Best Disease Presenting as Subretinal Pigment Epithelium Hyperreective Lesion on Spectral-Domain Optical Coherence Tomography: Multimodal Imaging Features Multimodal imaging in Best disease

Purpose To report clinical and multimodal imaging features of Best disease in patients presenting with subretinal pigment epithelium hyperreective lesion. Design Retrospective study. Methods Clinical examination ndings and multimodal imaging features, including color fundus photography, spectral-domain optical coherence tomography (SD-OCT), fundus autouorescence, uorescein and indocyanine green angiography (ICGA), and optical coherence tomography angiography (OCTA) images were evaluated retrospectively. Results We assessed 27 eyes of 16 patients with the diagnosis of Best disease. Only patients presenting with serous macular detachment and subretinal pigment epithelium hyperreective lesion in one or both eyes were included in this study. In 17 of 27 eyes (63%), brosis was identied by multimodal imaging techniques. Although there was no sign of active neovascularization on fundus examination or SD-OCT, a vascular network could be identied in 7 eyes (26%) (in 1 eye with OCTA only and in 6 with both OCTA and ICGA). Active neovascularization was seen in 3 eyes (11%). Treatment was recommended for eyes with active neovascularization, and follow-up was scheduled for eyes with quiescent neovascularization and brosis. Eyes and show brosis, quiescent neovascularization, or active neovascularization. Multimodal imaging techniques are very important for differentiation of these lesions. by split-spectrum amplitude-decorrelation angiography (SSADA) an visualize retinal and vasculature a 6 mm the retina. retina both and with correction for Automatic segmentation of retinal layers performed by to generate en face projection images of signal poor-quality diagnosis was conrmed by hyperautouorescent material on fundus autouorescence imaging in both an abnormal EOG with a low Arden ratio, and presence of a family history in the remaining 14 patients. Presence of hemorrhage on fundus examination, intraretinal uid or subretinal hyperreection above the RPE elevation or irregular RPE elevation on SD-OCT with or without increased hyperuorescence on uorescein angiography, and an early network with late staining plaque on ICGA were considered signs of active neovascularization. Only the presence of a vascular network on OCTA or FA and ICGA without signs of active neovascularization was accepted as quiescent neovascularization. Serous macular detachment was not accepted as a sign of active neovascularization because it could also be a primary result of Best disease. The maximum height of a hyperreective lesion was dened as the maximum distance between the RPE and Bruch’s membrane. Manual calipers provided with the software of the device were used for measurement. Values for each measurement were and vertical scans, and values were averaged Eyes with an any evidence of other retinal disorders, a history of previous treatment with antivascular endothelial growth factor, or photodynamic treatment were excluded from the study. Demographic and clinical data were collected from the medical chart, and data for all 27 eyes were used for statistical analysis. Descriptive statistical methods (mean, standard deviation) were used for characteristics such as age and sex. Wilcoxan signed rank test was used to compare rst and last examination ndings. Measurement values of the groups for continuous variables were compared using the Kruskal-Wallis test or Mann-Whitney U test and for categorical variables were compared using chi-square test. Statistical analyses used SPSS Version 20.0 (SPSS Inc, Chicago, IL, USA). P<0.05 was considered statistically signicant. eyes with quiescent neovascularization showed a late-phase ill-dened hyperuorescence on FA without pooling of dye, and early vascular network and late staining hyperuorescent plaque on ICGA. Two eyes with active neovascularization showed late-phase hyperuorescence related to leakage on FA, and early vascular network and late staining hyperuorescent plaque on ICGA. OCTA was performed in all eyes at least one examination. A neovascular complex was demonstrated on the OCTA en face projection image and ow signal on the cross-sectional OCTA in 3 eyes with active and in 7 eyes with quiescent neovascularization. central serous chorioretinopathy or age-related macular degeneration and had previously received multiple treatments, which were refractory to therapy and then therapies were deferred; 4 patients were then not treated. At last follow-up, patients’ visual acuity and retinal status remained unchanged. None of the eyes in their study displayed any signs of retinal hemorrhage or intraretinal uid. In our study 13 of 27 eyes with sub-RPE hyperreectivity and serous macular detachment were followed up for 1 to 11 years after diagnosis. Similar to the previous report [3], visual acuity and retinal lesions were stable in 12 of 13 eyes (92%) at last follow-up.


Introduction
Vitelliform macular dystrophy, also called Best disease, is an inherited retinal dystrophy caused by mutation in the BEST1 gene located on chromosome 11q12-q13. Although ve stages have been described, including previtelliform (stage I), vitelliform (stage II), pseudohypopyon (stage III), vitelliruptive (stage IV), and atrophic (stage V), the disease does not progress through each of these stages in every individual. [1][2] Indeed, the vitelliform stage, in which the macula has a characteristic yellowish appearance due to lipofuscin accumulation at the level of the retinal pigment epithelium (RPE), is frequently not present at the time of examination and sometimes it never occurs. [1][2][3] Especially during the vitelliform stage, diagnosis of Best disease is simple. However, during late stages, con rming the diagnosis by multimodal imaging techniques such as optical coherence tomography (OCT) and fundus auto uorescence is very useful. [4] Subretinal brosis and neovascularization are known as late-stage complications of the disease. Fibrotic scarring and choroidal neovascularization have similar presentation and they should be distinguished from each other. [3][4] Optical coherence tomography angiography (OCTA), which has become increasingly a useful and minimally invasive tool for monitoring new vessels, is a preferred imaging technique along with uorescein angiography (FA) and indocyanine green angiography (ICGA). [5][6] An abnormal electrooculogram (EOG) with a low Arden ratio and family history are helpful in supporting the diagnosis. [1,3] Although the appearance of lipofuscin accumulation between the RPE and the outer retina is very characteristic of Best Disease, presence of hyperre ective material between the RPE and Bruch's membrane, with serous macular detachment, may be the rst presenting nding in some patients. The diagnosis can be controversial in such cases. In this study we aimed to report clinical and multimodal imaging features of patients with Best disease presenting with a hyperre ective RPE elevation and serous macular detachment.

Methods
Patients with Best disease who had been diagnosed at Istanbul Retina Institute between January 2008 and December 2019 were evaluated retrospectively. Twenty-seven eyes of 16 patients presenting with a hyperre ective RPE elevation and serous macular detachment associated with Best disease were included in this study. Written informed consent was obtained prior to the diagnostic and therapeutic procedures. The study protocol was approved by the Institutional Review Board of Şişli Memorial Hospital, Istanbul. The study adhered to the tenets of the Declaration of Helsinki.
The initial diagnosis was based on clinical examination and spectral-domain OCT (SD-OCT) assessment. If fundus examination revealed an elevated hyperre ective lesion, further examination by OCT was required. Presence of hyperre ective material between the RPE and Bruch's membrane with serous macular detachment on SD-OCT were considered as inclusion criteria.
For all patients, medical record data, including measurement of best corrected visual acuity (BCVA) by the Early Treatment Diabetic Retinopathy Study chart, refraction, intraocular pressure measured by applanation tonometer, detailed anterior segment and fundus examination by slit-lamp biomicroscopy with a 90 diopter noncontact lens, and multimodal imaging features, were collected and analyzed. The Spectralis OCT system (Heidelberg Engineering, Heidelberg, Germany) was used for fundus auto uorescence, macular scans, digital uorescein angiography, and ICGA, and images were obtained by a single examiner experienced in performing scans using the SD-OCT device.
Patients identi ed by SD-OCT as having serous macular detachment and subretinal hyperre ective material underwent OCTA between March 2017 and November 2019. OCTA images were obtained by using the AngioVue (Optovue) OCTA system on a commercially available SD-OCT device (RTVue XR Avanti, Optovue, Fremont, CA, USA). The OCTA images were evaluated by using split-spectrum amplitude-decorrelation angiography (SSADA) and an algorithm to visualize retinal and choroidal vasculature in a 6 x 6 mm volume of the central part of the retina. The retina was scanned in both horizontal and vertical axes, with correction for movements of the eyes. Automatic segmentation of the retinal layers was performed by the viewing software and was used to generate en face projection images of the lesion. In the case of segmentation errors, the segmentation lines were manually adjusted to evaluate the lesion adequately. Images with signal strength index less than 40 were considered poor-quality images and excluded from the study.
The diagnosis was con rmed by hyperauto uorescent material on fundus auto uorescence imaging in both eyes, an abnormal EOG with a low Arden ratio, and presence of a family history in the remaining 14 patients. Presence of hemorrhage on fundus examination, intraretinal uid or subretinal hyperre ection above the RPE elevation or irregular RPE elevation on SD-OCT with or without increased hyper uorescence on uorescein angiography, and an early network with late staining plaque on ICGA were considered signs of active neovascularization. Only the presence of a vascular network on OCTA or FA and ICGA without signs of active neovascularization was accepted as quiescent neovascularization. 5 Serous macular detachment was not accepted as a sign of active neovascularization because it could also be a primary result of Best disease. The maximum height of a hyperre ective lesion was de ned as the maximum distance between the RPE and Bruch's membrane. Manual calipers provided with the software of the device were used for measurement. Values for each measurement were obtained from horizontal and vertical line scans, and values were averaged Eyes with an any evidence of other retinal disorders, a history of previous treatment with antivascular endothelial growth factor, or photodynamic treatment were excluded from the study. Demographic and clinical data were collected from the medical chart, and data for all 27 eyes were used for statistical analysis.
Descriptive statistical methods (mean, standard deviation) were used for characteristics such as age and sex. Wilcoxan signed rank test was used to compare rst and last examination ndings. Measurement values of the groups for continuous variables were compared using the Kruskal-Wallis test or Mann-Whitney U test and for categorical variables were compared using chi-square test. Statistical analyses used SPSS Version 20.0 (SPSS Inc, Chicago, IL, USA). P<0.05 was considered statistically signi cant.

Results
Twenty-seven eyes of 16 patients were appropriate to include in this study. Twelve (75%) were male. The age at rst examination was 22.7±12.6 years (range, 7-44 years). The mean initial logMAR visual acuity was 0.44±0.31 (range, 1-0 logMAR). The mean initial height of hyperre ective lesion was 289.8±183.4 mm (range 100 to 791 mm). Detailed demographic and clinical characteristics of patients are shown in Table 1.
At presentation, bilateral serous macular detachment and a hyperre ective lesion in the sub-RPE were identi ed in 11 patients (69%). In the remaining 5 patients, vitelliform-stage disease was found in one patient's fellow eye (6%) (patient 10), and vitelliruptive-stage disease in 4 patients' fellow eyes (25%) (patients 2, 9, 12, and 14). After detailed examination of patients with hyperre ective lesion, active neovascularization was identi ed in 2 eyes of 2 patients (patients 5 and 7) by both dye-based angiography and OCTA. Quiescent neovascularization was detected in 3 eyes of 2 patients (patients 3 and 11) by dyebased angiography alone, and was identi ed in 3 eyes of 2 patients (patients 7 and 15) by both dye-based angiography and OCTA. However, 3 patients (patients 1, 2, and 6) did not have dye-based angiography or OCTA at the rst examination, so presence of a vascular network inside the hyperre ective lesion could not be evaluated in 5 eyes, but OCTA assessment was performed at the nal examination. No neovascular network was detected in 3 eyes of 2 patients (patients 1 and 2), but neovascularization was identi ed in both eyes of patient 6 ( Figure 1).
Thirteen eyes of 8 patients were followed up for 1 to 11 years after diagnosis. Neovascularization developed in only one eye (patient 6) after 6 years of followup and the hyperre ective lesion was stable in the remaining 12 eyes. At last follow-up, the mean logMAR visual acuity was 0.45±0.27 (range, 1-0 logMAR) for those stable 12 eyes and there was no statistical difference in visual acuity between presentation and last follow-up (P=0.60). The mean height of the hyperre ective lesion was 333±3 mm (range 170 to 650 mm), and there was also no statistical difference in the height of the lesion between presentation ((330±7 mm (range 175 to 650 mm) and last follow-up (P= 0.09). As expected, the logMAR visual acuity had decreased, from 0.1 to 0.5, and the height of lesion had increased, from 100 to 128 mm, in the eye developing neovascularization. Figures 2 and 3 show images from patients 14 and 2 to indicate how the patient's retinal status remained unchanged during 1-year and 11-year follow-up, respectively.
At last evaluation, brosis was identi ed in 17 eyes (63%) by multimodal imaging. Although there was no sign of active neovascularization on fundus examination or SD-OCT, a vascular network could be identi ed in 7 eyes (26%) (in 1 eye only with OCTA and in 6 eyes with both OCTA and ICGA). Active neovascularization was seen in 3 eyes (11%). Treatment was recommended for eyes with active neovascularization, and follow-up was scheduled for eyes with quiescent neovascularization and brosis. Figure 4 shows images from patient 7 to demonstrate quiescent neovascularization in 1 eye and active neovascularization in the fellow eye.

Multimodal imaging features of patients
On fundus examination, all eyes demonstrated serous macular detachment, vitelliform accumulation at the posterior pole, and central macular elevation. In eyes with active neovascularization, hemorrhage was also seen. On SD-OCT, regular and hyperre ective RPE elevation in 24 eyes (89%), and a hypore ective subretinal space were identi ed in all eyes (100%). Outer retinal damage was seen in 7 eyes (26%) of 6 patients, and choroidal excavation in 8 eyes (30%) of 7 patients. However, all 3 eyes with active neovascularization had intraretinal uid and irregular RPE elevation, and 2 eyes had also subretinal hyperre ectivity above the RPE elevation. Fundus auto uorescence images identi ed hyperauto uorescence due to lipofuscin accumulation at the posterior pole in all eyes. The hyperre ective RPE elevations were hypoauto uorescent due to brosis and scarring in 21 eyes (78%) and related to neovascularization in 3 eyes (11%), and were hyperauto uorescent in the remaining 3 eyes (11%). Dye-based angiography was performed in 17 eyes (65%). Nine eyes with only regular and hyperre ective RPE elevation demonstrated early hypo uorescence and late hyper uorescence related to staining on FA and hypo uorescence on ICGA. Six eyes with quiescent neovascularization showed a late-phase ill-de ned hyper uorescence on FA without pooling of dye, and early vascular network and late staining hyper uorescent plaque on ICGA. Two eyes with active neovascularization showed late-phase hyper uorescence related to leakage on FA, and early vascular network and late staining hyper uorescent plaque on ICGA. OCTA was performed in all eyes at least one examination. A neovascular complex was demonstrated on the OCTA en face projection image and ow signal on the cross-sectional OCTA in 3 eyes with active and in 7 eyes with quiescent neovascularization.

Discussion
In Best disease, the appearance of hyperre ectivity between the RPE and Bruch's membrane has been identi ed as brotic pillar, brotic nodule, brotic scar, or sub-RPE deposition in previous studies. [7][8][9][10] In the study by Kumar et al [7], 38 eyes of 19 patients with Best disease were evaluated, and 12 showed brotic pillars on OCT. It was reported that a subretinal hypore ective space appeared in the pseudo-hypopyon stage and increased thereafter in the natural history of the disease. Then, brosis of subretinal vitelliform material occurred and localized vitelliform deposits and their brosis caused brotic pillars over time. [7] However, the vitelliruptive phase led to atrophy and brotic scarring, and brotic pillars occurred after a period of time. In our study, interestingly, those lesions were the presentation nding in 9 eyes of 5 patients at the mean age of 9.4 years (range, 7-13). In previous studies, vitelliform-stage disease was not detected at the time of examination, and sometimes did not occur, which is compatible with our nding.
It is known that these lesions have a relatively good prognosis. In the study of Zatreanu et al [3], 26 eyes of 13 patients with serous macular detachment with or without subretinal hyperre ective material were evaluated. Nine patients were incorrectly diagnosed as have central serous chorioretinopathy or age-related macular degeneration and had previously received multiple treatments, which were refractory to therapy and then therapies were deferred; 4 patients were then not treated. At last follow-up, patients' visual acuity and retinal status remained unchanged. None of the eyes in their study displayed any signs of retinal hemorrhage or intraretinal uid. In our study 13 of 27 eyes with sub-RPE hyperre ectivity and serous macular detachment were followed up for 1 to 11 years after diagnosis. Similar to the previous report [3], visual acuity and retinal lesions were stable in 12 of 13 eyes (92%) at last follow-up.
One issue in Best disease is active neovascularization. As the vitelliruptive phase leads to atrophy and brotic scarring, hemorrhage, either isolated or in association with neovascularization, may simulate the appearance of scar tissue. [4] However, active neovascularization can be differentiated from these lesions indirectly according to good response of treatment, or on multimodal imaging features such as hemorrhage on fundus image, intraretinal uid on SD-OCT, increased hyper uorescence related to leakage on FA, early network and late staining hyper uorescent plaque on ICGA, and neovascular complex and ow on OCTA. These are helpful for correct diagnosis. In the study by Khan et al [11], 14 eyes of 12 patients with neovascularization complicating BEST-1related retinopathy were identi ed. It was reported that active treatment with intravitreal anti-VEGF agents was associated with better functional outcomes than observation alone. In our study, active neovascularization was identi ed in 3 (11%) eyes on multimodal imaging and treatment was recommended. But our follow-up examination and treatment results are not available.
The similar appearance of sub-RPE epithelium hyperre ective material and quiescent neovascularization on fundus image and SD-OCT image is another issue. In some previous studies, the hyperre ective material between the RPE and Bruch's membrane was described as neovascularization. [3,12,13] However, not all highly re ective RPE elevations have an accompanying vascular network, and quiescent neovascularization can be distinguished from both these hyperre ective lesions easily by dye-based angiography and OCTA. In our study, in 17 of 27 eyes (63%), no neovascular network was seen on OCTA or ICGA and all those eyes were identi ed as brosis. Seven eyes (26%) with no hemorrhage on fundus image or no sign of activity except macular detachment on SD-OCT were classi ed as quiescent neovascularization based on OCTA and /or ICGA. No unnecessary treatment was offered for those eyes.
Kumar et al [7] concluded that histopathological examination of eyes in vitelliruptive stages also had brous scarring in accordance with the OCT appearance of the brotic pillar. [7,14] It is also known that vitelliform lesions show hyperauto uorescence, but areas of atrophy and brosis show low or absent auto uorescence signal on fundus auto uorescence imaging. [9] In the light of this information, the hyperre ectivity between the RPE and Bruch's membrane on SD-OCT can be classi ed as brotic pillar, brotic nodule, or brotic scar if the lesion is hypoauto uorescent on fundus auto uorescence imaging and if there is no nding of neovascularization. In our study, the hyperre ective lesion was hypoauto uorescent on fundus auto uorescence image in 20 of 24 eyes with regular and hyperre ective RPE elevation.
Fibrotic pillars also play a role in the formation of focal choroidal excavations. Kumar et al [7] reported that 8 of 12 eyes (66%) with brotic pillar had focal choroidal excavation directly beneath (33%) or in close approximation (33%) to the brotic pillar. It was concluded that brotic pillars lead to development of focal choroidal excavation. In our study, we also found focal choroidal excavation directly beneath the brotic pillar in 8 eyes (30%), a similar result.
In conclusion, in Best disease, hyperre ective RPE elevation on SD-OCT may represent brosis, quiescent neovascularization, or active neovascularization, and multimodal imaging techniques are very important for differentiation of these lesions.

Declarations
Disclosure and Acknowledgments a.Funding/Support: This research received no speci c grant from any funding agency in the public, commercial, or not-for-pro t sectors.
b.Financial Disclosure: No payment or services have been received from a third party for any aspect of the submitted work, including design, data collection, analysis, or interpretation of the data, writing of the report, or in the decision to submit the article for publication