The pathogenesis of amblyopia has been studied extensively. Studies in humans and animals point toward a cortical processing deficit in amblyopia, along with sensory defects including reduced spatial resolution, reduced contrast sensitivity, and a reduced number of binocular neural cells. Earliest reports have also suggested that the activation of cerebral blood flow by amblyopic eyes is decreased as compared to normal eyes.1,2 The retina has also been investigated in the pathogenesis of amblyopia. Macular sensitivity is reduced in patients with amblyopia but grossly; however, no other abnormalities have yet been reported consistently.
Optical coherence tomography angiography (OCTA) is a noninvasive technique that generates angiography images in a short span of time. It has a wide applicability in detecting both structural as well as blood flow information of the retina and choroid. Anecdotal reports have suggested microstructural changes in the retina and choroid in amblyopic patients. There is reportedly a decrease in vascular density in both superficial and deep retinal plexuses reported in amblyopic versus contralateral eyes.3,4 Functional parameters are also affected in the macula and the optic disc in both eyes of a unilateral amblyope.5 To improve our understanding of the pathogenesis of the disease, we aim to determine if any retinal or choroidal flow abnormalities exist in patients with amblyopia. The aim of our study was to calculate the foveal avascular zone and retinal and choriocapillaris (CC) vessel density in patients with amblyopia and compare it with the fellow normal eyes as controls.
Patients and Methods
The study conformed to the declaration of Helsinki and institutes ethical clearance was obtained. It was a prospective, observational study conducted at the Advanced Eye Center, Post Graduate Institute of Education and Research, Chandigarh, India. We included healthy subjects aged 6 years to 12 years with normal ophthalmic findings and unilateral anisometropic amblyopia. We included only treatment-naïve amblyopes with best-corrected visual acuity (BCVA) in the amblyopic eye between logMAR 0.3 and logMAR 0.7 (20/40 to 20/100) and logMAR 0.1 (20/25) or better in the contralateral eye. The visual acuity (VA) was chosen to include only moderate amblyopes who could give us a clear OCTA scan. Anisometropia was defined as an interocular cycloplegic spherical equivalent (SE) difference or astigmatism difference of at least 1.5 diopters (D). Cycloplegia was induced with 0.75% topical cyclopentolate put three times at an interval of 15 minutes. SE was calculated as the sum of the spherical plus half of the cylindrical error. The fellow eyes of the patients were enrolled in the control group. Patients with any ophthalmic structural abnormality; history of intraocular surgery, retinal or optic nerve disease, or any ocular media opacities including cataract were excluded from the study.
All patients underwent a complete ophthalmic examination, including BCVA, cycloplegic refraction, and biometry on the IOLMaster 500 (Zeiss, Oberkochen, Germany). OCTA imaging was performed using the RTVue XR Avanti spectral-domain OCT device with AngioVue (Optovue, Fremont, CA) with a light source at 840 nm, a bandwidth of 45 nm, and an A-scan rate of 70,000 scans per second. A 3 mm × 3 mm cube scan centered at the fovea was acquired containing 304 × 304 A-scans, with two consecutive B-scans captured at each fixed position to separate static tissue from blood flow signals. Two orthogonal volume scans (horizontal and vertical) were acquired for each eye and merged to reduce motion artifacts. The signal-to-noise ratio was increased using the split-spectrum amplitude decorrelation technology (SSADA).6
Each scan was segmented into en face images of the superficial capillary plexus (SCP), deep capillary plexus (DCP) of the retina, and choriocapillaris (CC) using the auto-segmentation feature in the AngioVue software. The limits for SCP were set with an inner boundary 3 μm below the internal limiting membrane and an outer boundary 15 μm below the inner plexiform layer (IPL). The DCP en face OCTA image was segmented with an inner boundary 15 μm below the IPL and an outer boundary at 70 μm below the IPL. The CC en face OCTA image was derived by the software with a slab 10-μm thick starting 31 deep to the retinal pigment epithelium–Bruch's membrane complex. All scans were reviewed to ensure correct segmentation and sufficient image quality and were repeated if deemed inadequate for analysis. Image size correction was applied to compensate for changes in axial length. Patients with poor quality image were excluded from the analysis based on one or more of the following criteria: low signal strength (< 50), presence of blink artifacts, and poor fixation leading to motion or doubling artifacts.
To quantify the OCTA images, we measured the foveal avascular zone (FAZ) area and vessel density in the acquired scans using ImageJ version 1.50 software (National Institute of Health, Bethesda, MD). The FAZ area was defined as an avascular central area within the innermost visible capillaries and the borders of the FAZ were traced manually in both the SCP and DCP using the freehand selection tool on ImageJ by a trained retina specialist who was blind to the clinical information. The shadow artifacts were ignored while marking the boundary of the deep FAZ as described previously.7 The area of the FAZ was determined in mm2 after calibrating the image by using the set scale parameter of ImageJ to define the 304-pixel width of the images as 3 mm. The vessel density values were calculated for the entire en face scans of the SCP, DCP, and CC by binarizing the raw images to obtain clear vessel boundaries and suppress the background noise.8 After comparing various binarization techniques, Niblack's auto local thresholding technique was adopted in our study as it considers the mean and standard deviation of all the pixels in the region of interest.9 Since there are no vessels in the FAZ, the background noise tends to increase in this area after local thresholding; hence, this area as delineated previously was cleared of the background noise before proceeding with analysis. Vessel density was then calculated as the percentage of the total area occupied by blood vessels, with the blood vessels being defined as the pixels having a decorrelation value above the threshold level.
Statistical analysis was done on the Statistical Package For Social Sciences (IBM SPSS Statistics for Windows, version 21.0; SPSS Inc., Armonk, NY). Means were expressed as + standard deviation (SD) along with percentages expressed as proportions. Categorical variables were compared using the Chi-square or Fischer's exact test and the means using t-test. Non-parametric correlations were calculated. A P value less than .05 was considered statistically significant. We kept results of Guo L et al.10 as baseline. In order to find a 68.2% difference as in their study at a power of 80%, we found the sample size needed was more than five in each group.
We recruited 14 patients of amblyopia. The mean age of the patients was 10.26 years ± 3.31 years.
The refractive error in the amblyopic eye ranged from −7 D to +6 D and in the fellow eye from −2.25 D to +4 D. Table 1 outlines the difference in the various parameters amongst the two eyes of the same patient along with their significance.
Difference in Parameters and Their Significance Between the Amblyopic and the Fellow Eye
There was a significant difference in the VA between two eyes. Biometric variables were comparable. The vessel density of the CC in the amblyopic eye was significantly reduced than the fellow eye (P = .005). However, there was no difference in the vessel density as well as the area of the FAZ between the two eyes.
Spearman correlations revealed the superficial as well as deep vessel density area positively correlated with the age (P = .00 and P = .04, respectively). However, there was no correlation of VA with the superficial (P = .51) or deep vessel density (P = .62). The CC vessel density correlated positively with the VA. (r = 0.41; P = .03). Figure 1 depicts a representative image of vessel density of superficial and deep vasculature. The CC were found to be attenuated as compared to the control eyes in all patients. (Figure 2).
Image depicts a representative image of vessel density of superficial and deep vasculature of the amblyopic eye. FAZ = foveal avascular zone
Image highlights attenuated choriocapillaris in the amblyopic eye in comparison to fellow eye.
We compared the retinal and CC vessel densities and flow in participants with amblyopia and compared to the normal eye. The superficial and the deep retinal flow was found to be comparable to the fellow normal eye. On the other hand, the CC flow was significantly less in the amblyopic eye than the normal fellow eye in patients with amblyopia. Moreover, the VA of the amblyopic eye correlated positively with the vessel density of the CC.
Although the debate is still on with regard to the exact location of defect in amblyopia, decreased VA is not the only abnormality in amblyopia. Other known abnormalities include reduced contrast, reduced visual fields and altered dark adaptation.11–13 Demer et al. demonstrated a reduction in relative cerebral blood flow and glucose metabolism in the amblyopic eye as compared to normal.1 Authors have described amblyopia as a complex disease that affects various levels of the visual pathway including the visual cortex, as well as the structure of the retina and choroid.14
Significant qualitative and quantitative differences in the retinal microstructures of the fovea have been noted in amblyopes from normal eyes. A significant correlation has been observed between the increased outer segment length and the improvement in the BCVA in amblyopic eyes following optical treatment.15 Abnormal central macula associated with myopic anisometropic amblyopia tended to be thinner and more normal following amblyopia treatment without affecting peripheral retina.16 Al-Haddad et al. observed the macula is immature in amblyopic eyes compared with fellow normal eyes.17 Most of these studies were based on the OCT features in amblyopia. With the advent of the new OCTA using SS-ADA algorithm, a couple of studies reciprocated these abnormalities in the vasculature of the macula, as well. A decrease in vessel density of the superficial as well as the deep vessel plexus was noted in the amblyopic eye as compared to controls in recent studies using the algorithm.3–5 However, none of the studies have documented the CC vessel density in these patients.
Our results are similar to the results of Guo et al.,10 who found no significant change in superficial and deep retinal plexus. We also found the choriodal vasculature to be attenuated. This attenuation was quantified as a decrease in vessel density of the CC in the amblyopic eye in our study. The choroid has also been shown to regulate ocular growth and refractive status of the eye.18,19 Choroidal thickness in amblyopic eyes was significantly thicker than in control eyes in many studies.14,20 Hence, these studies concluded that the amblyopic process may involve the choroid, and not only the retina. As with the pachychoroid spectrum of disease, we hypothesize that a thicker choroidal stroma could possibly be compressing the CC, leading to reduced vessel density on OCTA.21 The reduced blood flow in the CC maybe responsible for the amblyopia, but a direct cause-effect relationship cannot be ascertained from the present data.
Since there are still a number of amblyopes who are refractory to treatment, it may be that we are missing some pathogenetic steps that cause amblyopia. Difference of the choroidal vasculature may be one of the pathways to explore in this scenario. The OCTA is a noninvasive modality and can be used in larger number of children to reciprocate the results.
The study has the limitation of a small sample size. The nutritional status of the eye in terms of density of the CC may be the cause or the effect of amblyopia, which remains unanswered by our study. Temporal associations will be established through prospective studies undertaken before and after treatment.
There is yet no standardized or validated software to quantify the vascular densities by OCTA. So, comparing the FAZ area measurements and vessel density measurements amongst different studies computing results manually may be futile. Still, the present study to our knowledge is the first to quantitate CC vessel density and compare it with the fellow eye in patients with unilateral amblyopia. It will add to the data on clinical utility of OCTA in amblyopia.
- Demer JL, von Noorden GK, Volkow ND, Gould KL. Imaging of cerebral blood flow and metabolism in amblyopia by positron emission tomography. Am J Ophthalmol. 1988;105(4):337–347. https://doi.org/10.1016/0002-9394(88)90294-2 PMID: doi:10.1016/0002-9394(88)90294-2 [CrossRef]3258733
- Mizoguchi S, Suzuki Y, Kiyosawa M, Mochizuki M, Ishii K. Differential activation of cerebral blood flow by stimulating amblyopic and fellow eye. Graefes Arch Clin Exp Ophthalmol. 2005;243(6):576–582. https://doi.org/10.1007/s00417-004-1009-5 PMID: doi:10.1007/s00417-004-1009-5 [CrossRef]15650860
- LonngiVelez FG, Tsui I, et al. Spectral-domain optical coherence tomographic angiography in children with amblyopia. JAMA Ophthalmol2017;135:1086–91. doi:10.1001/jamaophthalmol.2017.3423 [CrossRef]28910439
- Yilmaz I, Ocak OB, Yilmaz BS, Inal A, Gokyigit B, Taskapili M. Comparison of quantitative measurement of foveal avascular zone and macular vessel density in eyes of children with amblyopia and healthy controls: an optical coherence tomography angiography study. J AAPOS. 2017;21(3):224–228. https://doi.org/10.1016/j.jaapos.2017.05.002 PMID: doi:10.1016/j.jaapos.2017.05.002 [CrossRef]28501447
- Sobral I, Rodrigues TM, Soares M, et al. OCT angiography findings in children with amblyopia. J AAPOS. 2018;22(4):286–289.e2. https://doi.org/10.1016/j.jaapos.2018.03.009 PMID: doi:10.1016/j.jaapos.2018.03.009 [CrossRef]30031875
- Jia Y, Tan O, Tokayer J, et al. Split-spectrum amplitude-decorrelation angiography with optical coherence tomography. Opt Express. 2012;20(4):4710–4725. https://doi.org/10.1364/OE.20.004710 PMID: doi:10.1364/OE.20.004710 [CrossRef]22418228
- Iafe NA, Phasukkijwatana N, Chen X, Sarraf D. Retinal Capillary Density and Foveal Avascular Zone Area Are Age-Dependent: Quantitative Analysis Using Optical Coherence Tomography Angiography. Invest Ophthalmol Vis Sci. 2016;57(13):5780–5787. https://doi.org/10.1167/iovs.16-20045 PMID: doi:10.1167/iovs.16-20045 [CrossRef]27792812
- Hartig SM. Basic image analysis and manipulation in ImageJ. Curr Protoc Mol Biol. 2013;Chapter 14:Unit14.15. https://doi.org/10.1002/0471142727.mb1415s102 PMID:23547012
- Niblack W. An Introduction to Digital Image Processing. Prentice Hall International Inc.: Upper Saddle River, NJ; 1986.
- Guo L, Tao J, Xia F, Yang Z, Ma X, Hua R. In vivo optical imaging of amblyopia: digital subtraction autofluorescence and split-spectrum amplitude-decorrelation angiography. Lasers Surg Med. 2016;48(7):660–667. https://doi.org/10.1002/lsm.22520 PMID: doi:10.1002/lsm.22520 [CrossRef]27075555
- Hess RF, Howell ER. The threshold contrast sensitivity function in strabismic amblyopia: evidence for a two type classification. Vision Res. 1977;17(9):1049–1055. https://doi.org/10.1016/0042-6989(77)90009-8 PMID: doi:10.1016/0042-6989(77)90009-8 [CrossRef]595414
- Donahue SP, Wall M, Kutzko KE, Kardon RH. Automated perimetry in amblyopia: a generalized depression. Am J Ophthalmol. 1999;127(3):312–321. https://doi.org/10.1016/S0002-9394(98)90327-0 PMID: doi:10.1016/S0002-9394(98)90327-0 [CrossRef]10088742
- Osman ZM, Abbas SR, Hefni W. Dark adaptation curve for different types of amblyopia. Bull Ophthalmol Soc Egypt. 1978;71(75):327–338. PMID:549727
- Xu J, Zheng J, Yu S, et al. Macular choroidal thickness in unilateral amblyopic children. Invest Ophthalmol Vis Sci. 2014;55(11):7361–7368. https://doi.org/10.1167/iovs.14-14439 PMID: doi:10.1167/iovs.14-14439 [CrossRef]25316720
- Nishi T, Ueda T, Hasegawa T, Miyata K, Ogata N. Retinal thickness in children with anisohypermetropic amblyopia. Br J Ophthalmol. 2015;99(8):1060–1064. https://doi.org/10.1136/bjophthalmol-2014-305685 PMID: doi:10.1136/bjophthalmol-2014-305685 [CrossRef]25680622
- Pang Y, Frantz KA, Block S, Goodfellow GW, Allison C. Effect of amblyopia treatment on macular thickness in eyes with myopic anisometropic amblyopia. Invest Ophthalmol Vis Sci. 2015;56(4):2677–2683. https://doi.org/10.1167/iovs.14-15532 PMID: doi:10.1167/iovs.14-15532 [CrossRef]25788654
- Al-Haddad CE, El Mollayess GM, Mahfoud ZR, Jaafar DF, Bashshur ZF. Macular ultrastructural features in amblyopia using high-definition optical coherence tomography. Br J Ophthalmol. 2013;97(3):318–322. https://doi.org/10.1136/bjophthalmol-2012-302434 PMID: doi:10.1136/bjophthalmol-2012-302434 [CrossRef]
- Pendrak K, Papastergiou GI, Lin T, Laties AM, Stone RA. Choroidal vascular permeability in visually regulated eye growth. Exp Eye Res. 2000;70(5):629–637. Erratum in: Exp Eye Res 2000;71:329. https://doi.org/10.1006/exer.2000.0825 PMID: doi:10.1006/exer.2000.0825 [CrossRef]10870521
- Troilo D, Nickla DL, Wildsoet CF. Choroidal thickness changes during altered eye growth and refractive state in a primate. Invest Ophthalmol Vis Sci. 2000;41(6):1249–1258. PMID:10798638
- Kara O, Altintas O, Karaman S, Emre E, Caglar Y. Analysis of Choroidal Thickness Using Spectral-Domain OCT in Children With Unilateral Amblyopia. J Pediatr Ophthalmol Strabismus. 2015;52(3):159–166. https://doi.org/10.3928/01913913-20150311-11 PMID: doi:10.3928/01913913-20150311-11 [CrossRef]26225384
- Ayachit G, Ayachit A, Nadgir H, Joshi S. Validating the pachychoroid disease spectrum using multimodal imaging. Indian J Ophthalmol. 2018;66(7):1022–1024. https://doi.org/10.4103/ijo.IJO_180_18 PMID: doi:10.4103/ijo.IJO_180_18 [CrossRef]29941762
Difference in Parameters and Their Significance Between the Amblyopic and the Fellow Eye
|Amblyopic Eye||Normal Eye||P Value|
|Mean refractive error (S.E)||+2.73 ± 5.86||+0.34 ± 2.45||.043*|
|Mean cylinder (D)||0.52 ± 1.29||0.04 ± 0.82||.204|
|Log Mar visual acuity||0.68 ± 0.18||.077 ± 0.17||.000*|
|Mean axial length (mm)||22.10 ± 2.59||20.87 ± 6.72||.578|
|Mean keratometry (D)||43.73 ± 1.31||40.17 ± 12.69||.366|
|FAZ Area (superficial) (mm2)||.33 .10 ±0.11||.33 .10 ± 0.12||.958|
|Vessel Density (superficial) (%)||33.51 ± 1.27||33.70 ± 1.55||.733|
|FAZ Area Deep (mm2)||0.62 ± 0.15||0.6179 ±0.16||.962|
|Vessel Density Deep (%)||34.55 ± 1.25||35.20 ± 0.41||.211|
|Vessel Density Choroid (%)||39.61 ± 0.45||44.10 ±0.37||.005*|