Vitelliform macular dystrophy, or Best disease, was first described by Friedrich Best in 1905.1 This disease is characterized by subretinal hyperautofluorescent vitelliform deposits in the posterior pole, resulting from the accumulation of lipofuscin between the retina pigment epithelium (RPE) and the neurosensory retina. Onset is usually in childhood and leads to visual loss. Burgess et al. described recently an autosomal recessive bestrophinopathy (ARB), caused by homozygous or compound heterozygous mutations in the BEST1 gene.2
The formation of choroidal neovascularization (CNV) is rare, with an estimated rate of 15%.3,4 It is a severe complication that must be suspected when a patient presents with a rapid deterioration of visual acuity (VA). An early diagnosis is important in order to promptly treat CNV because of its severe complications including hemorrhages, fluid exudation, and fibrosis resulting in photoreceptor damages, visual loss, and amblyopia in the pediatric population.
Fluorescein angiography (FA) and indocyanine green angiography (ICGA) are currently the methods of choice for diagnosing and classifying CNV. However, one of the major inconvenient is the difficulty to visualize and analyze CNV due to dye leakage and accumulation of subretinal material that can mask the vascular flow. Moreover, the administration of dyes can be difficult in child population. Optical coherence tomography angiography (OCTA) is a novel, noninvasive technology that can visualize the retinal and choroidal vascularization without administration of an intravenous dye. This imaging modality is faster than classical angiography and may be repeated frequently without side effects in a child population.
The present report describes a case series of pediatric cases of Best disease where OCTA contributed to the diagnosis of CNV and prompt treatment.
Patients and Methods
We retrospectively reviewed the records of patients younger than 18 years of age followed for neovascularized Best disease between November 2016 and November 2017 at Croix-Rousse University Hospital. Patients with other ophthalmological pathology were excluded.
To be enrolled in the study, patients underwent a complete ophthalmological examination including best-corrected VA (BCVA), biomicroscopy, and color fundus photographs (Topcon, Tokyo, Japan). Multimodal imaging included fundus autofluorescence (FAF), FA, ICGA, spectral-domain optical coherence tomography (SD-OCT) scans (Heidelberg Engineering, Heidelberg, Germany), and OCTA scans (Cirrus HD OCT Model 5000 with Angioplex or PlexElite; Carl Zeiss Meditec, Dublin, CA). OCTA is able to differentiate between static and dynamic structures by generating a decorrelation signal for visualization of vascular flow depth within a predetermined retinal area.
All patients included in this study had to meet the diagnostic criteria for Best disease, either autosomal-dominant or autosomal-recessive, as described by Burgess et al. in 2008.2 Patient characteristics including age, gender, symptoms, and medical history were collected. The findings from OCTA were collected for both the time of CNV diagnosis and during follow-up.
CNV diagnosis was made based on multimodal imaging findings, including leakage on FA, neovascular network on ICGA, or visualization of a neovascular membrane on OCTA.
Patient Characteristics and Best Disease History
Five eyes of three patients were included in the present study. The mean age at diagnosis was 6.8 years ± 1.8 years (range: 5 years to 10 years). All patients were girls and Caucasian. Two of the patients had an autosomal dominant Best disease and one had a recessive form. Two of the three cases had a bilateral CNV. At baseline, mean BCVA was 0.92 logMAR ± 0.12 logMAR corresponding to a Snellen equivalent of 20/160 (Table 1).
Demographics and Main Clinical Characteristics of Patients Included in the Study
Case 1 had an autosomic dominant Best disease for 1 year with foveal vitelliform deposit and a BCVA of 20/25 in both eyes. She presented with visual loss in her both eyes. BCVA was 20/160 in the right eye and 20/200 in the left eye.
Case 2 had no history of Best disease and presented with visual loss in her left eye for 6 months (BCVA: 20/100).
Case 3 also had no history of Best disease. She presented with visual loss in her left eye for 3 days (BCVA: 20/200 in both eyes). We diagnosed Best disease at the same time as CNV in two of the three patients.
CNV Diagnosis and Treatment Outcomes
CNV was diagnosed based on FA, ICGA, and OCTA. FA and ICGA showed a clear neovascular membrane in two eyes (case 2 in the left eye, and case 3 in right eye), whereas OCTA showed a neovascular membrane in all eyes. In eyes where angiography did not show clearly the neovascular membrane, pinpoints on FA were seen in one eye (case 3 in the left eye) and a hot spot on ICGA was seen in one eye (case 1 in the right eye). The remaining case showed subretinal pooling on FA without a clear neovascular feature on ICGA. B-scan OCT showed a subretinal hyporeflectivity in all cases, with intraretinal cysts in three eyes (case 1 in the right eye, case 3 in both eyes) (Figure 1).
Multimodal imaging at baseline. Fluorescein angiogram (FA), indocyanine green angiography (ICGA), optical coherence tomography angiography (OCTA), and B-scan OCT of the five eyes. (A) Case 1 right eye. (B) Case 1 left eye. (C) Case 2 left eye. (D) Case 3 right eye. (E) Case 3 left eye. FA and ICGA showed a clear neovascular membrane in two eyes (C, D). Pinpoints on FA were seen in one eye (E) and a hot spot on ICGA was seen in one eye (A). Subretinal pooling on FA without a neovascular membrane on ICGA was seen in one eye (B). OCTA demonstrated a well-circumscribed choroidal neovascularization with the typical “sea fan” shape in Case 1 (A, B). In Case 2, several long filamentous linear vessels with some anastomoses and loops were seen (C). The neovascular membrane was also clearly demonstrated in Case 3 images (D, E).
All cases were treated with intravitreal injections of anti-vascular endothelial growth factor (bevacizumab; Avastin; Genentech, South San Francisco, CA) with good anatomical and functional responses. Case 1 received four injections of bevacizumab in her right eye and six injections in her left eye. Case 2 was treated with five injections of bevacizumab in her left eye. Case 3 received three injections of bevacizumab in her right eye and two injections in her left eye.
In case 1, after a follow-up of 14 months, VA improved to 20/50 in the right eye and 20/63 in the left eye. VA also increased in case 2, with a final VA of 20/40 in her left eye after 24 months. Case 3 had a follow-up of 7 months, with a final VA of 20/100 in her right eye and 20/125 in her left eye.
VA improved in all eyes, with a mean final BCVA of 0.54 logMAR ± 0.19 logMAR, corresponding to a Snellen equivalent of 20/80 to 20/63 and representing a mean gain of 0.38 logMAR ± 0.12 logMAR. In all cases, SD-OCT scans showed a regression of the retinal fluid, and OCTA revealed a reduction of the CNV (case 1 did not beneficed of OCTA nor FA/ICGA during follow-up). A small persistent subretinal fluid was present in case 3 (both eyes) closed to the abrupt subfoveal lesion (Figure 2).
B-scan optical coherence tomography (OCT) and corresponding infrared imaging or OCT angiography (OCTA) at the end of follow-up. (A) Case 1 right eye. (B) Case 1 left eye. (C) Case 2 left eye. (D) Case 3 right eye. (E) Case 3 left eye. Serous retinal detachment and intraretinal cysts were totally absent in Case 1 (right and left eyes) and Case 2 (left eye). In Case 3, there was a chronical intraretinal cystic change in the right eye associated with a chronic, abrupt, tent-shaped subretinal detachment in both eyes that did not respond to anti-vascular endothelial growth factor injection. OCTA, when realized in patient, showed a decrease (C) or a stabilization of the neovascular membrane (D, E). However, OCTA was not performed during the follow-up in Case 1 (A, B).
Autosomal dominant Best disease and ARB are rare. Despite macular involvement, patients preserve good VA. Therefore, a neovascular complication should be discussed in case of sudden visual loss. The current gold standard for diagnosing CNV is FA and ICGA. However, these methods have multiple disadvantages. First of all, these are invasive techniques that require the administration of dyes, leading to a risk of side effects.5,6 The long-term side effects of FA in children are still unknown. Moreover, late hyperfluorescence of vitelliform material can mask the vascular flow or be misdiagnosed as a neovascularization.
OCTA seems to be a good alternative for diagnosing CNV, especially with children, as we showed in this case series. It is a noninvasive, rapid technique for imaging and does not require the administration of dyes. It allows visualization of the vascular flow, which is typically masked by the vitelliform material. It provides a three-dimensional map and en face visualization, whereas FA is limited to two dimensions, causing an overlap of superficial and deep capillary plexuses. Furthermore, with OCTA, there is no dye leakage, allowing us to better visualize and analyze CNV. OCTA can be repeated frequently and used as a routine examination.
In our three cases, we doubted the presence of CNV on FA because of dye leakage and accumulation of material, masking the vascular flow. OCTA scans passing throw the RPE illustrated a well-circumscribed CNV with the typical “sea fan” shape; several long filamentous linear vessels; and some anastomoses, loops, and arcades at the vessels' termini. It helped us for the diagnosis of CNV when FA imaging was difficult to analyze.
Limitations of using OCTA include the difficulty of fixation in very young children, the potential artifacts, and the impossibility to identify the breakdown of the blood-retinal barrier (possible with dye leakage).
Moreover, OCTA does not evaluate neovascular activity as we do not currently know the signs that demonstrate it. B-scan OCT allows a better assessment of disease activity than OCTA.
De Carlo et al. have demonstrated the high sensitivity and specificity of OCTA in detecting CNV in cases of age-related macular degeneration, central serous chorioretinopathy, and other different diseases.7 We can presume the same findings in Best disease, but further analysis is required to confirm this hypothesis.
To conclude, OCTA is an attractive alternative to angiography in children to detect CNV, guide treatment decisions, and assess response to therapy.
- Best F. Uber eine hereditare maculaafektion; Beitrage zur verergslehre. Zschr Augenheilk. 1905;13:199–212.
- Burgess R, Millar ID, Leroy BP, et al. Biallelic mutation of BEST1 causes a distinct retinopathy in humans. Am J Hum Genet. 2008;82(1):19–31. doi:10.1016/j.ajhg.2007.08.004 [CrossRef]18179881
- Miller SA, Bresnick GH, Chandra SR. Choroidal neovascular membrane in Best's vitelliform macular dystrophy. Am J Ophthalmol. 1976;82(2):252–255. doi:10.1016/0002-9394(76)90428-1 [CrossRef]949077
- Glacet-Bernard A., Soubrane G, Coscas G. [Macular vitelliform degeneration in adults. Retrospective study of a series of 85 patients]. [Article in French]. J Fr Ophtalmol. 1990;13(8–9):407–420.
- Kwiterovich KA, Maguire MG, Murphy RP, et al. Frequency of adverse systemic reactions after fluorescein angiography: Results of a prospective study. Ophthalmology. 1991;98:1139–1142. doi:10.1016/S0161-6420(91)32165-1 [CrossRef]1891225
- López-Sáez MP1, Ordoqui E, Tornero P, et al. Fluorescein-induced allergic reaction. Ann Allergy Asthma Immunol. 1998;81(5):428–430. doi:10.1016/S1081-1206(10)63140-7 [CrossRef]9860035
- de Carlo TE, Bonini Filho MA, Chin AT, et al. Spectral-domain optical coherence tomography angiography of choroidal neovascularization. Ophthalmology. 2015;122(6):1228–1238. doi:10.1016/j.ophtha.2015.01.029 [CrossRef]25795476
Demographics and Main Clinical Characteristics of Patients Included in the Study
|Case||Cage/Sex||Origin||Best Disease||Baselined BCVA (Snellen)||Duration of Best Disease (Months)|