Choroidal neovascularization (CNV) is a severe complication of different retinal diseases. Until 4 years ago, the detection of CNV vascular details was only possible through invasive dye injection, such as fluorescein angiography (FA) or indocyanine green (ICG) angiography.1,2 Optical coherence tomography angiography (OCTA) is a recent imaging technique that generates three-dimensional microvascular angiograms in vivo without dye injection.
OCTA may distinguish blood flow from static tissue in healthy eyes, as shown in previous studies.3–5 Furthermore, the visualization of CNV with split-spectrum amplitude-decorrelation angiography algorithm (SSADA) has been demonstrated with swept-source OCT (SS-OCT) and spectral-domain OCT (SD-OCT).6–9
The exact pathophysiology of CNV is complex and not yet completely understood, involving genetic factors and other processes, such as oxidative stress, inflammation, and hypoxia.10,11
Xu et al. demonstrated an association between decreased foveolar choroidal circulation and the development of AMD and CNV.12 The mechanism by which these age-related macular degeneration (AMD) risk factors and blood flow parameters affect each other remains unclear.
Before anti-vascular endothelial growth factor (VEGF) treatment we normally observe in almost all cases of CNV a large dark area around the new vessels. This low “decorrelated” signal area seems to coincide with an area of low choriocapillaris flow.
A dark perilesional of flow decrease has been frequently reported. Jia et al., for the first time, described areas of reduced choroidal flow adjacent to the CNV.7 Moult et al.13 and Coscas et al.2 described a dark perilesional area around CNV in approximately 80% of cases around CNV on OCTA. Lumbroso et al.14 hypothesized that choroidal hypoperfusion might be related to outer retinal ischemia, resulting in CNV development. They theorized that the dark area represents the affected zone of the choriocapillaris near the CNV. Several studies have identified abnormalities of choroidal flow associated with CNV in AMD.15–21
McLeod et al.22 and Lutty et al.23 provided histologic evidence of choriocapillaris atrophy close to the CNV. They reported intact choriocapillaris in geographic atrophy. Further, they observed choriocapillaris alteration consistently extending beyond the margins of CNV in patients with exudative AMD. Moult et al.13 and Lumbroso et al.14 similarly confirmed histopathology results by means of OCTA.
The aim of our study is to assess the “dark halo” area evolution before and after anti-VEGF treatment.
Patients and Methods
This research adhered to the tenets of the Declaration of Helsinki. Written informed consent was obtained from all of the involved patients.
In this prospective, observational, noncomparative case series, the RTVue XR-Avanti (Optovue, Fremont, CA) was used to perform OCTA of the retinal and choriocapillaris microvasculature in 43 examinations of 11 eyes with naïve or treated CNV. According to the classification of CNV based on OCT assessment, we observed four cases were CNV type 1 and seven were CNV type 2.
All patients were white. The mean age was 70.54 years ± 7.5 years. Two patients were affected by high blood pressure with good compensation by B-blockers. No other systemic disease has been reported.
The inclusion criteria were eyes with AMD with active CNV treatment-naïve or under treatment with anti-VEGF drugs: ranibizumab (Lucentis; Genentech, South San Francisco, CA) or aflibercept (Eylea; Regeneron, Tarrytown, NY). The exclusion criteria were as follows: any concomitant ocular disorder that might confuse the interpretation of the results; visual acuity reduction requiring surgical procedure including cataracts, retinal detachment, macular hole, etc; any use of eye drops, such as anti-glaucomatous or anti-inflammatory drugs for ocular diseases; any intraocular surgery within 3 months prior to enrollment; inability to comply with study procedures; and poor quality images due to inadequate cooperation.
CNV type labeling was determined on OCT B-scans as type 2 CNV if retinal edema and no retinal pigment epithelium (RPE) elevation were observed and type 1 if hyperreflective subretinal material anterior to an elevated, disrupted RPE was seen.24
The mean observation period ranged from 12 months to 24 months to understand the dark halo behavior.
The OCTA examination was planned to observe closely and precisely the early evolution of neovascularization after intravitreal aflibercept or ranibizumab injection between 6 days and 14 days after injection. In real life, it was not possible to examine all of the patients after exactly 14 days. We obtained measurements for 43 cycles before and after treatment in 12 eyes of 11 patients. This time strategy was selected to avoid possible CNV recurrence phenomena 3 weeks or more after the injection and late evolution features seen 30 days to 50 days after injection. We did not select early images 24 hours following treatment.
OCT Angiography Method Details
To perform OCTA, we used an ultrahigh-speed, wide-field, en face, high-resolution (70,000 A-scans/second), 840-nm wavelength OCT device (the XR-Avanti), based on a split-spectrum amplitude-decorrelation angiography algorithm. The macular angiography scan protocol covered a 3-mm × 3-mm area. The stack of angiography selection was applied in correspondence to default choriocapillaris visualization and then analyzed. An automated protocol was applied during follow-up. OCT angiograms were co-registered with standard structural OCT B-scans, allowing for contemporaneous visualization of both retinal flow and structure.
OCT angiograms were processed using AngioAnalytic software (official released in August 2017) by selecting choriocapillaris vessel density. The AngioAnalytic software reveals a nonhomogeneous pattern showing some colored areas related to the density flow. AngioAnalytic images of the choriocapillaris zone were exported using the default tool.
The CNV areas were measured using flow area of AngioAnalytic setting on “contour drawing” by two independent observers. The resulting measurements were compared by Cohen's kappa coefficient. The interobserver agreement of image analyzed was 0.92 (K = 0.225).
Dark Halo Assessment
As the dark halo is seen around CNV in choriocapillaris, scans from this region were collected for analysis (Figure 1).
Flow density analysis in pre- and post-intravitreal injection (A, A'). The blue area represents the “dark halo.” The images were exported and processed by Image J software (B, B'). The dark halo around the choroidal neovascularization were automatically identified (red area) by “adjust threshold” function (C, C'). The resulted area (red area) was automated quantified and used for the analysis.
For each image, flow density and scan segmentation were automatically selected and analyzed by the software. It was used to measure the blue space areas corresponding to dark halo. All the images were analyzed using Image J software (Version 1.50i; National Institutes of Health, Bethesda, MD) as previously described.25 Briefly, after conversion to 8-bit images, blue spaces were identified by the “adjust threshold” function with intensity threshold set from 0 to 50 for choriocapillaris OCTA images. The resulting binary images identified dark halo spaces as a red area in pixel2 and were then automatically quantified by Image J software and collected for the statistical analysis. In set scale, 1 mm corresponded to 80 pixels. The dark halo area obtained was converted from pixel2 to mm2.
The dark halo was assessed for each pre-injection evaluation and for each available examination between 6 days and 14 days after injection.
Data were analyzed by GraphPad PRISM Software (Version 6.0; GraphPad, La Jolla, CA). The changes of CNV and dark halo areas before and after treatment were assessed using t-test. To evaluate CNV and dark halo areas ratio, linear correlation test has been used. P values less than .05 were considered statistically significant.
OCTA has shown that the dark halo around the CNV is frequently observed. This dark area seems not to due to a masking effect, but to a real flow reduction, as shown by ICG.26
Dark halo evolution during and after treatment until recurrence is shown in Figure 2.
Optical coherence tomography angiography of normal halo evolution before treatment and at 1 day, 7 days, and 4 weeks after the intravitreal anti-vascular endothelial growth factor injection.
A large dark halo surrounds CNV before treatment, most likely due to choriocapillaris flow decrease. Twenty-four hours after the injection, CNV area and halo area begin to decrease. From 8 days to 12 days after the injection, the CNV continues to decrease, with a matching halo decrease. The maximum area reduction is observed approximately on day 10. On other scans recorded later, 4 weeks after injection, the halo increases again, and the CNV major vessels reappear again. We observed fluctuation in CNV, vascular density, and darkness of choriocapillaris area before and after intravitreal injections with area increasing and decreasing (Figure 3).
Flow signal on B-scan, optical coherence tomography angiography (OCTA), vascular density, and Image J processing are collected before and after three intravitreal injections of some eyes. “Dark halo” is seen on OCTA as a dark area and on vascular density a blue area. It is automatically identified and quantified by Image J software (red area).
We followed up at 12 months to 24 months all CNV cases, measuring vascular density (dark low perfusion halo) before treatment and at 7 days to 13 days after treatment (Figure 4).
The graphs show choroidal neovascularization area (left) and “dark halo” area (right) reduction after intravitreal anti-vascular endothelial growth factor injection. Both reductions were statistically significant (P < .001).
We observed a fluctuation of the dark halo around the CNV that decreased parallel to CNV activity reduction. In most of the cases (95.4%), the dark halo was larger before treatment and attained its smallest dimensions from 6 days to 13 days after injection. Parallel, the mean CNV area was reduced in 83.72% of the cases.
The mean CNV area before treatment was 0.44 mm2 (± 0.40 mm2) and 0.34 mm2 (± 0.29 mm2) after injection. The mean dark halo area before treatment was 0.34 mm2 (± 0.27 mm2) and 0.25 mm2 (± 0.23 mm2) after injection. CNV area and dark halo area reductions were statistically significant (P < .001) after the injections. The CNV-to-halo ratio was 2.06 before treatment and 2.46 after treatment with a P value less than .001 (Figure 5).
Correlation between the choroidal neovascularization (CNV) and “dark halo” ratio before and after intravitreal injection was statistically significant (P < .001).
The correlation between the CNV area and the dark halo was not statistically significant (P = .35). An association between CNV growth and intraretinal fluid development in many cases, but not in all cases (71%).
OCTA allows noninvasive monitoring of the retinal and choriocapillaris microvasculature in patients with CNV. OCTA may be repeated frequently to assess chorioretinal changes and to identify the alterations preceding and follow CNV development. In our study, we observed around the new vessels before the injection a large dark halo. This area seems to coincide with an area of choriocapillaris flow decrease. As mentioned by McLeod et al.22 and Lutty et al.23 there could be a reduction in choriocapillaris flow in AMD, and this loss of choriocapillaris flow is associated with Bruch's membrane deposits. They concluded that RPE could stimulate the formation and regression of CNV, and RPE loss could result in loss of choriocapillaris.
OCTA has shown that the dark halo around the CNV is frequently observed. This dark area seems not to be due to a masking effect, but to a real flow reduction, as shown by ICG.26
In our experience, the best method for studying and quantifying the dark halo by OCTA is to observe the choriocapillaris segmentation slab.
Although several artifacts by OCTA have been reported,27 our quantitative results seem to confirm observations by Jia et al.7 and Moult et al.13 that CNV high-flow lesions appear in areas of lower choriocapillaris flow to compensate the reduced circulation.
Recently, Seddon et al.28 described that histopathologic changes in the choriocapillaris might regulate the production of VEGF, providing the stimulus for new vessels growth. In previous studies, a distinct dark rim surrounding the CNV net was evident until the late phases of ICG angiography. They indicated that the dark rim corresponded to reactive hypertrophic / hyperplastic pigment epithelial cells enveloping the outer margin of the CNV and represents a favorable healing response.26
Jia et al.7 stated that OCTA could provide further insight into choroidal flow and AMD pathogenesis. They reported that deep choroidal vessels were more apparent and suspected that this finding was due to loss of choriocapillaris associated with AMD. The authors also suspected that these areas of possible ischemia might play a role in CNV pathogenesis. There were focal regions under and adjacent to the CNV, where there was greatly reduced flow in both the choriocapillaris and the deeper choroid.
Moult et al.13 observed that the region of CNV was encircled by a halo of choriocapillaris alteration. They reported CNV lesions surrounded by a region of severe choriocapillaris alteration.
Coscas et al.2 reported a large flow void and/or dark halo around exudative CNV eyes. Similarly, El Ameen et al.29 described surrounded CNV halo in cases of type 2 lesions detected by OCT angiography.
Dark halos do not seem to be due to segmentation error or to masking effects by fluid, hemorrhages, or pigment, nor to localized areas of atrophy, but rather to focal areas of choroidal low perfusion. The possibility of “false friends” described by Coscas et al.27 should always be considered during the interpretation of OCTA. However, these authors described misidentification of artifacts secondary to the RPE shadow effect of light passing through the blood vessels, as described previously by Spaide.30
We observed that the dark halo vessel density fluctuations last a few days and are too rapid to be due to blood or pigment masking. They seem related to flow fluctuations that are faster and possible during a few days' time frame. Our results agree with El Ameen et al.28 and Coscas et al.,2 who described slow flow in the choriocapillaris derived from outer retinal ischemia as the key factor in CNV development. As reported by Fiegel, ischemia due to slow choriocapillaris flow caused the release of vasogenic factors, causing CNV activation and growth.31 In our opinion, CNV may be considered a high-flow vascular structure that can induce a decreased flow of the choriocapillaris by blood deviation and blood sequestering.
In this study, for the first time, we were able to quantify the evolution of choriocapillaris vascular density around neovascularization before and after treatment.
Halo fluctuation seems to be parallel to CNV evolution after treatment. Our first results appear to confirm those of Jia et al.7 and Moult et al.,13 who suggested that CNV might form in the choriocapillaris region, as well as some alterations to compensate for the reduction in circulation. This feature could be due to blood flow sequestering (relative ischemia) around CNV when new vessels are at the highest peak of activity, followed by choriocapillaris increased perfusion (and decreased halo) when CNV activity abates after treatment. The dark area reaches its minimal surface at 6 days to 13 days after anti-angiogenic intravitreal injection. At that time point, the maximum CNV regression is attained when smaller vascular branches disappear, leaving a very low-density dark area. The variation in the obscure rim area, in dimensions, and in vascular density (darkness) is not stable or durable. There are constant fluctuations, increases, and decreases in the dark halo area in dimensions and in vascular density. Some areas seem to show even lower perfusion due to blood sequestering by high-flow CNV. In these areas, RPE cells could provide the stimulus for new CNV growth.
In our results, rapid dark halo fluctuations seem to show that vessels are present even if not visible, and blood flow reappears or is detected again during evolution. As described by Miere et al.,32 it is possible that we overlook the structures where blood flow does not reach the level of detection necessary with the device.
In conclusion, in our study, halo and CNV area evolution are closely parallel. We did not foresee this close agreement. Even less foreseen is the observation that the sensitivity of dark halo after injections is very high (95.4%) — more than CNV area reduction (83.7%). These findings could suggest that dark halo enlargement is more predictive of neovascularization activity than the CNV growth.
Larger studies are needed to confirm our preliminary results, which suggest that CNV lesions might develop in choriocapillaris areas to compensate the reduced circulation and the dark halo could be an early sign of CNV activation. In the future, the quantification of the flow variation around CNV could be a useful element either for assessing the result of treatments or for predicting the CNV reactivation of disease before fluid appears.
- Costanzo E, Miere A, Querques G, Capuano V, Jung C, Souied EH. Type 1 choroidal neovascularization lesion size: Indocyanine green angiography versus optical coherence tomography angiography. Invest Ophthalmol Vis Sci. 2016;57(9):307–313. doi:10.1167/iovs.15-18830 [CrossRef]
- Coscas GJ, Lupidi M, Coscas F, Cagini C, Souied EH. Optical coherence tomography angiography versus traditional multimodal imaging in assessing the activity of exudative age-related macular degeneration: A new diagnostic challenge. Retina. 2015;35(11):2219–2228. doi:10.1097/IAE.0000000000000766 [CrossRef]
- Spaide RF, Klancnik JM Jr, Cooney MJ. Retinal vascular layers imaged by fluorescein angiography and optical coherence tomography angiography. JAMA Ophthalmol. 2015;133(1):45–50. doi:10.1001/jamaophthalmol.2014.3616 [CrossRef]
- Savastano MC, Lumbroso B, Rispoli M. In vivo characterization of retinal vascularization morphology using optical coherence tomography angiography. Retina. 2015;35(11):2196–2203. doi:10.1097/IAE.0000000000000635 [CrossRef]
- Campbell JP, Zhang M, Hwang TS, et al. Detailed vascular anatomy of the human retina by projection resolved optical coherence tomography angiography. Sci Rep. 2017;7:42201. doi:10.1038/srep42201 [CrossRef]
- 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]
- Jia Y, Bailey ST, Wilson DJ, et al. Quantitative optical coherence tomography angiography of choroidal neovascularization in age-related macular degeneration. Ophthalmology. 2014;121(7):1435–1444. doi:10.1016/j.ophtha.2014.01.034 [CrossRef]
- Huang D, Jia Y, Rispoli M, Tan O, Lumbroso B. Optical coherence tomography angiography of time course of choroidal neovascularization in response to anti-angiogenic treatment. Retina. 2015;35(11):2260–2264. doi:10.1097/IAE.0000000000000846 [CrossRef]
- Lumbroso B, Rispoli M, Savastano MC. Longitudinal optical coherence tomography-angiography study of type 2 naive choroidal neovascularization early response after treatment. Retina. 2015;35(11):2242–2251. doi:10.1097/IAE.0000000000000879 [CrossRef]
- Ding X, Patel M, Chan CC. Molecular pathology of age-related macular degeneration. Prog Retin Eye Res. 2009;28(1):1–18. doi:10.1016/j.preteyeres.2008.10.001 [CrossRef]
- Nowak JZ. Age-related macular degeneration (AMD): Pathogenesis and therapy. Pharmacol Rep. 2006;58(3):353–363.
- Xu W, Grunwald JE, Metelitsina TI, et al. Association of risk factors for choroidal neovascularization in age-related macular degeneration with decreased foveolar choroidal circulation. Am J Ophthalmol. 2010;150(1):40–47.e2. doi:10.1016/j.ajo.2010.01.041 [CrossRef]
- Moult E, Choi W, Waheed NK, et al. Ultrahigh-speed swept-source OCT angiography in exudative AMD. Ophthalmic Surg Lasers Imaging Retina. 2014;45(6):496–505. doi:10.3928/23258160-20141118-03 [CrossRef]
- Lumbroso B, Rispoli M, Savastano MC, Jia Y, Tan O, Huang D. Optical coherence tomography angiography study of choroidal neovascularization early response after treatment. Dev Ophthalmol. 2016;56:77–85. doi:10.1159/000442782 [CrossRef]
- Hayashi K, de Laey JJ. Indocyanine green angiography of choroidal neovascular membranes. Ophthalmologica. 1985;190(1):30–39. doi:10.1159/000309489 [CrossRef]
- Melrose MA, Magargal LE, Goldberg RE, Annesley WH. Subretinal neovascular membranes associated with choroidal nonperfusion and retinal ischemia. Ann Ophthalmol. 1987;19(10):396–399.
- Staurenghi G, Flower RW. Clinical observations supporting a theoretical model of choriocapillaris blood flow in treatment of choroidal neovascularization associated with age-related macular degeneration. Am J Ophthalmol. 2002;133(6):801–808. doi:10.1016/S0002-9394(02)01436-8 [CrossRef]
- Flower RW, von Kerczek C, Zhu L, Ernest A, Eggleton C, Topoleski LD. Theoretical investigation of the role of choriocapillaris blood flow in treatment of subfoveal choroidal neovascularization associated with age-related macular degeneration. Am J Ophthalmol. 2001;132(1):85–93. doi:10.1016/S0002-9394(01)00872-8 [CrossRef]
- De Carlo TE, Bonini Filho MA, Chin AT. Spectral-domain optical coherence tomography angiography of choroidal neovascularization. Ophthalmology. 2015;122(6):1228–1238. doi:10.1016/j.ophtha.2015.01.029 [CrossRef]
- Theelen T1, Berendschot TT, Hoyng CB, Boon CJ, Klevering BJ. Near-infrared reflectance imaging of neovascular age-related macular degeneration. Graefes Arch Clin Exp Ophthalmol. 2009;247(12):1625–1633. doi:10.1007/s00417-009-1148-9 [CrossRef]
- Toju R, Iida T, Sekiryu T, Saito M, Maruko I, Kano M. Near-infrared autofluorescence in patients with idiopathic submacular choroidal neovascularization. Am J Ophthalmol. 2012;153(2):314–319. doi:10.1016/j.ajo.2011.06.026 [CrossRef]
- McLeod DS, Grebe R, Bhutto I, Merges C, Baba T, Lutty GA. Relationship between RPE and choriocapillaris in age-related macular degeneration. Invest Ophthalmol Vis Sci. 2009;50(10):4982–4991. doi:10.1167/iovs.09-3639 [CrossRef]
- Lutty G, Grunwald J, Majji AB, Uyama M, Yoneya S. Changes in choriocapillaris and retinal pigment epithelium in age-related macular degeneration. Mol Vis. 1999;5:35.
- Jung JJ, Chen CY, Mrejen S, et al. The incidence of neovascular subtypes in newly diagnosed neovascular age-related macular degeneration. Am J Ophthalmol. 2014;158(4):769–779.e2. doi:10.1016/j.ajo.2014.07.006 [CrossRef]
- Matet A, Savastano MC, Rispoli M, et al. En face optical coherence tomography of foveal microstructure in full-thickness macular hole: a model to study perifoveal Müller cells. Am J Ophthalmol. 2015;159(6):1142–1151. doi:10.1016/j.ajo.2015.02.013 [CrossRef]
- Gharbiya M, Pantaleoni FB, Grandinetti F, Gabrieli CB. Indocyanine green angiographic findings in idiopathic choroidal neovascularisation. Eye (Lond). 1999;13(Pt 5):621–628. doi:10.1038/eye.1999.170 [CrossRef]
- Coscas G, Lupidi M, Cagini C, Coscas F. ‘False-friend’ images on optical coherence tomography angiography: Early choroidal neovascularization or artefact?Acta Ophthalmol. 2018;96(2):200–202. doi:10.1111/aos.13078 [CrossRef]
- Seddon JM, McLeod DS, Bhutto IA, et al. Histopathological insights into choroidal vascular loss in clinically documented cases of age-related macular degeneration. JAMA Ophthalmol. 2016;134(11):1272–1280. doi:10.1001/jamaophthalmol.2016.3519 [CrossRef]
- El Ameen A, Cohen SY, Semoun O, et al. Type 2 neovascularization secondary to age-related macular degeneration imaged by optical coherence tomography angiography. Retina. 2015;35(11):2212–2218. doi:10.1097/IAE.0000000000000773 [CrossRef]
- Spaide RF. Optical coherence tomography angiography signs of vascular abnormalization with antiangiogenic therapy for choroidal neovascularization. Am J Ophthalmol. 2015;160(1):6–16. doi:10.1016/j.ajo.2015.04.012 [CrossRef]
- Feigl B. Age-related maculopathy - linking aetiology and pathophysiological changes to the ischaemia hypothesis. Prog Retin Eye Res. 2009;28(1):63–86. doi:10.1016/j.preteyeres.2008.11.004 [CrossRef]
- Miere A, Semoun O, Cohen SY, et al. Optical coherence tomography angiography features of subretinal fibrosis in age-related macular degeneration. Retina. 2015;35(11):2275–2284. doi:10.1097/IAE.0000000000000819 [CrossRef]