Primary inflammatory choriocapillaropathies are a group of chorioretinal inflammatory disorders in which choriocapillaris (CC) dysfunction has been implied as major etiopathogenetic mechanism. The clinical spectrum includes acute posterior multifocal placoid pigment epitheliopathy (APMPPE), multiple evanescent white dot syndrome (MEWDS), multifocal choroiditis (MFC), serpiginous choroiditis (SC), and other rarer entities.1,2
APMPPE typically is a bilateral disorder affecting young adults and characterized by multiple cream color placoid lesions on posterior pole at the level of retinal pigment epithelium (RPE) and adjacent tissue. The fluorescein angiography (FA) shows early hypofluorescence and later hyperfluorescence, whereas indocyanine angiography (ICGA) shows hypofluorescence in all phases.3
SC is another spectrum of inflammatory choriocapillaropathies featuring ill-defined, grayish to creamy-yellow patches of choroidal inflammation in peripapillary region that creep centrifugally during the course of disease. Angiographic patterns follow the typical descriptions of inflammatory choriocapillaropathies.4
Vogt-Koyanagi-Harada (VKH) disease is autoimmune derived, multisystem inflammatory disorder primarily classified as stromal choroiditis. Ocular features include posterior uveitis with multifocal serous retinal detachment in acute phase and progress to fundus depigmentation during the course of disease. Areas of delayed choroidal perfusion, pinpoint leakages and multilobulated dye pooling are demonstrated by FA in acute VKH. However, in the chronic phase of disease, RPE window defects and areas of spotted hyper- and hypofluorescence have been described. ICGA shows delayed choroidal perfusion, choroidal vasculopathy, and choroidal hypofluorescence are seen during acute and chronic stages.5
Optical coherence tomography angiography (OCTA) is a newer imaging modality that can provide depth-resolved, high-resolution vascular details of ocular tissue at desired depth plan based on principles of motion contrast.6 As, it is newly introduced, OCTA features of many ocular diseases in general and uveitis in particular are not yet widely described.
Before the advent of OCTA, CC flow characteristics could not be studied independently. OCTA allows us to study the CC separately. The CC in healthy eyes appears as a homogenous fine sand-like texture.6 The aim of this study is to investigate the textural properties and flow characteristic of CC in inactive APMPPE, SC, and VKH disease.
Patients and Methods
In this retrospective, cross-sectional, observational study, we included patients with inactive SC, APMPPE, and VKH disease. The study followed the tenets of the Declaration of Helsinki for research involving human subjects. The institutional review board at Aziz Fatimah Medical College and SEHHAT foundation approved this study, which was conducted from September 2017 through March 2018.
Inclusion criteria were: A diagnosis of inactive SC, APMPPE, or VKH disease with macular involvement resolved at least 6 months before the inception of our study. Exclusion criteria were: Any signs of active disease on clinical examination, FAF, or OCT; any disease other than mentioned in inclusion criteria; history of systemic/ocular tuberculosis; any past ocular surgery; and media opacities.
All patients underwent baseline comprehensive ophthalmic examination to look for any disease activity.
All eyes were scanned using Angiovue, RTVue XR Avanti (Optovue, Fremont, CA), except two eyes that were scanned using DRI OCT Triton swept-source OCT (Topcon, Tokyo, Japan).
Avanti utilizes a light source centered on 840 nm with a bandwidth of 45 nm and A-scan rate of 70k Hz. Using split-spectrum amplitude decorrelation algorithm, angiograms are constructed from scanned volume of 304 × 304 A-scans with two consecutive B-scans at each location.7
Triton uses a light source of 1,050 nm wavelength at an A-scan rate of 100 kHz. Each volume scan consists of 320 B-scans each repeated four times and each B-scan containing 320 A-scans. OCTA ratio analysis algorithm is employed to generate OCT angiograms from volumetric data.7
A 3 mm × 3 mm volume scan centered on fovea was obtained and displayed as enface angiograms adapted to default machine segmentation of CC. Structural OCT B-Scans and enface shadowgrams were used to look for any shadowing or signal attenuation at CC from RPE changes in study eyes. OCT angiograms with obvious shadowing were excluded from further analysis. The CC images were then analyzed and qualitatively classified for texture changes and low flow or flow void areas.
Dark spots on CC segmentation were classified as low flow areas, whereas texture heterogeneity was defined as change in the uniform homogenous sandy textural appearance of choriocapillaris to coarse appearance or presence of irregular capillaris (Figure 1).
Choriocapillaris segmentation demonstrating textural properties. (a) Fine homogenous texture. (b, c) Coarse and heterogeneous texture.
Based on the extent of area involved, the low flow areas were categorized into having normal flow, mild low flow (less than one-quarter low flow area), moderate (one quarter to one half low flow area), and severe low flow (more than half of total area) (Figure 2).
Choriocapillaris flow characteristics. (a) Mild, low flow. (b) Moderate and (c) severe low flow.
Statistical analysis was performed using statistical software (SPSS software Version 21; SPSS, Chicago, IL). Data were presented as mean (standard deviation [SD]) for quantitative variables and as counts (percentage) for categorical variables. Flow characteristics across groups were tested with Kurskal-Wallis test and Dunn's pairwise comparison (Bonferroni adjustment). Correlation between the flow characteristics and texture with the interval from the control of inflammation was determined using Spearman's correlation coefficient test. A P value of less than .05 was considered significant for statistical tests.
Demographics and Clinical Data
A total of 28 eyes of 16 patients (10 male and six female) with a diagnosis of APMPPE, SC, or VKH were evaluated for texture and flow defects. Mean ± SD age was 44.2 years ± 7.5 years (range: 32 years to 59 years). Fourteen (50%) of the eyes had SC, whereas APMPPE and VKH disease affected 10 (35%) and four (14.3%) of the total study eyes, respectively.
Mean ± SD duration from resolution of active disease was 14.6 months ± 5.70 months (range: 6 months to 24 months). The mean ± SD duration between the resolution of active inflammation and imaging was 15.7 months ± 5.7 months for SC, 14.3 months ± 6.5 months in the APMPPE group, and 11.7 months ± 2.6 months for VKH disease.
None of the patients with a history of APMPPE had been taking any drugs at the time of study, whereas all of the patients with SC or VKH disease were on low-dose oral steroids, immunosuppressants, or both. None of the eyes showed any active inflammatory lesions on clinical examination or FAF and structural OCT. A summary of patient demographics and clinical data are provided in Table 1.
Distribution of Patient Demographics, Clinical Features, and OCTA ParametersAmong Study Groups
We conducted qualitative analysis of CC texture for identification of alterations in the homogenous appearance of the CC slab. Texture heterogeneity was noted in a total of 22 eyes (79%) (Figure 3). All four eyes (100%) with VKH disease demonstrated texture heterogeneity, whereas four (40%) APMPPE eyes and 14 (100%) SC eyes demonstrated an altered texture (P = .001).
Distribution of textural properties across disease groups. APMPPE = acute posterior multifocal placoid pigment epitheliopathy; SC = serpiginous choroiditis
The CC segmentations were graded for flow characteristics as normal flow, and mild, moderate or severe low flow areas for each eye. Most of the eyes with VKH disease showed severe flow alterations (75%) compared to the SC and APMPPE eyes (Figure 4). Most SC and APMPPE eyes showed mild low flow areas (64.3% and 50%, respectively). A summary of the distribution of flow parameter and textural properties across the study groups are shown in Table 1.
Distribution of flow void areas across study groups. APMPPE = acute posterior multifocal placoid pigment epitheliopathy; SC = serpiginous choroiditis; VKH = Vogt-Koyanagi-Harada
The severity of low flow areas across the disease groups was statistically different (P = .021). Dunn's pairwise tests provided a string evidence (P = .018) of the difference between eyes with APMPPE and VKH disease.
Duration from the resolution of active inflammation and severity of low flow areas was negatively correlated (Figure 5).
Relationship between time after healing and low flow. APMPPE = acute posterior multifocal placoid pigment epitheliopathy; SC = serpiginous choroiditis
There existed a significant negative correlation (r = −0.43; P = .02) between severity of low flow areas and heal duration. However, the negative correlation between heal duration texture heterogeneity was very weak and statistically nonsignificant (r = −0.08; P = .71).
In this study we have analyzed CC changes in APMPPE, SC, and VKH disease using 3 mm × 3 mm fovea-centered OCTA images. The higher resolution of OCTA provided greater details of CC segmentation. Although in settings of uveitides, pigmentary changes and alterations in retinal morphology may give rise to false low flow areas on CC segmentations; careful evaluation of structural B-scans and enface shadowgrams adapted to RPE and choriocapillaris alongside helps better differentiate nonperfusion from false low flow.8,9
In other studies, acute lesions of APMPPE and SC have shown CC low flow that were, in part, attributed to the reduced OCTA signal penetration through inflamed RPE and partly to true CC hypoperfusion; as the flow void areas did not always correspond to the areas of RPE thickening or pigmentary changes.10–13
Persistent CC flow defects have been described in healed lesions of APMPPE. However, cases with follow-up data have demonstrated a gradual progressive resolution of CC perfusion defects.10,12,13
Similar findings have been described in patients with active and resolved serpiginous choroiditis. A marked loss of choriocapillaris flow has been described in acute stages followed by partial restoration of CC flow.14,15 With the control of inflammation and healing of active lesions, the flow void areas are replaced initially with visibility of larger choroidal vessels and later partial restoration of CC flow with irregular choriocapillaris and a heterogeneous texture arises.4,11,14–16
To date, few data have been published on the OCTA features of VKH. Cennamo et al.17 described a patchy background with speckled appearance of CC segmentation on 8 mm × 8 mm OCTA scans. Also, Giannakouras et al.18 demonstrated CC hypoperfusion in a case with acute disease using swept-source OCTA system. In another study comparing OCTA based features of acute central serous retinopathy with acute VKH disease, CC dark areas seen were in a consistent correlation with hypocyanescent ICGA lesions, in absence of any loss of signal transmission and considered as true flow void.19 To the best of our knowledge, only one study described the CC properties on OCTA both during acute and healed stages. In the acute stage, all eyes showed a severe CC hypoperfusion corresponding to hypocyanescent ICGA lesions. Partial or incomplete resolution of CC perfusion was seen in healed stages.20
In our study, CC hypoperfusion was least severe in APMPPE eyes compared to eyes with SC and VKH disease. VKH eyes showed more severe persistent flow defects. All VKH and SC eyes had a heterogeneous texture in the areas of presumed healed lesions compared to only 40% APMPPE eyes showing texture heterogeneity. We hypothesize that the texture heterogeneity may arise due to the greater visibility of choroidal vasculature or appearance of an irregular capillaris in the areas of previous activity or both. The negative correlation between time and severity of flow impairment our study also suggests an improvement in CC perfusion with increasing time after the control of inflammation and healing of lesions, presumably due to reperfusion of previously inflamed CC or formation of newer choriocapillaris in areas of absolute flow void. However, persistent texture heterogeneity and presence of residual low flow areas in eyes with healed APMPPE, SC, and VKH disease suggests some permanent damage to the CC, which may or may not be restored during the long term.
The limitations of our study include smaller sample size, diverse nature and variable course of studied disorders, retrospective design, and lack of data from acute stages to quiescence and qualitative nature of the analysis.
In conclusion, OCTA was able to noninvasively visualize the CC flow alterations and texture changes in eyes with APMPPE, SC, and VKH disease even after the resolution of the inflammation. The findings suggest that despite the resolution of active inflammation, partial CC hypoperfusion and texture disruptions persist for longer durations and may resolve in a time dependent manner.
- Herbort CP. Primary Inflammatory Choriocapillaropathies. In: Zierhut M, Pavesio C, Ohno S, Orefice F, Rao NA, eds. Intraocular Inflammation. Berlin, Heidelberg: Springer Berlin Heidelberg; 2016:545–560. doi:10.1007/978-3-540-75387-2_43 [CrossRef]
- Bouchenaki N, Cimino L, Resta A, Fontana P, Tran VT, Herbort CP. Primary inflammatory choriocapillaropathies, a new indocyanine green angiography derived concept. Invest Ophthalmol Vis Sci. 2002;43(13):4266–4266.
- Maggio E, Alfano A, Polito A, Pertile G. Choroidal perfusion abnormalities associated with acute posterior multifocal placoid pigment epitheliopathy: A case report. BMC Ophthalmol. 2018;18(1):87. doi:10.1186/s12886-018-0756-8 [CrossRef]29631552
- Khan HA, Shahzad MA. Multimodal imaging of serpiginous choroiditis. Optom Vis Sci. 2017;94(2):265–269. doi:10.1097/OPX.0000000000001015 [CrossRef]
- Baltmr A, Lightman S, Tomkins-Netzer O. Vogt-Koyanagi-Harada syndrome – Current perspectives. Clin Ophthalmol. 2016;10:2345. doi:10.2147/OPTH.S94866 [CrossRef]
- Khan HA, Mehmood A, Khan QA, et al. A major review of optical coherence tomography angiography. Expert Review of Ophthalmology. 2017;12(5):373–385. doi:10.1080/17469899.2017.1356229 [CrossRef]
- Munk MR, Giannakaki-Zimmermann H, Berger L, et al. OCT-angiography: A qualitative and quantitative comparison of 4 OCT-A devices. PLoS One. 2017;12(5):e0177059. doi:10.1371/journal.pone.0177059 [CrossRef]28489918
- Pichi F, Sarraf D, Arepalli S, et al. The application of optical coherence tomography angiography in uveitis and inflammatory eye diseases. Prog Retin Eye Res. 2017;59:178–201. doi:10.1016/j.preteyeres.2017.04.005 [CrossRef]28465249
- Kayat KV, Roisman L, Zett C, Novais EA, Farah ME. Choriocapillaris hypoperfusion artifact in OCT angiography. Ophthalmic Surg Lasers Imaging Retina. 2018;49(8):603–610. doi:10.3928/23258160-20180803-08 [CrossRef]30114305
- Heiferman MJ, Rahmani S, Jampol LM, et al. Acute posterior multifocal placoid pigment epitheliopathy on optical coherence tomography angiography. Retina. 2017;37(11):2084–2094. doi:10.1097/IAE.0000000000001487 [CrossRef]28151840
- Mangeon M, Zett C, Amaral C, et al. Multimodal evaluation of patients with acute posterior multifocal placoid pigment epitheliopathy and serpiginous choroiditis. Ocul Immunol Inflamm. 2018;26(8):1212–1218. doi:10.1080/09273948.2017.1335757 [CrossRef]
- Burke TR, Chu CJ, Salvatore S, et al. Application of OCT-angiography to characterise the evolution of chorioretinal lesions in acute posterior multifocal placoid pigment epitheliopathy. Eye (Lond). 2017;31(10):1399–1408. doi:10.1038/eye.2017.180 [CrossRef]
- Klufas MA, Phasukkijwatana N, Iafe NA, et al. Optical coherence tomography angiography reveals choriocapillaris flow reduction in placoid chorioretinitis. Ophthalmol Retina. 2017;1(1):77–91. doi:10.1016/j.oret.2016.08.008 [CrossRef]31047399
- Desai R, Nesper P, Goldstein DA, Fawzi AA, Jampol LM, Gill M. OCT angiography imaging in serpiginous choroidopathy. Ophthalmol Retina. 2018;2(4):351–359. doi:10.1016/j.oret.2017.07.023 [CrossRef]
- Montorio D, Giuffre C, Miserocchi E, et al. Swept-source optical coherence tomography angiography in serpiginous choroiditis. Br J Ophthalmol. 2018;102(7):991–995. doi:10.1136/bjophthalmol-2017-310989 [CrossRef]
- Ahn SJ, Park SH, Lee BR. Multimodal imaging including optical coherence tomography angiography in serpiginous choroiditis. Ocul Immunol Inflamm. 2017;25(2):287–291. doi:10.1080/09273948.2017.1288824 [CrossRef]28281863
- Cennamo G, Romano MR, Iovino C, de Crecchio G, Cennamo G. Optical coherence tomography angiography in incomplete acute Vogt-Koyanagi-Harada disease. Int J Ophthalmol. 2017;10(4):661–662.28503445
- Giannakouras P, Andreanos K, Giavi B, Diagourtas A. Optical coherence tomography angiography: Employing a novel technique for investigation in Vogt-Koyanagi-Harada disease. Case Report Ophthalmol. 2017;8(2):362–369. doi:10.1159/000477611 [CrossRef]
- Aggarwal K, Agarwal A, Deokar A, et al. Distinguishing features of acute Vogt-Koyanagi-Harada disease and acute central serous chorioretinopathy on optical coherence tomography angiography and en face optical coherence tomography imaging. J Ophthalmic Inflamm Infect. 2017;7(1):3. doi:. Epub 2017 Jan 13. doi:10.1186/s12348-016-0122-z [CrossRef]28091938
- Aggarwal K, Agarwal A, Mahajan S, et al. The role of optical coherence tomography angiography in the diagnosis and management of Acute Vogt–Koyanagi–Harada Disease. Ocul Immunol Inflamm. 2018;26(1):142–153. doi:10.1080/09273948.2016.1195001 [CrossRef]
Distribution of Patient Demographics, Clinical Features, and OCTA ParametersAmong Study Groups
|Characteristic||APMPPE (10 Eyes)||SC (14 Eyes)||VKH Disease (4 Eyes)||Total (28 eyes)|
|Age, Years; Mean (±SD)[range]||40.0 (±6.3)[32–50]||45.9 (±8.1)[37–59]||48.8 (±0.5)[48–49]||44.2 (±7.5)[32–59]|
|Involvement, No. of Eyes (%)|
| Bilateral||10 (100)||12(86)||2 (50)||24 (86)|
| Unilateral||0 (0)||2 (14)||2 (50)||4 (14)|
| Texture, No. of Eyes (%)|
| Homogenous||6 (60)||0 (0)||0 (0)||6 (21)|
| Heterogeneous||4 (40)||14 (100)||4(100)||22 (79)|
|Low Flow Areas, No. of Eyes (%)|
| None||3 (30)||0 (0)||0 (0)||3 (11)|
| Mild||5 (50)||9 (64)||1 (25)||15 (54)|
| Moderate||2 (20)||4 (29)||0 (0)||6 (21)|
| Severe||0 (0)||1 (7)||3 (75)||4 (14)|