Optical Coherence Tomography Angiography (OCT-A) in Choroideremia (CHM) carriers

Purposes: To explore OCT-A abnormalities in CHM carriers Methods: CHM carriers and age-matched controls were consecutively enrolled at the Eye Clinic in Florence. All patients underwent a complete ophthalmic examination, fundus photography, fundus autofluorescence (FAF) and OCT examinations. OCT-A images of the superficial capillary plexus (SCP), deep capillary plexus (DCP) and choriocapillaris slab (CC) were acquired and analyzed using ImageJ software to detect and quantify vascular density. Results: Six patients (12 eyes) and 8 age-matched controls (16 eyes) were included in our study. The mean age was 45.5 ± 16.3 years (range 15–61) for the CHM carriers and 46.6 ± 12.2 (range 18–54) for controls. All CHM carriers showed fundus abnormalities. The detected mean central retinal thickness (CRT) (220 ± 18.34 vs 227 ± 15.46; p =.342) and choroidal central thickness (CCT) (271 ± 54.28 vs 275 ± 38.36; p =.760) did not differ between the carrier and the control group, respectively. Quantitative analysis of the inner retinal vasculature disclosed no significant difference of both SCP (p =.437) and DCP (p =.859) vessel density compared to the control group. Of note, a mild reduction on the vascular flow of the CC could be detected in the carrier group compared to the control group (78.896 ± 13.972 vs 80.008 ± 10.862; p =.045). Conclusions: OCT-A allows us to underline the role of the retinal pigment epithelium in the CHM pathophysiology. Central inner retinal and choriocapillaris vascularization were preserved although the RPE was always involved in the CHM carrier: this could support a secondary role of vascular impairment in the natural history of the disease.

Keywords: Choroideremia; carrier; OCT-A; CHM; optical coherence tomography; OCT

Choroideremia (CHM, OMIN 303100) is a rare (1: 50,000 males) X-linked recessive degenerative disease characterized by progressive atrophy of the photoreceptors, RPE, and the choroid in affected males. It is caused by defects in the CHM gene, on chromosome Xq21.2, which encodes Rab escort protein-1 (REP-1), a key mediator of membrane trafficking in the retina and RPE. Affected males experience childhood-onset nyctalopia, followed by loss of peripheral visual field in their teenage years while female carriers retain good visual function throughout life, with either no symptoms or mild to moderate night blindness ([1]). Female carriers of choroideremia often demonstrate characteristic fundus changes including areas of chorioretinal degeneration, pigmentary granularity in the periphery, and yellowish drusen-like deposits ([2],[3]). Little is known about the degree of retinal and choroidal vasculature involvement in CHM female carriers. OCT-A is a novel and noninvasive diagnostic technique used to investigate both the retinal capillary and the choriocapillaris (CC) plexi ([4]). The contribution of OCT-A in understanding the pathophysiology of different ocular conditions such as adult onset vitelliform macular dystrophy ([5]), Stargardt disease ([6]), Retinitis Pigmentosa ([7]) and Best vitelliform macular dystrophy ([8],[9]) has been extensively described in literature. Using OCT-A, we have recently shown an early vascular involvement in both the inner retinal layers and choriocapillaris in young CHM patients ([10]). The vascular changes in CHM carriers have not been fully analyzed yet. The purpose of our work is to study vascular abnormalities of the retina and choroid vasculature in female carriers of choroideremia by means of OCT-A.

Six female CHM carriers were consecutively enrolled at the Regional Reference Center for Hereditary Retinal Degenerations at the Eye Clinic in Florence. The considered criteria for CHM diagnosis was the following: history of night blindness; appearance of the fundus characterized by varying degrees of chorio-retinal atrophy and with evidence of large choroidal vessels at the posterior pole and mid-periphery; ERG alterations (abnormal scotopic responses or abnormal scotopic and photopic components); VF abnormalities (scotomas that correspond to the observed areas of degeneration and progressive visual field constriction); X-linked inheritance; identification of pathogenic mutations in the CHM gene. Age-, sex- and ethnicity-matched healthy subjects without ocular or systemic diseases were considered as a control group. An accurate molecular and family history was taken into consideration. Both patients and controls received a comprehensive ophthalmological examination including best corrected visual acuity (BCVA), measurement of intraocular pressure, biomicroscopy of the anterior segment and fundus examination. Fundus autofluorescence imaging (FAF) (ultra-wide-field digital scanning laser technology, Daytona, Optos), color fundus photographs and OCT (swept source OCT Triton, Topcon Medical Systems Inc, Oakland, NJ, USA), full-field standard ERG (FF-ERG according to ISCEV protocols using RETIMAX (Roland Consult, Brandenburg, Germany), and Goldman Visual Field were performed. Moreover, OCT-A was performed using a swept source DRI OCT Triton (SS-DRI OCTA Triton; Topcon Corporation, Tokyo, Japan) with a scanning area of 3 × 3 mm centered on the fovea. Automated and manual segmentation of the full thickness retinal scans in the SCP and DCP, outer avascular retina, and CC was performed. SCP, DCP, and CC slabs were considered for quantitative and qualitative analysis.

We analyzed the OCT-A slabs according to the same procedure used in a previous work ([10]). Each OCT-A slab was exported from the Topcon software as a.TIFF format and loaded in ImageJ software (https://imagej.net/Welcome) to calculate vessel density. Vessel density was quantified using "adjust threshold tool" in ImageJ to classify white pixels as the flow signal and black pixels as background. Each image was binarized according to this threshold so as to have white colored vessels on a black background. The measurements obtained were expressed as the ratio between white pixels (flow) and the total pixel area of the 3 × 3 mm section scan. The SCP, DCP, and CC of the patients and the controls were analyzed using this method (Figure 1). The FAZ area based on SCP and DCP was manually outlined using a free-hand selection tool included in the OCT-A. Central retinal thickness (CRT) and central choroidal thickness (CCT) were measured on the OCT B-scan. CRT was recorded using the software SS-OCT Triton whereas CCT was assessed by manually measuring the subfoveal distance between Bruch's membrane interface and the sclero-choroidal interface to identify the inner and outer boundaries of the choroid, respectively.

PHOTO (COLOR): Figure 1. (A,F,M) Fundus color photographs of the right eye, left eye and right eye of patient P2 (51-year-old), patient P4 (42-year-old) and control C4 (54- year-old).(B,G,N) High resolution OCT B-scans of patients P2, P4 and C4, respectively. In particular, OCT scan of patient P3 (G) shows a hyper-reflective alteration on the RPE-Bruch's band which correspond to a yellowish macular dot. 3 × 3 mm OCT-A scans centered on the fovea (red square) showing the retinal vascular texture of the superficial capillary plexus (C,H,O); deep capillary plexus (D,I,P); and (E,L,Q) choriocapillaris layer of patients P2 (C,D,E), P4 (H,I,L) and C4 (O,P,Q), respectively, with corresponding images processing the algorithm extracting flow signal from the three different slabs. Vessel density was calculated using the ImageJ thresholding tool in order to have white pixels (flow signal) on a black background (flow signal voids).

To achieve good visualization of the choroid, high-resolution imaging-OCT was used in all OCT scans. According to our previous work ([3]), we classified CHM female carriers in 4 different fundus pattern (A,B,C,D) according to the different appearance of the posterior pole according to the presence of RPE abnormalities in order to evaluate a possible relationship between fundus pattern and OCT-A abnormalities: no clear posterior pole alterations in pattern A, RPE dystrophy with small hypo-pigmented RPE areas in pattern B, RPE dystrophy with small yellowish well-defined dots (drusen-like deposits) in pattern C, and RPE dystrophy with yellowish well-defined dots and large hypo-pigmented RPE areas in pattern D.

The genetic examination was performed at the Department of Genetic Diagnosis at the Careggi Teaching Hospital in Florence. All statistical analyses were performed using a Student independent samples t-test. Results of the analyses are expressed as mean ± SD for quantitative variables. A p-value less than 0.05 was considered statistically significant.

This study was approved by the local research Ethics Committee in accordance with the principles of the Declaration of Helsinki. We included in this study 12 eyes of 6 CHM female carriers with a mean age of 45.5 ± 16.3 years (range, 15–61 years) and 16 eyes of 8 age-matched control subjects (with a mean age of 46.6 ± 12.2 years; range, 18–54 years). The demographic characteristics of the patients and the controls with mutations are listed in Table 1. The BCVA was 20/20 in all patients and controls. The mean spherical equivalent was 0.09 ± 0.34 D (range 0.75 − 0.25) and −0.62 ± 1.30 D (range 1, −2.75) of the CHM female carriers and controls, respectively. The anterior segment was unremarkable except for the lens status: 5 patients (10 eyes) were phakic with no opacities, 1 patient (2 eyes) was pseudo-phakic. Goldmann visual field was normal in all patients. Fundoscopic examination revealed variable pigmentary alterations in the mid and far periphery in all the CHM female carriers. More specifically, in 3 patients, yellowish well-defined dots were present in the macular area and in 1 patient, RPE dystrophy characterized by well-defined hypopigmented RPE areas were also detectable in the macular area (Figure 1). We classified 2 CHM carriers as pattern A, 1 patient as pattern B, 2 patients as pattern C and 1 patient as pattern D in our series. The fundus pattern did not influence the vascular features obtained using OCT-A, in particular there was no relationship between the fundus pattern and OCTA choriocapillaris phenotype.

Table 1. The demographic characteristics of the patients and the controls.

1 SE: spherical equivalent; FAZS: superficial foveal avascular zone; FAZD: deep foveal avascular zone; RE: right eye; LE: left eye; VD: vessel density; SCP: superficial capillary plexus; DCP: deep capillary plexus; CC: choriocapillaris slab; CRT: central retinal thickness; CCT: central choroidal thickness; VA: visual acuity

The mean CRT was 220 ± 18.3 μm in CHM carriers and 227 ± 15.4 μm in controls (p =.342) whereas the mean CCT was 271 ± 54.2 μm and 275 ± 38.3 μm (p =.760) in the CHM carriers and control groups, respectively. The qualitative analysis of the OCT-A images revealed no statistically significant differences between the FAZ area of the patients and the age-matched control group. In fact, the FAZ of the SCP and DCP was 299.22 ± 33.79 and 293.92 ± 33.30, respectively in the CHM carriers; the mean FAZ of the SCP and DCP was 306.94 ± 37.96 and 320.72 ± 37.57, respectively in the control group. There were no statistical differences between the CHM carriers and controls (FAZ SCP, p =.681 vs FAZ DCP p =.212). Moreover, no significant differences were recorded in the mean VD of the SCP (108.31 ± 4.23 vs 106.89 ± 3.67; p =.437) and VD of DCP (102.828 ± 2.04 vs 102.700 ± 1.87; p =.859) between the study groups. Quantitative analysis of the CC slab on OCT-A showed a tendency toward reduction of the VD of the CC in the CHM carrier group (78.9 ± 13.97 vs 80.01 ± 10.86; p =.045).

In this study, we studied 6 CHM female carriers and compared them to age-matched control subjects in order to identify OCT-A based vascular abnormalities. OCT-A based methods represent a noninvasive way to study vascular features in CHM female carriers and could provide new insight on retinal and choroidal vasculature even in the absence of significative anatomical changes, as usually happens in these patients. So far, few reports have studied vascular abnormalities in CHM affected males and female carriers ([10]–[14]). More specifically, Abbouda et al. ([13]) did not find vascular abnormalities in CHM female carriers using OCT-A examinations, in comparison with a control group. However, they did not describe the CHM carrier fundus features. In fact, we know that variable pigmentary alterations and different degrees of retinal atrophy are the funduscopic hallmarks of CHM carriers ([3],[15],[16]). Furthermore, in the macular area and at the posterior pole, yellowish well-defined dots, resembling subretinal drusenoid deposits can be detected ([3]). Histopathologic studies show irregular thickness with depigmentation of the RPE, thickened Bruch's membrane and a thinned and hypocellular choroid in CHM carriers ([17],[18]). In our series we have included CHM female carriers who presented peripheral abnormalities but also RPE alterations in the macular area. Our comparative analysis between female carriers and healthy subjects revealed a preserved retinal vasculature and FAZ region. In fact, no differences were detected between the VD and FAZ area on OCT-A, both evaluated at the SCP and DCP levels, between the patients and control groups. Furthermore, we did not observe statistically significant differences in CRT between the female carriers and control group (p =.342). These data are in agreement with the results replaced by Abbouda et al ([13]). More specifically, the fundus pattern did not influence the vascular features obtained using OCT-A. In addition, we did not find statistically significant differences between the CCT of the CHM carriers and controls as previously reported ([3],[19]).

However, a trend toward reduction in the CC area was detectable in CHM carriers of our series compared to the control group (P =.045). The interpretation of this finding may be challenging. We know from previous histological studies that the Rab escort (REP-1) protein is normally expressed at the level of the retina, choroid and RPE, and loss of function of this protein causes slow degeneration of these structures ([20],[21]). Moreover, in most female carriers the severity of findings can vary greatly, with several retinal abnormalities probably due to unbalanced X-chromosome inactivation. Therefore, we can assume that a different degree of expression of the protein in ocular tissues such as the retina, RPE and choroid, dependent on random X chromosome inactivation, could be responsible for our results in the CHM carriers examined using OCT-A. A normal expression of this protein (REP-1) is thought to regulate both the growth of the retinal capillary network and CC vessels while a downregulation of the protein at the level of the RPE could lead to the peculiar RPE abnormalities frequently seen in these patients ([13]). Contrary to what was found in young CHM patients ([10]), where OCT-A revealed a significant impairment of the vessel density of the SCP, DCP and CC layers compared to the age-matched control group, the OCT-A parameters of the CHM female carriers (who often showed evident fundus abnormalities involving both RPE and the neuroepithelium, sometimes similar to the young CHM patients) were not different from controls. One of the possible interpretations for these important findings could be the secondary involvement of the choriocapillaris in the natural history of CHM in comparison with the RPE abnormalities and, in particular, the choriocapillaris would be the vascular layer involved before the retinal vascular layers. A recent report of 2 female CHM carriers followed for 11 and 17 years showed a slow progression of retinal and RPE abnormalities during the long – term follow-up ([22]). The tendency toward reduction of the VD of the CC VD detectable in the patients of our series (even though these results were not significant) may suggest a possible progressive choriocapillaris atrophy during the follow-up. In this hypothesis, OCT-A could be an important tool to characterize and quantify the extent of choriocapillaris atrophy during the follow-up. The limit in our study is the small sample of patients; however, we have to consider that CHM is a rare condition and we took only into consideration female CHM carriers.

In conclusion, OCT-A allows us to underline the role of the retinal pigment epithelium in the CHM pathophysiology. In our study, central inner retinal and choriocapillaris vascularization were preserved although the RPE was always involved in the CHM carriers: this could support a primary role of the RPE rather than vascular impairment in the natural history of choroideremia.

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.