Amblyopia is defined as a condition with a minimum of two lines of visual loss without any organic pathology that occurs in 1% and 4% of individuals younger than 17 years. Visual acuity can be improved if detected and properly treated early.1
In previous studies, deviations from the normal structure were demonstrated in the choroid, fovea, macula, and peripapillary areas of amblyopic eyes by optic coherence tomography angiography (OCTA) imaging.2,3 OCTA is a non-invasive and dye-free imaging technique that detects the velocity of erythrocytes in the vessel and processes it to convert volumetric data and calculate the vessel density quantitatively.4
Because specifically measuring the superficial and deep vessel densities of the retina was not possible until recently, data on the superficial and deep vessel densities in amblyopic eyes are limited. There are a few studies about OCTA imaging of amblyopic eyes, but no consensus was achieved about the alterations of retinal microvascular structure in amblyopia. In addition to studies that revealed decreased vascular densities in the superficial and deep capillary plexus, there are reports that identified a decreased vascular density in the superficial capillary plexus only or no significant change in both layers.5–9
This study aimed to use OCTA to detect the existence of any vascular dysfunction in superficial and deep retinal layers that accompanies visual impairment in amblyopic eyes compared to fellow and control eyes.
Patients and Methods
This cross-sectional study was conducted in an outpatient hospital clinic with participants between 6 and 18 years of age. Informed consent was obtained from the families of all participants. This study was approved by Alaaddin Keykubat University's Ethics Committee.
Amblyopia was considered to be a minimum of two Snellen lines of difference in visual acuity between the two eyes and the lack of any organic pathology on biomicroscopy and funduscopy that may cause any visual impairment. Eyes with strabismic amblyopia were excluded from the current study to avoid inaccuracies on OCTA measurements caused by fixation anomalies. Eyes with deprivation amblyopia, nystagmus, bilateral amblyopia, corneal opacity, and a history of intraocular surgery were also excluded. Eyes with spherical refractive errors of 5.00 diopters (D) or greater and cylindrical refractive errors of 3.00 D or greater were not included in the study because of their disrupting effect on OCTA measurement quality. The authors randomly selected one eye from age- and gender-matched individuals referred to our outpatient clinic for a routine eye examination who were recruited as control eyes.
A complete ophthalmic examination including cycloplegic autorefractometry, best corrected visual acuity (determined 3 days after cycloplegic autorefractometry), and anterior segment and dilated posterior segment examination was performed in all individuals.
OCTA Study Protocol
A 3 × 3 mm scale map was used for all macular measurements. Vascular density was measured by the Topcon DRI OCT Triton device (Topcon Corporation, Tokyo, Japan), which used the internal limiting membrane as the reference plane and considered the area from 2.6 to 15.6 µm as the superficial retinal layer. Using the inner plexiform layer as the reference plane, the area from 15.6 to 70.2 µm was considered the deep retinal layer. The vascular densities of the superficial and deep capillary plexus were calculated as percentages.
The distribution of values was analyzed by the Kolmogorov–Smirnov test and a categorization of parameters was analyzed by the chi-square test. Study data were compatible with the normal distribution. Study parameters were analyzed with an independent t test between eyes in all groups. SPSS software (version 21.0; SPSS, Inc., Chicago, IL) was used for statistical analysis. P values less than .05 were considered statistically significant.
Thirty-seven patients between 6 and 18 years of age with anisometropic amblyopia and 37 age- and gender-matched individuals without amblyopia were included in the current study. No significant differences in age or gender were observed between the participants (P > .05). Demographic and clinical parameters for all of the participants are listed in Table 1.
Demographic and Clinical Parameters of Participants
Visual acuity was lower and spherical equivalent of cycloplegic refractive error was significantly higher in amblyopic eyes than fellow eyes (P < .05). Visual acuity was significantly lower in amblyopic eyes than control eyes (P < .05). The spherical equivalent of cycloplegic refraction was significantly higher in amblyopic eyes than control eyes (P < .05) (Table 1).
The superficial and deep vascular densities of all eyes in the superior, inferior, temporal, and nasal quadrants of the foveal and parafoveal areas obtained by OCTA are listed in Table 2. A significant decrease in the superficial and deep capillary plexus in the superior quadrant was seen in amblyopic eyes when compared to fellow and control eyes, whereas no significant differences were seen in the other quadrants (Table 2). Box-plot analyses of the superficial and deep retinal plexus in the superior quadrants of all eyes are demonstrated in Figure 1.
SCP and DCP Vascular Densities of All Eyes in the Superior, Inferior, Temporal, and Nasal Quadrants of the Foveal and Parafoveal Areas
Box-plot analysis of the (A) superficial (SCP) and (B) deep (DCP) retinal plexus in the superior quadrants of all eyes.
No significant differences in the foveal avascular zone was observed between the groups (P > .05).
The American Academy of Ophthalmology defined amblyopia as visual loss that cannot be improved by eyeglasses in one or two eyes without any detectable organic pathology.10 Insufficient stimulation of the foveal area in early ages, abnormal binocular vision, asymmetry in visual acuity between two eyes, or a combination of those factors were identified as possible causes of amblyopia in experimental studies.11
Superficial and deep vascular densities of the retina in amblyopic eyes were non-invasively evaluated by OCTA in the current study. Mean vascular densities of both superficial and deep layers in the foveal and parafoveal areas were not significantly different in amblyopic eyes when compared to fellow and control eyes. However, when divided into quadrants, vascular densities of the superficial and deep layers were significantly decreased in the parafoveal superior quadrant in amblyopic eyes.
There are studies evaluating the retinal vascular structure of amblyopic eyes in the literature. Yilmaz et al.5 reported a significant decrease in vascular densities of the superficial and deep capillary plexus in eyes with strabismic amblyopia measured by OCTA in 1, 2, and 3 mm areas. Lonngi et al.6 reported a significant decrease in vascular density in the superficial and deep layers that was prominent in 6 × 6 mm measurements of the deep capillary plexus in their study comparing 13 eyes with strabismic or anisometropic amblyopia with 50 control eyes. Demirayak et al.7 found no significant difference in the vascular densities of the superficial and deep capillary plexus between 17 amblyopic and 21 control eyes.
Chen et al.8 reported the mean decrease in vascular density of the superficial capillary plexus as 3.17% in the foveal area and 2.76% in the parafoveal area, whereas the difference in the deep capillary plexus was not significant in their study of 52 anisometropic, 16 strabismic, 17 bilateral amblyopic, and 66 healthy control eyes. A positive correlation was found between inner retinal thickness and vascular density of the superficial and deep capillary plexus of the foveal area in the same study. Furthermore, deterioration was observed in all quadrants. In Chen et al.'s study, deteriorations in all quadrants except the superior quadrant of the superficial capillary plexus were found to be correlated with visual acuity.
Pujari et al.9 reported no significant difference between eyes with strabismic amblyopia and the fellow eyes of 10 participants in their study that evaluated the superficial capillary plexus in all quadrants with a 4.5 × 4.5 mm map using the Triton DRIOCTA (Topcon Medical Systems, Inc., Oakland, NJ). Our study is compatible with studies by Yilmaz et al.,5 Lonngi et al.,6 and Demirayak et al.7 in that we observed no significant differences in mean vascular densities of the superficial and deep capillary plexus in amblyopic eyes.
In a study combined with the split-spectrum amplitude-decorrelation angiography (SSADA) and fundus autofluorescence, a decrease in macular signal autofluorescence induced with blue light was detected in 72% of 22 amblyopic eyes, but no significant change in vascular structure was detected by SSADA in the superficial and deep layers of the retina.12 An increase in choroidal thickness and patchy atrophic areas in 68.2% of amblyopic eyes was reported by the same authors. Sobral et al.13 observed a significant decrease in the vascular densities of the superficial and deep capillary plexus in amblyopic eyes when compared to healthy control eyes, but they reported a significant decrease in the vascular density of the superficial capillary plexus in amblyopic eyes when compared to fellow eyes. Their study demonstrates a vascular dysfunction in both amblyopic and fellow eyes.
In the report by Doguizi et al.,14 40 eyes with anisometropic amblyopia with a mean refractive error (+4.96 ± 1.43 D) were compared to 57 healthy control eyes with the RTVue XR Avanti (AngioVue; Optovue Inc., Fremont, CA). They reported a significantly lower vascular density in the foveal, parafoveal, and perifoveal areas of the superficial and deep capillary plexus (except for the inferior perifoveal region) in eyes with hyperopic and anisometropic amblyopia compared to fellow and control eyes before and after analysis adjusted for axial length and refraction. No comment is available about separating the inferior quadrant of the perifoveal area from vascular deterioration in amblyopia.
We observed deterioration in the superior quadrant of the superficial and deep vascular layers in amblyopic eyes. Similarly, Li et al.15 demonstrated a vascular deterioration in the superficial and deep capillary plexus of myopic eyes except in the superior-temporal quadrant. Doguizi et al.14 demonstrated a deterioration in microcirculation of macula conserving the inferior perifoveal region in hyperopic eyes, whereas Li et al.15 demonstrated a deterioration in macular microcirculation conserving the superior-temporal region in myopic eyes. Accordingly, refractive status may be related to the localized damage of macular microcirculation. Further studies combined with histological analysis are needed to understand if some retinal diseases have any tendency to hold specific regions of the macula.
We showed deterioration only in the superior quadrant of the parafoveal superior and deep capillary plexus of the amblyopic eyes in our study. Although it is unclear that the damage in vascular structure exhibits a specific localization, there are reports pointing out localized defects in vascular structure. Chen et al.8 reported that retinal perfusion deteriorated in all quadrants in amblyopic eyes, but the deterioration in the superior quadrant was not correlated with visual acuity. In their study evaluating microvasculature of the macula in myopic eyes, Li et al.15 reported vascular deterioration in deep and superficial layers, except in the superior temporal quadrant. Yu et al.16 evaluated the macular microvasculature in 18 perfusion-labeled human donor eyes using confocal microscopy histologically and OCTA. They ranked vessel density in decreasing order as inferior, superior, temporal, and nasal in the superficial and deep layers in their report. Additionally, the age-related increase in vessel density was observed in the superficial layers of macula. The authors indicated that this finding may indicate the importance of a stable blood supply for the human macula. Based on these results, they postulated that regional differences may reflect a well-matched vascular supply and neuronal demands. Although results of previous studies suggest that pathology is location specific in some macula-related diseases, additional studies in large study groups and histological correlations of the results obtained from optical imaging techniques are needed.
Although amblyopia is defined as visual insufficiency with the lack of detectable organic pathology, cellular contraction and shrinkage in the lateral geniculate nucleus, an increase in amacrine cells synapses in inner plexiform layer of the retina, a decrease in bipolar cells synapses, a decrease in inner plexiform layer thickness, and Müller cell density were reported in histological studies, indicating a pathology at the cell level.17–19 The deterioration of vascular structure observed in our study is compatible with the previous studies that reported damage in amblyopic eyes histopathologically.
Our study and previous studies in the literature reveal localized and layer-by-layer deterioration of the retinal vascular structure in amblyopic eyes. Discordancy of the data in the literature may originate from various reasons. First, it was demonstrated that macular functions measured by multifocal electro-retinography had a different response of magnocellular and parvocellular neurons in the cortical visual pathway, and alterations of the response to the visual stimulation in the central nervous system differed between the patients with anisometropic amblyopia and strabismic amblyopia.20–22 Similarly, we believe that retinal vascular structure may alternate according to type of amblyopia. For this reason, we included only patients with anisometropic amblyopia to achieve a pure study group. Second, any eye movement away from the fixation point during the OCTA procedure may cause misalignment problems and affect the test results dramatically, even if it is minor in degree. Combining eyes with anisometropic amblyopia and strabismic amblyopia in the same study group may result in inaccurate OCTA measurements. Third, the reproducibility, repeatability, and quality of measurements within the 3 × 3 mm area is better than the 6 × 6 mm area.23 We used the 3 × 3 mm measuring scale in the current study to achieve more qualified results. Trachsler et al.24 reported significant differences in the foveal avascular zone parameters obtained from four different OCTA devices. Considering the type of OCTA device may be recommended when comparing the results of different studies in the literature.
The limitations of our study are the cross-sectional design and small sample size. Patients with anisometropic amblyopia were not divided into subgroups of patients who did and did not apply eye patching, but we plan to evaluate this issue in our future studies. Due to the exclusion of patients who had a spherical refractive error greater than 5.00 D and cylindrical refractive error greater than 3.00 D, our study may not reflect the entire amblyopic population.
The current study may reveal an anatomical pathology in amblyopic eyes by identifying localized dysfunction in retinal vasculature, compatible with the literature. Although no detectable organic pathology has been defined in amblyopia, in light of new technological developments, it may be possible to demonstrate anatomical and functional defects in future studies.