Association of Optical Coherence Tomography Angiography Metrics With Detection of Impaired Macular Microvasculature and Decreased Vision in Amblyopic Eyes: The Hong Kong Children Eye Study | Macular Diseases | JAMA Ophthalmology | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Optical Coherence Tomography Angiography (OCT-A) of an Amblyopic Eye and Control Eye From 2 Participants
Optical Coherence Tomography Angiography (OCT-A) of an Amblyopic Eye and Control Eye From 2 Participants

A and B, Original OCT-A outputs of the 6 × 6 area centered on the foveal avascular zone (FAZ; indicated by yellow arrowheads). C and D, In the vessel density map with color grading, both eyes have similar vessel density (97.1% and 97.3%, respectively).

Figure 2.  Optical Coherence Tomography Angiography (OCT-A) of an Amblyopic Eye and Control Eye From 2 Participants
Optical Coherence Tomography Angiography (OCT-A) of an Amblyopic Eye and Control Eye From 2 Participants

A more irregular shape of the foveal avascular zone circularity is seen in the amblyopic eye (A) compared with the control eye (B). C and D, Skeletonization (represented by red tracing) of the OCT-A images for the measurement of fractal dimension and vessel density index.

Table 1.  Demographics of Amblyopic Cases and Control
Demographics of Amblyopic Cases and Control
Table 2.  Multivariable Analysis Comparing Amblyopic Eyes vs Control Eyes in Terms of FAZ, FAZ Circularity, Fractal Dimension, Vessel Diameter Index, and Vessel Densitya
Multivariable Analysis Comparing Amblyopic Eyes vs Control Eyes in Terms of FAZ, FAZ Circularity, Fractal Dimension, Vessel Diameter Index, and Vessel Densitya
Table 3.  Linear Regression Analysis Between FAZ Area, FAZ Circularity, Vessel Density, Vessel Diameter Index, and Fractal Dimension With Best-Corrected Visual Acuity After Adjusting for Covariatesa
Linear Regression Analysis Between FAZ Area, FAZ Circularity, Vessel Density, Vessel Diameter Index, and Fractal Dimension With Best-Corrected Visual Acuity After Adjusting for Covariatesa
1.
McKean-Cowdin  R, Cotter  SA, Tarczy-Hornoch  K,  et al; Multi-Ethnic Pediatric Eye Disease Study Group.  Prevalence of amblyopia or strabismus in Asian and non-Hispanic white preschool children: multi-ethnic pediatric eye disease study.   Ophthalmology. 2013;120(10):2117-2124. doi:10.1016/j.ophtha.2013.03.001 PubMedGoogle Scholar
2.
Chia  A, Dirani  M, Chan  YH,  et al.  Prevalence of amblyopia and strabismus in young Singaporean Chinese children.   Invest Ophthalmol Vis Sci. 2010;51(7):3411-3417. doi:10.1167/iovs.09-4461 PubMedGoogle Scholar
3.
Peterseim  MM, Papa  CE, Wilson  ME,  et al.  The effectiveness of the Spot Vision Screener in detecting amblyopia risk factors.   J AAPOS. 2014;18(6):539-542. doi:10.1016/j.jaapos.2014.07.176 PubMedGoogle Scholar
4.
Amblyopia  CE.  Amblyopia.   Surv Ophthalmol. 1995;40(1):23-39. doi:10.1016/S0039-6257(95)80044-1 PubMedGoogle Scholar
5.
Stryker  MP, Harris  WA.  Binocular impulse blockade prevents the formation of ocular dominance columns in cat visual cortex.   J Neurosci. 1986;6(8):2117-2133. doi:10.1523/JNEUROSCI.06-08-02117.1986 PubMedGoogle Scholar
6.
Levi  DM.  Visual processing in amblyopia: human studies.   Strabismus. 2006;14(1):11-19. doi:10.1080/09273970500536243 PubMedGoogle Scholar
7.
Huberman  AD, Feller  MB, Chapman  B.  Mechanisms underlying development of visual maps and receptive fields.   Annu Rev Neurosci. 2008;31:479-509. doi:10.1146/annurev.neuro.31.060407.125533 PubMedGoogle Scholar
8.
Bruce  A, Pacey  IE, Bradbury  JA, Scally  AJ, Barrett  BT.  Bilateral changes in foveal structure in individuals with amblyopia.   Ophthalmology. 2013;120(2):395-403. doi:10.1016/j.ophtha.2012.07.088 PubMedGoogle Scholar
9.
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. doi:10.1136/bjophthalmol-2014-305685 PubMedGoogle Scholar
10.
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. doi:10.1136/bjophthalmol-2012-302434 PubMedGoogle Scholar
11.
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. doi:10.1016/j.jaapos.2017.05.002 PubMedGoogle Scholar
12.
Lonngi  M, Velez  FG, Tsui  I,  et al.  Spectral-domain optical coherence tomographic angiography in children with amblyopia.   JAMA Ophthalmol. 2017;135(10):1086-1091. doi:10.1001/jamaophthalmol.2017.3423 PubMedGoogle Scholar
13.
Sobral  I, Rodrigues  TM, Soares  M,  et al.  OCT angiography findings in children with amblyopia.   J AAPOS. 2018;22(4):286-289.e2. doi:10.1016/j.jaapos.2018.03.009PubMedGoogle Scholar
14.
Borrelli  E, Lonngi  M, Balasubramanian  S,  et al.  Increased choriocapillaris vessel density in amblyopic children: a case-control study.   J AAPOS. 2018;22(5):366-370. doi:10.1016/j.jaapos.2018.04.005 PubMedGoogle Scholar
15.
Demirayak  B, Vural  A, Onur  IU, Kaya  FS, Yigit  FU.  Analysis of macular vessel density and foveal avascular zone using spectral-domain optical coherence tomography angiography in children with amblyopia.   J Pediatr Ophthalmol Strabismus. 2019;56(1):55-59. doi:10.3928/01913913-20181003-02 PubMedGoogle Scholar
16.
Chen  W, Lou  J, Thorn  F,  et al.  Retinal microvasculature in amblyopic children and the quantitative relationship between retinal perfusion and thickness.   Invest Ophthalmol Vis Sci. 2019;60(4):1185-1191. doi:10.1167/iovs.18-26416 PubMedGoogle Scholar
17.
Tang  FY, Ng  DS, Lam  A,  et al.  Determinants of quantitative optical coherence tomography angiography metrics in patients with diabetes.   Sci Rep. 2017;7(1):2575. doi:10.1038/s41598-017-02767-0 PubMedGoogle Scholar
18.
Cheung  CY, Li  J, Yuan  N,  et al.  Quantitative retinal microvasculature in children using swept-source optical coherence tomography: the Hong Kong Children Eye Study.   Br J Ophthalmol. 2018;103:672-679. doi:10.1136/bjophthalmol-2018-312413 PubMedGoogle Scholar
19.
Hsu  ST, Chen  X, Ngo  HT,  et al.  Imaging infant retinal vasculature with OCT angiography.   Ophthalmol Retina. 2019;3(1):95-96. doi:10.1016/j.oret.2018.06.017 PubMedGoogle Scholar
20.
Sun  Z, Tang  F, Wong  R,  et al.  OCT angiography metrics predict progression of diabetic retinopathy and development of diabetic macular edema: a prospective study.   Ophthalmology. 2019;126(12):1675-1684. doi:10.1016/j.ophtha.2019.06.016 PubMedGoogle Scholar
21.
Kim  AY, Chu  Z, Shahidzadeh  A, Wang  RK, Puliafito  CA, Kashani  AH.  Quantifying microvascular density and morphology in diabetic retinopathy using spectral-domain optical coherence tomography angiography.   Invest Ophthalmol Vis Sci. 2016;57(9):OCT362-OCT370. doi:10.1167/iovs.15-18904 PubMedGoogle Scholar
22.
Yam  JC, Tang  SM, Kam  KW,  et al.  High prevalence of myopia in children and their parents in Hong Kong Chinese Population: the Hong Kong Children Eye Study.   Acta Ophthalmol. 2020. Published online January 24, 2020. doi:10.1111/aos.14350PubMedGoogle Scholar
23.
Stanga  PE, Tsamis  E, Papayannis  A, Stringa  F, Cole  T, Jalil  A.  Swept-Source Optical Coherence Tomography Angio™ (Topcon Corp, Japan): technology review.   Dev Ophthalmol. 2016;56:13-17. doi:10.1159/000442771 PubMedGoogle Scholar
24.
Tang  F, Cheung  CY.  Quantitative retinal optical coherence tomography angiography in patients with diabetes without diabetic retinopathy.   Invest Ophthalmol Vis Sci. 2017;58(3):1766. doi:10.1167/iovs.17-21552 PubMedGoogle Scholar
25.
Domalpally  A, Danis  RP, White  J,  et al.  Circularity index as a risk factor for progression of geographic atrophy.   Ophthalmology. 2013;120(12):2666-2671. doi:10.1016/j.ophtha.2013.07.047 PubMedGoogle Scholar
26.
Cleland  BG, Crewther  SG, Crewther  DP.  The cat as a model for visual deprivation.   Aust N Z J Ophthalmol. 1985;13(3):263-269. doi:10.1111/j.1442-9071.1985.tb00433.x PubMedGoogle Scholar
27.
Crewther  SG, Crewther  DP, Cleland  BG.  Convergent strabismic amblyopia in cats.   Exp Brain Res. 1985;60(1):1-9. doi:10.1007/BF00237012 PubMedGoogle Scholar
28.
Sacconi  R, Borrelli  E, Querques  G.  Reproducibility of vessel density, fractal dimension, and foveal avascular zone using 7 different optical coherence tomography angiography devices.   Am J Ophthalmol. 2018;192:252-253. doi:10.1016/j.ajo.2018.05.020 PubMedGoogle Scholar
29.
Mo  S, Krawitz  B, Efstathiadis  E,  et al.  Imaging foveal microvasculature: optical coherence tomography angiography versus adaptive optics scanning light ophthalmoscope fluorescein angiography.   Invest Ophthalmol Vis Sci. 2016;57(9):OCT130-OCT140. doi:10.1167/iovs.15-18932 PubMedGoogle Scholar
30.
Bhardwaj  S, Tsui  E, Zahid  S,  et al.  Value of fractal analysis of optical coherence tomography angiography in various stages of diabetic retinopathy.   Retina. 2018;38(9):1816-1823. doi:10.1097/IAE.0000000000001774 PubMedGoogle Scholar
31.
Nguyen  TT, Wong  TY.  Retinal vascular manifestations of metabolic disorders.   Trends Endocrinol Metab. 2006;17(7):262-268. doi:10.1016/j.tem.2006.07.006 PubMedGoogle Scholar
32.
Sun  C, Wang  JJ, Mackey  DA, Wong  TY.  Retinal vascular caliber: systemic, environmental, and genetic associations.   Surv Ophthalmol. 2009;54(1):74-95. doi:10.1016/j.survophthal.2008.10.003 PubMedGoogle Scholar
33.
Joyal  JS, Gantner  ML, Smith  LEH.  Retinal energy demands control vascular supply of the retina in development and disease: the role of neuronal lipid and glucose metabolism.   Prog Retin Eye Res. 2018;64:131-156. doi:10.1016/j.preteyeres.2017.11.002 PubMedGoogle Scholar
34.
Holmen  IC, Konda  MS, Pak  JW,  et al.  Prevalence and severity of artifacts in optical coherence tomographic angiograms.   JAMA Ophthalmol. 2019;138(2):119-126. doi:10.1001/jamaophthalmol.2019.4971 PubMedGoogle Scholar
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    June 25, 2020

    Association of Optical Coherence Tomography Angiography Metrics With Detection of Impaired Macular Microvasculature and Decreased Vision in Amblyopic Eyes: The Hong Kong Children Eye Study

    Author Affiliations
    • 1Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong
    • 2Hong Kong Eye Hospital, Kowloon, Hong Kong
    • 3Zhejiang University School of Medicine, Department of Ophthalmology, Affiliated Hangzhou First People’s Hospital, Hangzhou, China
    • 4Department of Ophthalmology and Visual Sciences, Prince of Wales Hospital, Sha Tin, Hong Kong
    JAMA Ophthalmol. 2020;138(8):858-865. doi:10.1001/jamaophthalmol.2020.2220
    Key Points

    Question  Are macular microvascular abnormalities in children with amblyopia still present after adjusting for all known confounders and do they correlate with visual acuity?

    Findings  In this population-based nested case-control study of 30 children with amblyopia and 1045 controls, amblyopic eyes had abnormal macular microvascular represented by decreased foveal avascular zone circularity, decreased fractal dimension, and increased vessel diameter index. LogMAR visual acuity correlated with foveal avascular zone circularity and vessel diameter index.

    Meaning  Quantitative optical coherence tomography angiography metrics detect impaired microvasculature in amblyopic eyes that correlates with visual acuity, suggesting a potential role in amblyopia assessment.

    Abstract

    Importance  Microvascular abnormalities in amblyopia are becoming evident with high-resolution imaging, such as optical coherence tomography angiography (OCT-A); however, to our knowledge, the clinical significance and use of these findings are unknown.

    Objective  To assess changes in quantitative OCT-A metrics in amblyopic eyes and explore their association with visual acuity in children.

    Design, Setting, and Participants  This population-based nested case-control study included children aged 6 to 8 years who were consecutively recruited between January 2016 and July 2017 from the population-based Hong Kong Children Eye Study (HKCES) at the Chinese University of Hong Kong Eye Centre. All participants underwent OCT-A with a swept-source OCT and detailed ophthalmic investigations. Macular microvasculature of the superficial capillary plexus was quantified by a customized automated image analysis program. A multivariable linear regression was conducted to evaluate the differences in OCT-A metrics between amblyopic and nonamblyopic eyes after adjustment for all known confounders. Data analysis was conducted from September to November 2018.

    Main Outcomes and Measures  Differences in OCT-A metric (foveal avascular zone [FAZ]) area, FAZ circularity, vessel density, vessel diameter index, and fractal dimension between amblyopic and nonamblyopic eyes.

    Results  There were 30 participants with amblyopia (mean [SD] age, 7.57 [1.2] years; 16 girls [53.3%]) and 1045 controls (mean [SD] age, 7.65 [1.0] years; 580 girls [55.5%]) in this cohort. Compared with control eyes, amblyopic eyes had decreased FAZ circularity (−0.058; 95% CI, −0.096 to −0.021, P = .002), decreased fractal dimension (−0.014; 95% CI, −0.024 to −0.003; P = .01), and increased vessel diameter index (0.002; 95% CI, 0.002 to 0.003; P < .001). A difference was not identified between FAZ area and vessel density. LogMAR visual acuity was associated with FAZ circularity (sβ, −0.133; P < .001) and vessel diameter index (sβ, 0.097; P = .001) but not with vessel density nor FAZ area.

    Conclusions and Relevance  The results of this population-based study in children supports the presence of macular microvascular abnormalities in amblyopic eyes. Such changes as measured by OCT-A metrics are associated with visual acuity, inferring retinal involvement in the development of amblyopia and suggesting a potential role of quantitative OCT-A metrics in the diagnosis and recognition of amblyopia.

    Introduction

    Amblyopia is one of the most commonly encountered ophthalmic conditions in children, with the population-based prevalence ranging from 0.5% to 3.5%.1-3 Strabismus, anisometropia, and visual deprivation are common risk factors of amblyopia thought to result in functional and morphological effects on the visual cortex and lateral geniculate nucleus.4-7 Recent studies have started to look into the involvement of the retina in the development of amblyopia with evaluations of retinal nerve fiber layer, macular and choroidal thickness, and ganglion cell complex thickness with intriguing yet inconsistent results.8-10 Up to now, the exact nature and structural functional correlation of retinal microvasculature involvement in amblyopia has not been well defined.

    With the increased availability of optical coherence tomography angiography (OCT-A), detailed microvascular networks can now be quantified quickly and noninvasively. Several studies using OCT-A in children have reported abnormal microvasculature in amblyopic eyes compared with normal controls.11-16 However, because of heterogenous study designs and the presence of several confounders of OCT-A,17-19 the results have been inconsistent. Furthermore, these abnormal OCT-A parameters have not been demonstrated to be associated with visual acuity, raising questions regarding the clinical significance of these findings in amblyopia.

    In this study, we used a fully automated custom MATLAB (MathWorks) program20 to measure several unitless OCT-A metrics from OCT angiograms.17,21 These metrics quantify the extent of microvascular damages sensitively with good repeatability and reproducibility and have been widely used in our previous studies.17,21 To eliminate bias, all known ocular and systemic confounders were adjusted.18,19 We also explored whether macular microvasculature measured by OCT-A metrics is associated with visual acuity.

    Methods
    Study Population

    Data for this analysis were derived from the Hong Kong Children Eye Study (HKCES), a population-based cohort study of eye conditions in children from first grade to third grade (aged about 6-8 years) from primary schools in Hong Kong.22 In brief, the HKCES was designed to determine the prevalence of various eye disorders in children, including refractive errors, strabismus, amblyopia, and allergic eye diseases, and identify their environmental and genetic determinants. Sample selection was based on a stratified and clustered randomized sampling frame. In Hong Kong, all primary schools (n = 571) registered in the Education Bureau were stratified into 7 cluster regions according to population densities. This division into 7 cluster regions is determined by the Hong Kong government according to an even distribution of population density in each cluster region. In HKCES, the schools in each cluster region were randomly assigned an invitation priority according to the ranking numbers generated by computer. Invitations to participate in the cohort were sent according to the ranking numbers until the required sample was achieved in each cluster region.

    This study included a subpopulation of HKCES, and participants were consecutively recruited from January 2016 to July 2017 to undergo OCT-A imaging. All children were ethnic Chinese. Exclusion criteria included the presence of deprivation amblyopia, nystagmus, retinal diseases, intraocular inflammation, and media opacity, such as corneal disease or cataract. Patients with a history of prematurity, neurologic disease, or systemic conditions that could alter the microvasculature were excluded. In addition, patients with head, neck, or other injuries preventing proper positioning or with an inability to maintain retinal fixation were excluded.

    The study adhered to the Declaration of Helsinki and the study protocol was approved by the ethics committee board of the Prince of Wales Hospital, the Chinese University of Hong Kong. All parents signed an informed consent document on participation in the study. Participants were not offered payment or incentives to participate in the study.

    Definitions of Amblyopia

    Amblyopia was defined as a best-corrected visual acuity (BCVA) between 20/40 and 20/200 OD/OS or OU without an identifiable organic cause for the decreased vision. Only strabismic or anisometropic amblyopia were included in this study. For patients with bilateral amblyopia, the eye with poorer vision was selected for OCT-A imaging.

    Children with a BCVA of 20/20 or greater were included in the control group after completing a full ophthalmic examination with no evidence of any ocular abnormality. Only the right eye was chosen for analysis. If the right eye OCT-A image was excluded from analysis during quality control or missing, measurement would be performed on the left eye. The fellow eye of patients with unilateral amblyopia was not included in the control group even if the BCVA of fellow eye was 20/20 or greater.

    Ophthalmic Examinations

    All children underwent a full ophthalmic examination by an ophthalmic specialist. Visual acuity (VA) was measured with (aided VA) or without eyeglasses (unaided VA) using a logMAR chart (Nidek). Refractive status was measured before and after cycloplegia using an autorefractor (Nidek ARK-510A). Two cycles of cyclopentolate, 1% (Alcon), and tropicamide, 1% (Santen), were administered 10 minutes apart. Three or more readings of spherocylindrical autorefraction were taken at least 30 minutes after the last drop of cycloplegic agent. Axial length (AL) was measured with an IOL master (Carl Zeiss Meditec).

    OCT-A Imaging

    The protocol of OCT-A imaging and analysis has been previously reported.18 All study participants underwent swept-source OCT-A (Triton DRI-OCT; Topcon, Inc), which contains a swept source with a wavelength of 1050-nm light source and a speed of 100 000 A scans per second. Volumetric OCT scans centered on the fovea were obtained with a scan area of 6 mm × 6 mm containing 320 × 320 A scans. We used the built-in software (IMAGEnet6) to generate OCT-As, which can provide a high-detection sensitivity of low blood flow and reduced motion artifacts without compromising axial resolution.23 An image quality score ranging from 0 to 100 was given by the software for each volumetric OCT scan. Only the superficial capillary plexus was analyzed to avoid confounding from artifacts resulting from the shadow graphic projection from the former to the deep capillary plexus.24

    OCT-A Quality Control

    Two dedicated readers (J.L. and Y.N.) carefully evaluated each OCT-A image and OCT cross-sectional B scan image in the Chinese University Hong Kong Ocular Reading Centre. The readers were masked to all information of the participants. The OCT-A images with significant image artifacts or poor image quality were excluded from analysis, including (1) quality score below 40, (2) motion artifacts (eg, vessel discontinuity or significant residual motion lines), (3) inaccurate segmentation of tissue layers or slabs (a tolerance of 10% was allowed for inaccurate segmentation; manual segmentation was not done), (4) blur affecting 20% or more of the image (eg, because of tilted images, ocular media opacity, defocus, or axial movement), (5) signal loss (eg, due to eye blinking), or (6) poor centration (ie, fovea not at center).

    Quantification of Retinal Microvasculature

    A customized automated MATLAB program (2017 version; MathWorks) was used for all image analyses. The details of the quantification of retinal capillary network from OCT-A and its reliability assessment have been reported before.17 The area of the FAZ (mm2) was calculated by counting the total numbers of pixels within the region in scale. The FAZ circularity index was calculated as the ratio of the measured area of FAZ to the area of the projected circle (same perimeter as the identified FAZ) with a range of 0 to 1. A ratio closer to 0 indicated an irregular shape of FAZ, whereas closer to 1 indicated a circular shape.25 Total vessel density (VD) was calculated as the percentage of area not defined as nonperfusion regions over the total area excluding the central Early Treatment Diabetic Retinopathy Study 1-mm circle. The binarized image was also skeletonized and the fractal dimension (FD) was then calculated by using the box-counting method. Vessel diameter index (VDI) was calculated as the area occupied by blood vessels from the binarized image over the total length of blood vessels from the skeletonized image.21

    Statistical Analysis

    Statistical analysis was performed using SPSS, version 24 (IBM). An independent-sample t test was used to compare baseline demographics and evaluate differences in OCT-A metrics between amblyopic and nonamblyopic eyes. The OCT-A metrics (FAZ area, FAZ circularity, VD, FD, and VDI) were log-transformed because of their skewed distribution. A 2-sided P value of .05 was used to test for statistical significance. Linear regression analyses were performed to determine BCVA associations with log-transformed OCT-A metrics (independent variables) after adjustment for confounders, including age, sex, body mass index (calculated as weight in kilograms divided by height in meters squared), spherical equivalent, AL, OCT-A quality score, and central subfield thickness. We compared standardized regression coefficients (sβ) with higher sβ values, indicating stronger associations with OCT-A metrics.

    Results

    A total of 1218 participants were recruited for this study. Of these, 143 (11.7%) were excluded because of low-quality scores (30 [21.0%]), motion artifacts (45 [31.5%]), blurry images or tilted images (60 [42.0%]), signal loss (5 [3.5%]), and poor centration (3 [2.1%]), leaving a total of 1075 eyes from 1075 participants for the final analysis. The mean (SD) age of the study participants was 7.64 (0.98; range, 5.52-8.77) years. There were 530 boys (49.3%) and 545 girls (50.7%). Among them, 30 (2.8%) had amblyopia and 1045 (97.2%) were controls. There were 3 cases of bilateral amblyopia, and the worse eye was chosen for analysis. Comparative analyses of the baseline demographic characteristics of the participants in this report (OCT-A cases) with those in the whole HKCES did not show a difference (eTable in the Supplement).

    Table 1 shows the demographic characteristics of the amblyopic cases and controls. The baseline characteristics, including age, sex, AL, body mass index, central subfield thickness, and quality score of OCT-A imaging, were not different between groups. As expected, there was a difference between the spherical equivalent of amblyopic eyes and control eyes.

    Histogram plots of OCT-A metrics showed skewed distribution. Therefore, log-transformed OCT-A metrics were used in a linear regression analysis. Table 2 shows the comparison between the 2 groups in terms of FAZ, FAZ circularity, fractal dimension, VDI, and VDI after adjustment for age, sex, AL, quality score of OCT-A, and central subfield thickness. The mean (SD) FAZ circularity was lower in the amblyopic group compared with controls (0.549 [0.145] vs 0.611 [0.103]; P = .002), indicating an irregular shape of the FAZ. Furthermore, the amblyopic eyes had a lower mean (SD) FD (1.721 [0.15] vs 1.734 [0.28]; P = .008) and larger mean (SD) VDI (0.021 [0.005] vs 0.018 [0.003]; P < .001) when compared with the control eyes. However, we did not find a difference between amblyopic eyes and control eyes in mean (SD) FAZ area (0.290 [0.118] mm2 vs 0.323 [0.137] mm2; P = .24) nor mean (SD) VD (97.03 [2.02] vs 97.20 [1.66]; P = .61). Figure 1 and Figure 2 show the differences between OCT-A output in amblyopic and control eyes.

    A linear regression analysis between OCT-A metrics and BCVA was conducted with adjustment of known covariates as described previously (Table 3). Poorer visual acuity was correlated with reduced FAZ circularity (sβ, −0.133; P < .001) and increased vessel diameter index (sβ, 0.097; P = .001). Best-corrected visual acuity was not associated with FAZ area (sβ, −0.045; P = .09), FD (sβ, −0.029; P = .35), and VD (sβ, −0.009, P = .75).

    Discussion

    This population-based study found a reduced FAZ circularity and FD and an increased VDI in amblyopic eyes compared with control eyes as measured by swept-source OCT-A. Reduced FAZ circularity and increased VDI was associated with poorer visual acuity. This article identified new OCT-A metrics with a potential to be used as objective tools in the diagnosis and monitoring of amblyopia and its treatment in children.

    Foveal avascular zone area and VD were the main parameters used in previous OCTA studies to quantify microvasculature in amblyopia, but they have been shown to be affected significantly by confounders and may not be the most sensitive markers to assess foveal health and correlate with central vision.17 We believe unitless parameters, such as FAZ circularity, VDI, and FD, represented as a ratio are more reliable and reproducible measurements to monitor structural changes in the retinal microvasculature. We did not find a difference between amblyopic and control eyes in FAZ area and VD in the superficial capillary layer. We also did not find a correlation between these 2 parameters with logMAR visual acuity. Prior studies analyzing FAZ and VD alone have given inconsistent results. Yilmaz et al11 and Lonngi et al12 reported a lower VD in the superficial and deep capillary plexus of amblyopic eyes. A recent cross-sectional study with 85 children with amblyopia with 66 age-matched controls showed reduced VD in the superficial capillary plexus only but not the deep capillary plexus.16 However, like this study, Demirayak et al15 did not find a difference in retinal capillary plexus densities between amblyopic eyes, controls, and fellow eyes of patients with unilateral amblyopia. Animal studies also did not reveal a difference in VD between amblyopic eyes and controls.26,27 These inconsistencies could be attributed to small sample size, heterogenous study populations, and assessment methods. In addition, it is known that several covariates can be significantly associated with FAZ and VD measurements and should be adjusted before comparing data across cases.18,19 The VD of the deep capillary plexus was also included in many studies, and artifacts from projections from the superficial layer could lead to bias.18 High interobserver variability using the same machine when measuring the FAZ in correspondence to the deep capillary plexus has been reported.28 A study by our group on adult diabetic retinopathy using the Triton DRI OCT with OCT-A metrics obtained by the same algorithm showed high intrasession repeatability and intersession reproducibility.17 In diabetic retinopathy, decreased FAZ circularity was also shown to be associated with a reduction in visual function.17 Identifying reliable and reproducible data from OCT-A images is the one of the milestones for clinically applying OCT-A in amblyopia management.

    It remains to be determined whether the documented abnormal microvascularization seen in the amblyopic eye is a cause or consequence of amblyopia, and this study cannot imply a causal relationship. Loss of FAZ circularity has been shown to be a good indicator of vascular dropout and is associated with disease progression in vascular maculopathy.29 Similarly, FD quantifies the geometric branching network of retinal vasculature and reflects microvascular dropout.30 The reduced FAZ circularity and FD in amblyopic eyes, as shown in our study, may indicate ischemia in the fovea and less efficient blood distribution in their retinal branching network, respectively. In addition, a higher VDI was found in amblyopic eyes that was associated with poorer vision, suggesting that retinal vessel widening is seen in amblyopic eyes. Wider retinal vessels have been shown in previous studies to be associated with diabetic retinopathy severity and endothelial dysfunction, inflammation, and microvascular hypoxia.31,32 To our knowledge, our study is one of the first to document such microvascular changes in amblyopia. While the underlying pathophysiology is uncertain, the vascular changes in amblyopic eyes may possibly be an early sign of changes in retinal neuron metabolism. It has been shown in animal studies that in a maturing retina, neurons are an important source of vascular endothelial growth factor to control the development of the superficial and deep vascular layers.33 Changes in the neuronal level, with lengthening of the photoreceptors outer segment, can occur in the presence of monocular deprivation.8 Studies using high-resolution OCT revealed that the normal bulge at the central foveal junction between the inner and outer segments was more likely to be absent in amblyopic eyes, with an increase in foveal to parafoveal thickness ratios in amblyopic eyes.10 These findings suggest foveal immaturity and underdeveloped photoreceptors in amblyopic eyes. In fact, a trend toward reduction in the foveal thickness in the amblyopic eye was reported by Nishi et al9 after successful optical treatment. It has been shown in animal models that impaired photoreceptor retinal ganglion cell and photoreceptor survival are associated with abrogated vascular development,33 again suggesting that the metabolic needs of the neural retina can be associated with blood vessel supply in development. Future work looking into the changes of the abnormal vasculature, structural macular, and neuronal layer over time with amblyopia treatment is required to clarify the underlying pathophysiology of this phenomenon in amblyopia.

    Strengths and Limitations

    Strengths of our study include a standardized image acquisition protocol and strict definition of image artifacts for quality control in a reading center. The fully automated custom MATLAB program that we recently developed is able to quantify detailed information from the macula capillary network from OCT-As.20 Two dedicated assessors of the images were also masked to the study participants’ initial diagnoses. Based on the results from our previous population-based study on the distributions of quantitative OCT-A metrics, we ensured accuracy of the results by adjusting for all known covariates.

    Limitations of this study should be noted. We did not use a specific scoring system for artifacts, such as image tilt,34 for our OCT-A quality control. In the calculation of vessel density, our measurement excluded the central Early Treatment Diabetic Retinopathy Study 1-mm circle at the center but not specifically the FAZ. This reduces the effect of FAZ as an interacting variable but does not completely eliminate its effect. Furthermore, the measurement of OCT-A metrics includes retinal vessels and retinal capillaries and it does not separate arterioles and venules. Changes in OCT-A metrics may reflect changes in arterioles or venules or both. Associated with study design, only 1 eye of each patient was recruited in the study and interocular correlation cannot be commented in this study. Because of the cross-sectional nature of our data, we could only establish an association, but not a causal relationship, between amblyopia and OCT-A metrics. We also did not explore specific OCT macula findings of foveal pit excavation, foveal bulge, and ganglion cell analysis in association with amblyopia, which have been previously shown to be different in amblyopic eyes.8-10 Future studies are needed to determine how these primary structural changes may affect the association with macular microvasculature as measured by OCT-A. Lastly, our study population includes a narrow racial and age band, hence the study results may not be applicable to other populations.

    Conclusions

    This population-based study supports the presence of morphological changes in the macular microvasculature in amblyopic eyes of children. Such changes are associated with poorer visual acuity, and we believe that OCT-A and quantitative OCT-A metrics, such as FAZ circularity, VDI, and FD, have a potential to be used as a reliable, objective, and automated tool in the diagnosis and recognition of amblyopia.

    Back to top
    Article Information

    Accepted for Publication: May 9, 2020.

    Corresponding Author: Jason C. Yam, FRCSEd, Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong Eye Hospital, 147K Argyle St, Kowloon 999077, Hong Kong (yamcheuksing@cuhk.edu.hk).

    Published Online: June 25, 2020. doi:10.1001/jamaophthalmol.2020.2220

    Author Contributions: Dr Yam had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Drs Wong and Zhang are co–first authors and contributed equally.

    Concept and design: Wong, Li, Pang, Chen, Tham, Yam.

    Acquisition, analysis, or interpretation of data: Wong, Zhang, Yuan, Li, Cheung, Yam.

    Drafting of the manuscript: Wong, Yam.

    Critical revision of the manuscript for important intellectual content: Wong, Zhang, Yuan, Li, Pang, Chen, Tham, Cheung.

    Statistical analysis: Wong, Zhang.

    Obtained funding: Pang, Tham, Yam.

    Administrative, technical, or material support: All authors.

    Supervision: Li, Tham, Yam.

    Conflict of Interest Disclosures: Dr Tham reports consultant fees from Alcon Laboratories, Allergan, Bausch + Lomb, IOPtima, Merck, Pfizer, Santen Pharmaceutical, Sensimed, and C-MER Eye Care Holdings; nonfinancial support for serving expert testimony; grants from Aeon Astron, Alcon Laboratories, AMO Asia, Icare Finland, Pfizer, Santen Pharmaceutical, and Sensimed; lecture fees from Alcon Laboratories, Allergan, Merck, Pfizer, and Santen Pharmaceutical; and meeting fees from Alcon Laboratories, Allergan, Merck, Pfizer, and Santen Pharmaceutical outside the submitted work. No other disclosures were reported.

    Funding/Support: The study was supported in part by Chinese University of Hong Kong Jockey Club Children Eye Care Programme; the General Research Fund; Research Grants Council, Hong Kong (grant 14111515 [Dr Yam]); UBS Optimus Foundation (grant 8984 [Dr Yam]) and the direct grants of the Chinese University of Hong Kong (grants 4054121 and 4054199 [Dr Yam]); and the Health and Medical Research Fund, Hong Kong (grant 07180826 [Dr Zhang]).

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: We thank the children who participated in our study and the families who gave them support in the Hong Kong Children Eye Study. We also thank our colleagues and volunteers for all their hard work and dedication in collecting the data for this study. We thank Mandy P. H. Ng, MHM, Department of Ophthalmology and Visual Sciences, the Chinese University of Hong Kong, for her help in logistics and coordinating the study. None of these individuals were compensated for their contributions.

    References
    1.
    McKean-Cowdin  R, Cotter  SA, Tarczy-Hornoch  K,  et al; Multi-Ethnic Pediatric Eye Disease Study Group.  Prevalence of amblyopia or strabismus in Asian and non-Hispanic white preschool children: multi-ethnic pediatric eye disease study.   Ophthalmology. 2013;120(10):2117-2124. doi:10.1016/j.ophtha.2013.03.001 PubMedGoogle Scholar
    2.
    Chia  A, Dirani  M, Chan  YH,  et al.  Prevalence of amblyopia and strabismus in young Singaporean Chinese children.   Invest Ophthalmol Vis Sci. 2010;51(7):3411-3417. doi:10.1167/iovs.09-4461 PubMedGoogle Scholar
    3.
    Peterseim  MM, Papa  CE, Wilson  ME,  et al.  The effectiveness of the Spot Vision Screener in detecting amblyopia risk factors.   J AAPOS. 2014;18(6):539-542. doi:10.1016/j.jaapos.2014.07.176 PubMedGoogle Scholar
    4.
    Amblyopia  CE.  Amblyopia.   Surv Ophthalmol. 1995;40(1):23-39. doi:10.1016/S0039-6257(95)80044-1 PubMedGoogle Scholar
    5.
    Stryker  MP, Harris  WA.  Binocular impulse blockade prevents the formation of ocular dominance columns in cat visual cortex.   J Neurosci. 1986;6(8):2117-2133. doi:10.1523/JNEUROSCI.06-08-02117.1986 PubMedGoogle Scholar
    6.
    Levi  DM.  Visual processing in amblyopia: human studies.   Strabismus. 2006;14(1):11-19. doi:10.1080/09273970500536243 PubMedGoogle Scholar
    7.
    Huberman  AD, Feller  MB, Chapman  B.  Mechanisms underlying development of visual maps and receptive fields.   Annu Rev Neurosci. 2008;31:479-509. doi:10.1146/annurev.neuro.31.060407.125533 PubMedGoogle Scholar
    8.
    Bruce  A, Pacey  IE, Bradbury  JA, Scally  AJ, Barrett  BT.  Bilateral changes in foveal structure in individuals with amblyopia.   Ophthalmology. 2013;120(2):395-403. doi:10.1016/j.ophtha.2012.07.088 PubMedGoogle Scholar
    9.
    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. doi:10.1136/bjophthalmol-2014-305685 PubMedGoogle Scholar
    10.
    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. doi:10.1136/bjophthalmol-2012-302434 PubMedGoogle Scholar
    11.
    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. doi:10.1016/j.jaapos.2017.05.002 PubMedGoogle Scholar
    12.
    Lonngi  M, Velez  FG, Tsui  I,  et al.  Spectral-domain optical coherence tomographic angiography in children with amblyopia.   JAMA Ophthalmol. 2017;135(10):1086-1091. doi:10.1001/jamaophthalmol.2017.3423 PubMedGoogle Scholar
    13.
    Sobral  I, Rodrigues  TM, Soares  M,  et al.  OCT angiography findings in children with amblyopia.   J AAPOS. 2018;22(4):286-289.e2. doi:10.1016/j.jaapos.2018.03.009PubMedGoogle Scholar
    14.
    Borrelli  E, Lonngi  M, Balasubramanian  S,  et al.  Increased choriocapillaris vessel density in amblyopic children: a case-control study.   J AAPOS. 2018;22(5):366-370. doi:10.1016/j.jaapos.2018.04.005 PubMedGoogle Scholar
    15.
    Demirayak  B, Vural  A, Onur  IU, Kaya  FS, Yigit  FU.  Analysis of macular vessel density and foveal avascular zone using spectral-domain optical coherence tomography angiography in children with amblyopia.   J Pediatr Ophthalmol Strabismus. 2019;56(1):55-59. doi:10.3928/01913913-20181003-02 PubMedGoogle Scholar
    16.
    Chen  W, Lou  J, Thorn  F,  et al.  Retinal microvasculature in amblyopic children and the quantitative relationship between retinal perfusion and thickness.   Invest Ophthalmol Vis Sci. 2019;60(4):1185-1191. doi:10.1167/iovs.18-26416 PubMedGoogle Scholar
    17.
    Tang  FY, Ng  DS, Lam  A,  et al.  Determinants of quantitative optical coherence tomography angiography metrics in patients with diabetes.   Sci Rep. 2017;7(1):2575. doi:10.1038/s41598-017-02767-0 PubMedGoogle Scholar
    18.
    Cheung  CY, Li  J, Yuan  N,  et al.  Quantitative retinal microvasculature in children using swept-source optical coherence tomography: the Hong Kong Children Eye Study.   Br J Ophthalmol. 2018;103:672-679. doi:10.1136/bjophthalmol-2018-312413 PubMedGoogle Scholar
    19.
    Hsu  ST, Chen  X, Ngo  HT,  et al.  Imaging infant retinal vasculature with OCT angiography.   Ophthalmol Retina. 2019;3(1):95-96. doi:10.1016/j.oret.2018.06.017 PubMedGoogle Scholar
    20.
    Sun  Z, Tang  F, Wong  R,  et al.  OCT angiography metrics predict progression of diabetic retinopathy and development of diabetic macular edema: a prospective study.   Ophthalmology. 2019;126(12):1675-1684. doi:10.1016/j.ophtha.2019.06.016 PubMedGoogle Scholar
    21.
    Kim  AY, Chu  Z, Shahidzadeh  A, Wang  RK, Puliafito  CA, Kashani  AH.  Quantifying microvascular density and morphology in diabetic retinopathy using spectral-domain optical coherence tomography angiography.   Invest Ophthalmol Vis Sci. 2016;57(9):OCT362-OCT370. doi:10.1167/iovs.15-18904 PubMedGoogle Scholar
    22.
    Yam  JC, Tang  SM, Kam  KW,  et al.  High prevalence of myopia in children and their parents in Hong Kong Chinese Population: the Hong Kong Children Eye Study.   Acta Ophthalmol. 2020. Published online January 24, 2020. doi:10.1111/aos.14350PubMedGoogle Scholar
    23.
    Stanga  PE, Tsamis  E, Papayannis  A, Stringa  F, Cole  T, Jalil  A.  Swept-Source Optical Coherence Tomography Angio™ (Topcon Corp, Japan): technology review.   Dev Ophthalmol. 2016;56:13-17. doi:10.1159/000442771 PubMedGoogle Scholar
    24.
    Tang  F, Cheung  CY.  Quantitative retinal optical coherence tomography angiography in patients with diabetes without diabetic retinopathy.   Invest Ophthalmol Vis Sci. 2017;58(3):1766. doi:10.1167/iovs.17-21552 PubMedGoogle Scholar
    25.
    Domalpally  A, Danis  RP, White  J,  et al.  Circularity index as a risk factor for progression of geographic atrophy.   Ophthalmology. 2013;120(12):2666-2671. doi:10.1016/j.ophtha.2013.07.047 PubMedGoogle Scholar
    26.
    Cleland  BG, Crewther  SG, Crewther  DP.  The cat as a model for visual deprivation.   Aust N Z J Ophthalmol. 1985;13(3):263-269. doi:10.1111/j.1442-9071.1985.tb00433.x PubMedGoogle Scholar
    27.
    Crewther  SG, Crewther  DP, Cleland  BG.  Convergent strabismic amblyopia in cats.   Exp Brain Res. 1985;60(1):1-9. doi:10.1007/BF00237012 PubMedGoogle Scholar
    28.
    Sacconi  R, Borrelli  E, Querques  G.  Reproducibility of vessel density, fractal dimension, and foveal avascular zone using 7 different optical coherence tomography angiography devices.   Am J Ophthalmol. 2018;192:252-253. doi:10.1016/j.ajo.2018.05.020 PubMedGoogle Scholar
    29.
    Mo  S, Krawitz  B, Efstathiadis  E,  et al.  Imaging foveal microvasculature: optical coherence tomography angiography versus adaptive optics scanning light ophthalmoscope fluorescein angiography.   Invest Ophthalmol Vis Sci. 2016;57(9):OCT130-OCT140. doi:10.1167/iovs.15-18932 PubMedGoogle Scholar
    30.
    Bhardwaj  S, Tsui  E, Zahid  S,  et al.  Value of fractal analysis of optical coherence tomography angiography in various stages of diabetic retinopathy.   Retina. 2018;38(9):1816-1823. doi:10.1097/IAE.0000000000001774 PubMedGoogle Scholar
    31.
    Nguyen  TT, Wong  TY.  Retinal vascular manifestations of metabolic disorders.   Trends Endocrinol Metab. 2006;17(7):262-268. doi:10.1016/j.tem.2006.07.006 PubMedGoogle Scholar
    32.
    Sun  C, Wang  JJ, Mackey  DA, Wong  TY.  Retinal vascular caliber: systemic, environmental, and genetic associations.   Surv Ophthalmol. 2009;54(1):74-95. doi:10.1016/j.survophthal.2008.10.003 PubMedGoogle Scholar
    33.
    Joyal  JS, Gantner  ML, Smith  LEH.  Retinal energy demands control vascular supply of the retina in development and disease: the role of neuronal lipid and glucose metabolism.   Prog Retin Eye Res. 2018;64:131-156. doi:10.1016/j.preteyeres.2017.11.002 PubMedGoogle Scholar
    34.
    Holmen  IC, Konda  MS, Pak  JW,  et al.  Prevalence and severity of artifacts in optical coherence tomographic angiograms.   JAMA Ophthalmol. 2019;138(2):119-126. doi:10.1001/jamaophthalmol.2019.4971 PubMedGoogle Scholar
    ×