1, F1) shows some irregular “multimodal” reflectivity, with dark cones again visible in the RP subject (F2). Split-detector imaging (E2, F2) again shows relatively normal inner segments. Panels C and G correspond to IZ and EZ dropout. In choroideremia, EZ dropout coincides closely with the loss of the ELM and presence of an interlaminar bridge (ILB). On AOSLO, there is an abrupt termination to the photoreceptor mosaic (C2), with no cone inner segments present past this point (D2). In RP, ILBs are generally not seen, and the ELM continues well past EZ dropout. On AOSLO, remnant cone inner segments similarly persist well past the end of the EZ (G2), finally disappearing at or near the point of ELM termination (H2). Scale bars: OCTs, 100 μm lateral, 100 μm axial; AO, 10 μm.
Fig 4.
Variability in photoreceptor density near lesion borders.
Panel A displays a border region from KS_0044 approximately 1° nasal from the area shown in Fig 2, located at approximately 6.5° to 8.5° eccentricity. Arrowheads indicate two regions within ~100 μm of the lesion edge; the cone mosaic at the open arrowhead is noticeably sparser (B, 8,049 cones/mm2) than at the closed arrowhead (C, 15,474 cones/mm2), which is approximately 395 μm (or approximately 1.36°) more eccentric from the fovea. For comparison, normal mean ± standard deviation cone density measured from 9 normal subjects at an eccentricity of 7.36° (~2,142 μm) is 11,833 ± 1,816 cones/mm2.[36] Scale bars: 50 μm.
Table 2.
Cone Density Variability Near Lesion Borders.
Fig 5.
Bubble lesions correspond to subretinal hyporeflective regions on SD-OCT.
SD-OCT and confocal AOSLO imaging of bubble lesions are presented here from two subjects, JC_0618 (left) and JC_0752 (right). The lateral distance subtended by the AOSLO imaging windows are indicated by two black arrows in frames A, C, D, and F, while the location of the SD-OCT B-scans are indicated by white arrows in frames B and E. In JC_0618, an ORT is indicated by the asterisk (*) allowing mapping of the bubble-like lesions seen in B to the regions indicated by black arrowheads in A and C. In JC_0752, double asterisks (**) indicate pigment clumps, while triple asterisks (***) mark a retinal blood vessel. These features help localize the bubble-like lesions in E to the region indicated by the black arrowhead in D. In F, where the plane of the B-scan does not cross the bubble-like lesions, and the spot in D is no longer visible. Scale bars: A, C, D and F, 500 μm; B and E, 100 μm.
Fig 6.
ORTs viewed en face on SD-OCT are contiguous with central regions of preserved retina.
Four sets of SD-OCT en face and B-scan images are presented here for subjects JC_0621, JC_0699, JC_0754, and JC_0942. En face projections are aligned to averaged line scans passing horizontally (bottom) and vertically (right) through the fovea. Orthogonal lines on each en face OCT indicates the locations at which line scans were taken. B-scan cross sections of ORTs can be seen as distinct hyperreflective ovaloid structures superficial to Bruch’s membrane; in en face projection, ORTs appear as long, thin protrusions. Most ORTs are contiguous with the central region of preserved retina, though remnant “island” ORTs can also be seen amidst surrounding atrophy. Scale bars: all lateral & en face OCTs (L/E) = 1,000 μm, axial OCTs = 200 μm.
Fig 7.
A-G present a vertical SD-OCT B-scan (A), an en face SD-OCT projection (B), a split-detector AOSLO montage (C), and zoomed-in views of the montage (D-G) of an ORT in subject JC_0778. Labeled arrowheads indicate corresponding locations in different modalities. The ORT indicated by the filled black arrowhead is structurally distinct on B-scan (A) from the wider patch of preserved retina just superior (indicated by the open black arrowhead). Top and bottom arrowheads in C indicate the top and bottom edges of the zoomed-in 100 x 100 μm images provided by panels D to G. Relatively normal cones are seen in the region of preserved retina (D) while sparse, morphologically abnormal remnant cone inner segments are seen throughout the ORT (E, F, G). Scale bars: A, lateral (vertical) & axial (horizontal) = 250 μm, axial (horizontal) = 100 μm; B, 250 μm; C, 50 μm; D-G, 50 μm.