Assessment of Foveal Avascular Zone and Macular Vascular Plexus Density in Children With Unilateral Amblyopia: A Systemic Review and Meta-Analysis

Purpose: To examine the quantitative measurements of OCTA in children with amblyopia using the meta-analysis methodology. Methods: PubMed, Embase, and Cochrane library were searched for available papers up to March 2021. Weighted mean differences (WMD) were used to compare the retina parameters between the eyes with amblyopia and the contralateral eyes or healthy control eyes. Results: Twelve studies were included. When considering the parafovea, the microvessel density was reduced in amblyopic eyes compared with healthy control eyes in the superficial capillary plexus (SCP) in 6 × 6 volume scan (WMD = −2.12, 95%CI: −3.24, −0.99) but not SCP in 3 × 3 volume scan (WMD = −1.43, 95%CI: −2.96, 0.11). In the deep capillary plexus (DCP), amblyopia did not decrease vessel density in the 6 × 6 volume scan (WMD = −2.22, 95%CI: −5.86, 1.42; I2 = 79.6%, P = 0.008), but a difference was observed in the whole eye 3 × 3 (WMD = −1.95, 95%CI: −3.23, −0.67; I2 = 27.5%, P = 0.252). There were no significant differences in the foveal avascular zone area and foveal thickness between amblyopic eyes and healthy control eyes. There were no significant differences in microvessel density, foveal avascular zone area, and foveal thickness between amblyopic eyes and fellow eyes. Conclusion: According to OCTA, amblyopic eyes had lower vessel density in parafoveal SCP and DCP compared with healthy control eyes, but not compared with fellow eyes. There were no significant differences regarding the foveal avascular zone area and foveal thickness between amblyopic and non-amblyopic eyes.


INTRODUCTION
Amblyopia is defined as a reduction in best-corrected visual acuity (2-line difference between the two eyes) secondary to neurological deficits in visual output caused by abnormal brain stimulation during critical periods of visual development (1)(2)(3). It is usually unilateral and is the most common cause of vision loss and mononuclear blindness in children. It remains the most common cause of preventable vision loss in children (4). The worldwide prevalence is 1-5%, and the prevalence in North America is 2-4% (2,3).
In animal models of amblyopia, fewer ganglion cells project to the central visual nuclei (5), and it has been suggested that the postnatal process of ganglion cell reduction is affected by amblyopia and should result in a thicker circumpapillary retinal nerve fiber layer (6). A meta-analysis of optical coherence tomography (OCT) studies revealed that a thicker foveola was found in amblyopic eyes compared with normal eyes (7). On the other hand, it is unknown whether this thicker fovea is associated with changes in blood flow or microvessel density.
OCT angiography (OCTA) is a relatively new and noninvasive imaging technique that uses motion control contrast imaging to obtain high-resolution volumetric blood flow information and generate angiographic images of the retina and choroid (8). The advantages of OCTA over the traditional fluorescein and indocyanine-green angiography are its fast acquisition time and non-invasiveness (9,10). In addition, it provides quantitative measurements (11). It can be used to assess retinal diseases, glaucoma, and uveitis (9,10). Importantly, OCTA can provide quantitative assessments of the superficial capillary plexus (SCP) and deep capillary plexus (DCP) (12). Some studies examined the use of OCTA for amblyopic eyes, but the results are inconsistent, and no meta-analysis is available to synthesize the quantitative measurements of OCTA in children with amblyopia.
Therefore, the aim of the present meta-analysis is to examine the quantitative measurements of OCTA in children with amblyopia. The results could provide a firmer basis for the use of OCTA for the management of amblyopia.

Study Selection
This meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The study eligibility criteria were: (1) children with amblyopia in one eye as the case group, compared with the contralateral eyes or healthy control eyes; (2) quantitative measurement of the foveal avascular zone and macular vascular plexus density using OCTA; (3) language limited to English; and (4) the outcomes could be extracted.
PubMed, Embase, and Cochrane library were searched for available papers up to March 2021 using the MeSH terms and key words pertinent to "amblyopia" and "optical coherence tomography angiography."

Data Extraction and Quality Assessment
The selection and inclusion of studies were performed in two stages by two independent reviewers (Lei Gao and Xiangwen Shu). This included, first, the analysis of the titles and abstracts, followed by the full texts. Any disagreements were resolved by a third reviewer (Peng Zhang).
Data, including names of authors, publication year, study design, sample size, type of amblyopia, OCTA algorithm, and spherical equivalent, were gathered with a structured data collecting form. The observational studies were evaluated according to the Newcastle-Ottawa scale (NOS) (13,14).

Statistical Analysis
All analyses were performed using STATA SE 14.0 (StataCorp, College Station, Texas, USA). Weighted mean differences (WMD) and corresponding 95% confidence intervals (CIs) were used to compare the retina parameters between the eyes with amblyopia and the contralateral eyes or healthy control eyes. Statistical heterogeneity among these studies was calculated using Cochran's Q-test and the I 2 index. Q-test with P < 0.10 and I 2 > 50% indicated high heterogeneity, and the random-effects model was then used; otherwise, the fixed-effects model was used. The publication bias was analyzed using funnel plots. P-values < 0.05 were considered statistically significant.

Foveal Avascular Zone Area in Amblyopic Eyes Compared With Healthy Control Eyes
There was no significant difference between amblyopia eyes and healthy control eyes in foveal avascular zone area in the assumed SCP 6

Macular Vessel Density in Amblyopic Eyes Compared With Fellow Eyes
In the SCP, amblyopia decrease the vessel density when considering the whole eye in 6 × 6 volume scan (WMD = −2.23, 95%CI:  Figure 1C).
In the DCP, amblyopia did not decrease the vessel density when considering the whole eye in 6 × 6 volume scan (WMD = −1.37, 95%CI:     Figure 1D).

Publication Bias
No significant publication bias was observed (Supplementary Figure 2).

DISCUSSION
OCTA is a novel rapid and non-invasive method for the angiography of the retina and choroid (9, 10), but the studies yield conflicting results. Therefore, the aim of this meta-analysis was to examine the quantitative measurements of OCTA in children with amblyopia. The results indicate that according to OCTA, amblyopic eyes had lower vessel density in parafoveal SCP and DCP compared with healthy control eyes, but not compared with fellow eyes. There were no significant differences regarding the foveal avascular zone area and foveal thickness between amblyopic and non-amblyopic eyes.
Because of the small number of patients included in the previous studies, the present meta-analysis increased the power to provide a more convincing assessment of OCTA in amblyopia. Among the parameters measured by OCTA, the parafoveal vessel density was different between amblyopic and healthy control eyes, and this difference was not observed when compared with the fellow eye. This suggests that the pathogenesis of amblyopia involves a reduced vascularization of the affected eye but that this process could also be observed in the fellow normal eye. This was observed by Sobral et al. (23). Chatzistefanou et al. (27) and Meier and Giaschi (28) also observed that unilateral amblyopia affected the fellow eye and that the fellow eye cannot be considered equivalent to a normal healthy eye. Of note, different results were obtained when considering the whole eye in 6 × 6 and 3 × 3 volume scan, suggesting that different measurement methodologies affected the results.
In the present meta-analysis, no difference was observed at the fovea level between amblyopic eyes and normal healthy or fellow eyes. This is in contradiction with a meta-analysis of OCT that  showed thicker fovea in amblyopic eyes compared with control eyes (7). This discrepancy could be due to the differences in the methods of measurement and the included studies.
Nevertheless, there is literature supporting abnormal eye development in amblyopia. Indeed, rapid visual development is observed in the sensitive visual development period, which ends by 6-7 years of age (29). During this critical period, any obstacles to the normal visual environment increase the risk of amblyopia, leading to changes in cortical neurons and their synapses (29). The blurred vision will lead to insufficient stimulation in the amblyopic eye, which affects the maturation of the macula (30), which could contribute to decreased need for oxygen supply and decreased vascularization. The lack of normal vision could also lead to reduced apoptosis of the ganglion cells in the amblyopic eyes, leading to the thicker fovea (6), as observed by Li et al. (7), but this was not observed in the present study.
Of course, the results of the present meta-analysis must be considered in relation to the study limitations. First, unpublished studies, conference abstracts, and non-English studies were not included. In addition, because of the small number of studies, bias might still be present. Second, because of the small number of studies, the influence of treatments could not be assessed. Such treatments have been shown to influence foveal thickness in amblyopia (31), and only two of the included studies reported treatments. Miki et al. (32) and Tugcu et al. (33) reported differences in foveal thickness between different types of amblyopia, but this could not be examined here. Furthermore, all included studies were cross-sectional and case-control studies, and the cause-to-effect relationship of amblyopia and reduced vascularization could not be examined. Third, the correlations between the severity of the amblyopia, the types of amblyopia, and the microvessel density of the amblyopic eyes could not be investigated because such data were not reported in the included studies. We performed subgroup analyses for the types of amblyopia, but the numbers of studies in some subgroups were small. Nevertheless, OCTA is a recent technology, the area is still being explored, and the number of studies is small. Future studies should provide additional data, especially clinical trials.
In conclusion, according to OCTA, amblyopic eyes had lower vessel density in parafoveal SCP and DCP compared with healthy control eyes, but not compared with fellow eyes. There were no significant differences regarding the foveal avascular zone area and foveal thickness between amblyopic and nonamblyopic eyes.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

AUTHOR CONTRIBUTIONS
LG, YG, and XS: substantially contributed to conception or design.
LG and PZ: contributed to acquisition and analysis of data.
LG: drafted the manuscript for important content and gave final approval. YG, FH, PZ, and XS: critically revised the manuscript for important intellectual content. All authors contributed to the article and approved the submitted version.