Refractive Development and Choroidal Vascularity in the Form-Deprivation Pigmented Rabbit Model

Abstract

Purpose: This study assessed the characteristics of refractive development and choroidal vasculature in the form-deprivation (FD) pigmented rabbit model.

Methods: Monocular FD was performed in three-week-old pigmented rabbits (n = 18 for FD, n = 12 for control). Throughout the eight-week rearing period, refractive errors, corneal curvature radius (CCR), ocular biometric parameters, retinal thickness (RT), and choroidal thickness (ChT) were measured every two weeks using cycloplegic retinoscopy, keratometer, A-scan ultrasonography, and optical coherence tomography (OCT). The choroidal vascularity index (CVI) was calculated from OCT images by measuring the total choroidal area (TCA), stromal area (SA), and luminal area (LA). At the end of the form deprivation, the vitreous dopamine level was measured using an enzyme-linked immunosorbent assay kit.

Results: Relatively myopic refraction was induced in FD eyes after two, four, six, and eight weeks (interocular differences: −1.48 ± 0.88, −1.92 ± 0.90, −1.95 ± 0.80, and −2.00 ± 0.83 diopter; P < 0.001). Furthermore, FD eyes showed significantly longer axial length (AL) and vitreous chamber depth after eight weeks, with mean differences of 0.32 ± 0.03 and 0.32 ± 0.05 mm, respectively (P < 0.001). There were no significant differences in anterior chamber depth, lens thickness, CCR, and RT among the three groups through the intervention (all P > 0.05). After eight weeks, the average ChT of FD eyes was thinner than contralateral eyes (−19.37 ± 7.01 µm; P < 0.001). Additionally, the TCA, SA, and LA in FD eyes were smaller after four, six, and eight weeks (all P < 0.05, week 8: 0.3697 ± 0.0639 vs. 0.4272 ± 0.0968, 0.1047 ± 0.0221 vs. 0.1233 ± 0.0328, and 0.2650 ± 0.0459 vs. 0.3039 ± 0.0659 mm², respectively). However, the CVI showed no significant difference among the three groups (P > 0.05). Finally, the concentration of vitreous dopamine was lower in the FD eyes, compared with contralateral and control eyes: 0.18 ± 0.20, 0.40 ± 0.67, and 0.33 ± 0.06 ng/mL, respectively (P < 0.05).

Conclusions: Form deprivation led to a relatively myopic shift in pigmented rabbits and a decrease in vitreous dopamine levels. In addition, with the lengthening of AL, the choroid thinned, but CVI remained unchanged.

Translational Relevance: Our study offered data about the refractive characteristics of pigmented rabbits to investigate myopia mechanisms. The modified method imaged the choroid of the inferior species more clearly, achieving in exploring the changes of choroidal vasculature in vivo.

Introduction

The etiology of myopia is intricate, and both genetic and environmental factors are involved in the development of myopia.¹ However, it is unclear how visual input and subsequent signaling cascade affect eye growth and refractive development. 

The dysfunction of the choroidal vascular system leading to scleral hypoxia has been speculated to be the underlying mechanism of myopia.² The choroid, sandwiching between the retinal pigment epithelium (RPE) and sclera, is the most pigmented and vascular structure of the eye, which is mainly composed of the choriocapillaris and two vascular layers: Sattler's layer and Haller's layer.³ Choroidal thickness (ChT) indicates the choroidal vascular status in various ophthalmic diseases, especially myopia. Recent advancements in optical coherence tomography (OCT) technology have revealed a close link between ChT and refractive development over the past decade.⁴ The ChT change is a proxy for predicting future axial length (AL) in myopia development.⁵ There is some evidence that dopamine, one of the neurotransmitters released from the retina, is involved in the signaling cascade that controls myopia. Retinal dopamine levels have also been found to be positively correlated with ChT, supporting the hypothesis that dopamine is related to retinal signals controlling choroidal response.⁶ Nevertheless, ChT fails to present the details of choroidal structure changes. The choroidal vascularity index (CVI) is a ratio of vascular area to the total choroidal area, as determined from OCT images. It is a reliable way to assess changes in the choroidal vasculature and is now considered a novel parameter for evaluating choroidal changes in myopia progression.⁷ 

Because of their close phylogenetic characteristics and anatomical similarities to human eyes, rabbits are commonly used as a model for retinal and choroidal diseases.^8,9 However, no previous studies have analyzed the vascular layers of choroid using OCT in rabbits, particularly in relation to refractive development. As we know, a relatively myopic change of 1.35 diopter (D) was induced in New Zealand white (NZW) rabbits for six-week monocular form deprivation.¹⁰ Monocular form deprivation represents the classic model of myopia induction.¹¹ NZW rabbit is an albino phenotype caused by the absence or inactivation of tyrosinase,¹² which is involved in dopamine synthesis, leading to functional abnormalities in the RPE and impaired emmetropization in chickens and guinea pigs.^13,14 Pigmented rabbits with healthy fundus tissue may be more suitable for investigating choroidal changes in myopia progression. 

In this study, we evaluated the refractive eye growth in the pigmented rabbit model of form-deprivation (FD) and quantitatively assessed choroidal vasculature by OCT scans. Furthermore, we investigated whether vitreous dopamine in rabbit eyes exhibited similar changes in other animal models of myopia.^15,16 The findings contribute to the possibility of integrating additional research on myopia and fundus disease. 

Material and Methods

Animals

In this study, 30 three-week-old healthy pigmented rabbits (15 male) were used. Animals were housed in independent cages (a base area of 0.3 m², 0.5 m × 0.6 m, and a height of 0.4 m) under fluorescent lamps on a 12-hour light/dark cycle, with a room temperature maintained at 24°C. The standard food and water were available as desired for them. The infant rabbits were provided twice-daily sheep milk before the age of seven weeks.⁸ All procedures and care of animals were approved by the Animal Ethical Committee at Zhongshan Ophthalmic Center, Sun Yat-Sen University (Guangzhou, China), and adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Rabbit Model of FD Myopia

Eighteen rabbits were randomly selected to undergo eight weeks of monocular form deprivation, using opaque white facemasks that covered the right eyes, leaving the mouths, noses, ears, and contralateral eyes uncovered. The facemask, made of sterile latex gloves, was placed around the neck of the animal during the experimental period (Fig. 1A), with a twice-daily check to avoid shedding. Meanwhile, the right eyes of 12 normal age-matched rabbits served as controls. The two-week intervals (weeks 2, 4, 6, and 8) were selected to observe the ocular growth of pigmented rabbits based on our preliminary observation that a steady refractive state of the rabbit eye would appear when the age reached 11 weeks. 

Figure 1.

 

(A) The laboratory photographs of the experimental subjects, including the special occluders in pigmented rabbits. (B) The dopamine levels of vitreous in the FD eyes of pigmented rabbits showing a significant reduction at week 8. ^NSP > 0.05, *P < 0.05.

Refraction and Ocular Measurements

Refractive errors were measured with cycloplegia using a streak retinoscope (YZ24; 66 Vision Tech, Jiangsu China) and trial lenses on awake animals in a dim room after the measurements of corneal curvature radius (CCR) by an experienced optometrist using a handheld keratometer (ARK-30, Nidek, Japan). Rabbits with spherical equivalent (SE) less than 3.00 D or with anisometropia of more than 0.50 D at the first time were excluded from this study. 

AL, anterior chamber depth (ACD), lens thickness (LT), and vitreous chamber depth (VCD), were measured by A-scan ultrasonography (SDK/180/ABS; SDK, Suzhou, China) with a 10 MHz hand-held ultrasound probe in the appropriate place with an accuracy of ± 0.01 mm. The sound velocities of the rabbit eyes were set: the aqueous humor received 1532 m/s, the lens received 1641 m/s, and the vitreous received 1532 m/s. After a drop of 0.4% oxybuprocaine hydrochloride (Benoxil; Santen, Osaka, Japan), the probe tip was placed centrally perpendicular to the corneal plane. Five readings were recorded for each measurement to calculate the mean values for statistical analysis. The ratio of VCD to AL (VCD/AL) was also calculated to better assess the refractive eye growth. 

Optical Coherence Tomography

The retinal and choroidal vascular layers were visualized by 870-nm spectral-domain OCT (SD-OCT, Spectralis; Heidelberg Engineering, Heidelberg, Germany), with enhanced depth imaging (EDI) in anesthetized animals, on the same day after refraction measurements. Horizontal raster scans (30° × 15°, 9.0 mm × 4.5 mm) were acquired from the visual streak, inferior to the medullary rays, tangential to the optic disc border. An eye-tracking system (follow-up mode) was applied during the follow-up visits. Additionally, all eyes were lubricated with 0.9% saline solution to reduce image noise. Images with a quality index < 30 were excluded for further analysis.¹⁷ ImageJ 1.8.0 software (National Institutes of Health, Bethesda, MD, USA) was used for all OCT imaging analyses. The accuracy of the measurement parameters in OCT images was ± 1 µm. 

Retinal thickness (RT, defined as the linear distance between the internal limiting membrane and the surface of RPE) and ChT (defined as the linear distance between the RPE/Bruch's membrane complex and the choroidal-scleral junction) were quantitatively evaluated by two trained examiners. The central location below the optic disc border at a distance of one optic disc diameter and its nasal and temporal locations were manually selected for evaluation (Fig. 2). The averaged values of the two independent measurements for each position served as the final values. 

Figure 2.

 

Measurements of the RT and ChT in the temporal, central, and nasal positions of the OCT image in vivo. (A) En face fundus image of pigmented rabbit right eye; The temporal, central, and nasal locations in the visual streak (approximately one pupil distance below the optic nerve head). (B) The corresponding measurements of ChT and RT in the OCT image.

CVI was calculated from OCT images. The total choroidal area (TCA), the stromal area (SA), and the luminal area (LA) for the center of the OCT images within a horizontal 6000 µm wide area with 3000 µm nasal and 3000 µm temporal borders were determined using the previous method.¹⁸ The OCT images were binarized with the optimal Niblack threshold. The light pixels were defined as the SA, whereas the dark pixels were defined as the LA. LA was obtained by calculating the difference between TCA and SA. The CVI was defined as the ratio of LA to TCA (Fig. 3). 

Figure 3.

 

The analysis of the choroidal vascularity index of the right eye in a pigmented rabbit in vivo. (A) The area of the middle two-thirds of the OCT image (3000 µm on either side of the center) was added to calculations of CVI, TCA, LA, and SA. (B) The binarized OCT image after Niblack's auto local threshold tool. (C) The overlay of the binarized image to the raw OCT image shows the choroidal structure components. The whole area represents TCA, the light area (surrounded by yellow lines) represents SA, the dark area represents LA, and CVI is the ratio of LA to TCA.

Measurement of Dopamine Levels

To minimize the impact of circadian rhythms, we collected vitreous humor from rabbits among the three groups under anesthesia between 8 and 10 AM at the end of 8 weeks of monocular form deprivation.¹⁹ Vitreous humor 300-500 µL per eye was extracted using a 2 mL syringe with a 23 gauge needle inserted into the vitreous cavity 1 mm behind the limbus. After extraction, the vitreous samples were transferred into sterile tubes and kept at −80 °C. All samples were measured simultaneously according to the manufacturer's protocol using a dopamine enzyme-linked immunosorbent assay kit to detect the dopamine concentration (BA E-5300R; Labor Diagnostika Nord, Nordhorn, Germany). 

Statistical Analysis

Statistics were performed with SPSS software version 26.0 (SPSS, Inc., Chicago, IL). Comparisons of SE, CCR, AL, ACD, LT, VCD, VCD/AL, and dopamine levels among the FD, contralateral, and control eyes were made by one-way analysis of variance (ANOVA). Comparisons of ChT and RT in temporal, central, and nasal positions were done by two-way ANOVA. Repeated measures ANOVA were used to compare TCA, LA, SA, and CVI among the FD, contralateral and control eyes. Bonferroni's test or Dunnett’ s test was applied for post-hoc analyses. P < 0.05 was considered statistically significant. 

Results

Refraction

The mean refractions of all eyes in 3-week-old pigmented rabbits without FD interventions were moderate hyperopia. There were no differences in the mean SE in the FD, contralateral, and control eyes (P > 0.05, Table 1; Fig. 4A). Relatively more myopic refraction rates were induced in the FD eyes, compared with the contralateral eyes after 2, 4, 6, and 8 weeks of form deprivation (−1.48 ± 0.49, −1.92 ± 0.51, −1.94 ± 0.41, and −2.01 ± 0.35 D, all P < 0.001, Table 1; Fig. 4A). Form deprivation induced rapid myopia rates in pigmented rabbits with a refractive change rate of −1.436 versus − 0.551 D/month or the FD and contralateral eyes. The development of refraction errors in control eyes was similar to that of the contralateral eyes during the experimental period. 

Table 1.

 

Changes in Spherical Equivalent and Ocular Biometric Parameters Across Time

 

Figure 4.

 

Changes in ocular parameters from week 0 to 8. (A) The spherical equivalent was more myopic in the FD eyes than contralateral and control eyes from week 2 to 8. (B) The cornea became more flattened over time and the corneal curvature radius showed no significance among the FD, contralateral, and control eyes. (C) Axial length in the FD eyes was longer than the contralateral and control eyes from week 2 to 8. (D) Vitreous chamber depth in the FD eyes was elongated than the contralateral and control eyes from week 2 to 8. *P < 0.05, **P < 0.01, ***P < 0.001 between the FD and contralateral eyes; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001, between the FD and control eyes.

Ocular Biometric Parameters

There were no significant differences in the AL, ACD, LT, VCD, and CCR baseline among the FD, contralateral, and control eyes (all P > 0.05; Table 1 and Supplementary Table S1; Fig. 4). After two weeks of form deprivation, AL significantly prolonged in the FD eyes compared to the contralateral and control eyes (13.89 ± 0.26, 13.64 ± 0.22, and 13.63 ± 0.20 mm, respectively; P = 0.003; Table 1; Fig. 4C). The corresponding mean of VCD was 6.66 ± 0.19, 6.43 ± 0.18, and 6.43 ± 0.22 mm (P < 0.001, Table 1; Fig. 4D). With a longer deprivation period of 8 weeks, the mean differences of AL and VCD between FD and contralateral eyes widen to 0.32 ± 0.11 and 0.32 ± 0.11 mm (both P < 0.001, Table 1). The ratio of VCD/AL was higher in the FD eyes at four time points (P < 0.001, Supplementary Table S1). Mean ACD and LT increased over time; however, there were no statistical differences among the FD, contralateral, and control eyes through the intervention. (P > 0.05; Supplementary Tables S1). The AL increase rate of FD and contralateral eyes were 1.562 versus 1.414 mm/month, respectively. Over eight weeks, the mean of CCR in the FD, contralateral, and control eyes dropped from 66.54 ± 0.95, 66.39 ± 0.96, and 66.65 ± 1.29 D to 52.40 ± 1.03, 52.47 ± 1.03, and 52.58 ± 1.43 D, respectively (P < 0.05 between week 0 and week 8). However, there were no statistical differences among the three groups (P > 0.05, Table 1; Fig. 4B). 

Retinal Thickness and Choroidal Thickness

The mean of RT in FD eyes of three-week-old pigmented rabbits was 179.61 ± 8.59, 179.67 ± 7.32, and 179.22 ± 8.82 µm at the temporal, central, and nasal locations, respectively (measuring in Fig. 2), with no differences from the contralateral and control eyes (P > 0.05; Table 2). After eight weeks, RT at all three locations decreased slightly, but there were no significant differences between FD and contralateral eyes (P > 0.05, Table 2). There were no significant differences in ChT among the FD, contralateral, and control eyes at three locations of OCT scans at baseline (week 0) (P > 0.05; Table 2; Fig. 5, measuring in Fig. 2). 

Table 2.

 

Measurements of Choroidal Thickness and Retinal Thickness in Temporal, Central, and Nasal Positions

 

Figure 5.

 

Changes in the choroidal thickness in the temporal, central, and nasal position below one pupil distance from the optic nerve head from week 0 to 8. (A) The choroidal thickness in the temporal side of FD eyes was significantly thinner than those of contralateral and control eyes from week 2 to 8. (B) The choroidal thickness in the center of FD eyes was significantly thinner than those of contralateral and control eyes from week 2 to 8. (C) The choroidal thickness in the nasal side of FD eyes was significantly thinner than those of contralateral and control eyes from week 2 to 8. (D) Changes in the average choroidal thickness of the temporal, central, and nasal positions from week 2 to 8. T: temporal; C: center; N: nasal. *P < 0.05, **P < 0.01, ***P < 0.001 between the FD and contralateral eyes; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001, between the FD and control eyes.

The FD eyes showed a decrease in the averaged ChT from 58.01 ± 6.32 µm to 57.66 ± 6.72 µm after two weeks and then increased slowly to 62.80 ± 8.80 µm after eight weeks. On the other hand, the contralateral eyes showed a significant increase in the averaged ChT from 57.65 ± 6.03 µm to 67.51 ± 8.37 µm after two weeks, to 75.14 ± 10.42 µm after four weeks, to 76.69 ± 10.32 µm after six weeks and to 82.17 ± 12.24 µm after eight weeks (all P < 0.01; Table 2; Fig. 5D). The mean differences of ChT continue to increase from 0.36 ± 2.60 µm to −19.37 ± 7.01 µm after eight weeks (P < 0.001, Table 2). Similarly, the control eyes showed a continuous increase in ChT in week 2, week 4, week 6, and week 8 (all P < 0.05; Table 2; Fig. 5). In three groups, all averaged ChT showed a strongly positive correlation with refraction (R² = 0.43, P = 0.002) and a negative correlation with AL (R² = −0.29, P = 0.04). 

Choroidal Vascularity Parameters

TCA, LA, SA, and CVI were measured at baseline (Fig. 3), and all results were not statistically different (all P > 0.05, Fig. 6). During the experimental period, the TCA continuously increased in the FD, contralateral, and control eyes. However, a significant reduction was found in FD eyes, compared with contralateral eyes (week 2: 0.3160 ± 0.0618 vs. 0.3517 ± 0.0596 mm²; week 4: 0.3278 ± 0.0655 vs. 0.3814 ± 0.0731 mm²; week 6: 0.3480 ± 0.0650 vs. 0.3864 ± 0.0723 mm²; week 8: 0.3697 ± 0.0639 vs. 0.4272 ± 0.0968 mm²; all P < 0.001, Fig. 6A). The SA also showed a continuous increase in three groups. Significant reduced SA was found in the FD eyes after 4 weeks, compared with contralateral eyes (week 4: 0.0970 ± 0.0212 vs. 0.1088 ± 0.0247 mm²; week 6: 0.1022 ± 0.0202 vs. 0.1115 ± 0.0235 mm²; week 8: 0.1047 ± 0.0221 vs. 0.1233 ± 0.0328 mm²; all P < 0.05, Fig. 6B). The LA consistently increased in both the contralateral and control eyes from the beginning of the experiment, while the FD eyes remain stable for the first two weeks, increasing slowly from week 4 onward. LA in the FD eye was smaller than in the contralateral eyes at all time points, and the difference was statistically significant after 4 weeks (Fig. 6C). During the experimental period, the CVI calculated from TCA, SA, and LA showed no significant difference among the FD, contralateral, and control eyes (week 0: 73.74% ± 4.37% vs. 72.92% ± 4.42% vs. 71.30% ± 3.72%; week 2: 71.84% ± 3.51% vs. 71.72% ± 3.43% vs. 71.26% ± 2.83%; week 4: 70.34% ± 3.77% vs. 71.50% ± 3.23% vs. 70.13% ± 3.06%; week 6: 70.60% ± 2.64% vs. 71.07% ± 3.81% vs. 69.68% ± 2.83%; week 8: 71.72% ± 3.20% vs. 71.29% ± 2.55% vs. 70.05% ± 2.80%; all P > 0.05, Fig. 6D). 

Figure 6.

 

The comparison of changes in choroidal vascularity parameters from week 0 to 8. (A) The total choroidal area in the FD and contralateral eyes was enlarged, and the area in the FD eyes was smaller than that in the contralateral and control eyes. (B) The stromal area in the FD eyes was smaller than in the contralateral and control eyes. (C) The luminal area in the FD eyes was smaller than in the contralateral and control eyes. (D) There was no significance in the choroidal vascularity index between the FD and contralateral eyes. *P < 0.05, **P < 0.01, ***P < 0.001 between the FD and contralateral eyes; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001, between the FD and control eyes.

Dopamine Levels

At the end of the experiment, dopamine concentrations of vitreous humor from the FD (n = 8), contralateral (n = 8), and control eyes (n = 4) were quantified. The vitreous dopamine level showed a significant decline in FD eyes, compared with contralateral and control eyes: 0.18 ± 0.20, 0.40 ± 0.67, and 0.33 ± 0.06 ng/mL, respectively (P < 0.05, Fig. 1B). 

Discussion

In the present study, we succeeded in validating a relatively myopic shift model induced by form deprivation in pigmented rabbits by measuring ocular biometric parameters and analyzing choroidal structure changes to assess myopization. To the best of our knowledge, we reported here the first quantitative analysis of the choroidal structural and vascular changes in rabbits. Overall, the results of monocular form-deprivation relative myopia seen in nonhuman primates were reproduced in our pigmented rabbits. 

All pigmented rabbits in the FD group showed great myopic shift susceptibility induced by form deprivation during the experimental period. Early monocular form deprivation in infant pigmented rabbits resulted in the FD eye developing a longer AL and manifesting more myopic characteristics than the contralateral and control eyes. The relative myopic change in SE induced by the intervention was about 1.95 D in pigmented rabbits at week 6, which was higher than the average change of 1.35 D in the FD group of NZW rabbits, as the previous study reported.¹⁰ A remarkable rise in AL from 12.61 mm to 15.32 mm was observed between week 0 and week 6, similar to that in albino rabbits. Form deprivation induced rapid myopia progression in pigmented rabbits, with a refractive change rate of −1.436 versus − 0.551 D/month and an AL change rate of 1.562 versus 1.414 mm/month for the FD and contralateral eyes, respectively. Nevertheless, interocular differences in these parameters in pigmented rabbits were more pronounced. Additionally, the same operator performed the measurements of A-scan ultrasonography to ensure that the probe was positioned at the center of the cornea without exerting pressure and to reduce the measurement error, like in guinea pigs²⁰ and chicks.²¹ From weeks 6 to 8, the changes in SE, VCD, and AL remained relatively stable, suggesting that form deprivation induction in rabbits played a significant role in the initial six weeks. Furthermore, the corneas became flattened but showed no difference between the two subspecies. However, Wang et al.²² reported that the mean CCR of adult pigmented rabbits was lower than that of NZW rabbits and even more similar to that of humans. 

The ratio of VCD to AL demonstrated that the primary determinant of enlargement of ocular AL during myopia development was an elongated vitreous chamber cavity observed in human ocular myopization.²³ In addition, physical and functional differences caused by RPE pigmentation patterning also reflect the heterogeneity of the two species on the same disease.²⁴ Solans et al.²⁵ reported that ocular therapeutic drug concentrations are higher in pigmented rabbits than in albino ones due to the combination with melanin. It has also been reported that atropine, the recent pharmacological agent to control myopia, binds melanin in the RPE of rabbits to become functional.²⁶ The findings suggested that pigmented rabbits were prone to be a decent animal model for studying human myopia genesis and treatment because of physiological and metabolic manifestations. 

The current research reported that a reduction in ocular dopamine levels or the inhibition of the dopamine signal pathway could contribute to the development of myopia.²⁷ Reduced dopamine levels have been observed in the eyes of experimental myopia animal models of various species, including multiple species such as monkeys²⁸ and guinea pigs.²⁹ Retinal dopamine may directly regulate the expression of BMP2 protein in RPE to accelerate or inhibit myopia progression.³⁰ In our study, the reduced dopamine levels in the form-deprivation pigmented rabbit model were consistent with the previous studies, indicating the role of the dopaminergic signaling pathway during postnatal refractive development in the species. 

Myopia remains a complex and elusive condition characterized by excessive ocular elongation, accompanied by scleral remodeling and thinning of the choroid during development. Although the relationship between myopia and the choroid remains incompletely understood, the present study sheds light on this issue. In our study, ChT of pigmented rabbits increased with aging in the control and contralateral eyes in consonance with the trend in humans during childhood.³¹ By contrast, ChT in the FD eyes decreased rapidly from week 2 after the imposition of myopia-inducing form deprivation. However, the speed of choroidal thinning and component atrophy was relatively slower from week 4 to 6, which implied the upregulation of some signal protective mechanisms in inhibiting myopic processes in ocular growth.³² Visual signals are transmitted into the eye to stimulate the retina, with the collaboration of the RPE and choroid, ultimately regulating changes in the biological behavior of cells in the scleral tissue, the classic signal cascade transmission pathway.³³ With increasing myopia severity, the volume of choroidal blood perfusion accompanying changes in ChT further alters the molecular signaling pathways.³⁴ The hypoxia-inducible factor-1 alpha signaling pathway underlying response to hypoxia and metabolism facilitates myopia progression.³⁵ During the early period of myopia development, the choroidal blood perfusion appears to decrease, resulting in the reduction of ChT and the onset of scleral hypoxia. This is consistent with findings in human pathological myopia, where dramatic decreases in ChT and choroidal blood flow have been observed.³⁶ Previous studies in humans with high myopia found notable reductions in the choriocapillaris and medium-sized vessels, as well as thinning of the layer of large choroidal vessels.^37,38 Our study found that the choroidal atrophied by fit in all choroidal conponents and started enveloping the aggravation of myopia in pigmented rabbits. It has been reported that ChT and choroidal vascularity parameters also show significant reductions in human eyes with high myopia.^36,39 Thus our study demonstrated that the choroidal perfusion changes in pigmented rabbits are consistent with those in human myopia. 

EDI-OCT is a powerful imaging modality that enhances better visualization of the morphological vascular structures of the choroid.⁴⁰ Furthermore, binarization analysis of choroidal structure is an innovative noninvasive method capable of providing more detailed information on retinal and choroid microvascular anatomy and excellent quantitative capability. The standard clinical OCT technology is sometimes superior for imaging the choroid of rabbits owing to the imaging resolution and retinal anatomy varies across species.⁴¹ We modified EDI-OCT scanning parameters and binarization analysis method to analyze ChT, TCA, LA, SA, and CVI to evaluate the effects of myopic induction in pigmented rabbits. We also monitored longitudinal data from individual pigmented rabbits and found that ChT, TCA, LA, and SA were significantly decreased in the FD eyes with myopia development compared with the contralateral and control eyes, consistent with previous myopia research in humans.⁷ During ocular development, natural growth might partially offset the loss of choroidal shrinking, resulting in a slight increase in the TCA, SA, and LA in FD eyes. Maintaining choroidal blood perfusion is essential for providing sufficient oxygen to the outer retina, which is exceptionally susceptible to hypoxia because of its high metabolic demand.⁴² The reduced choroidal blood flow affects the oxygen and nutrient supply to the overlying sclera, leading to a hypoxic environment in the sclera and eventually promoting the myopia process. It is plausible to hypothesize that the choroidal stroma accesses inappropriate nutrition with progressive hypoxia, resulting in choroidal stromal atrophy. Another possibility may be that the bridging tissues between RPE and choroid are subjected to expansive centrifugal forces, leading to choroidal mechanical stretching and, in turn, thinning of both vascular and stromal components.⁴³ 

During the intervention, the CVI in the FD and contralateral eyes remained stable. No statistical significance was found for CVI, suggesting that CVI might not be applicable in predicting myopia development in pigmented rabbits. A plausible explanation is that the choroidal stroma and blood flow increase during normal development, which could somewhat counteract the effect of form deprivation on the choroid. Additionally, the sample size of pigmented rabbits in this study might not have been sufficient to detect differences in CVI. However, the CVI in pigmented rabbits is higher than that in humans, 65.6% and 61.5% for adults⁴⁴ and children,⁴⁵ respectively, suggesting it is still a promising indicator in myopia research. Our findings indicated that the pigmented rabbit myopia model closely resembled human myopia. 

In summary, we assessed the refractive development of the pigmented rabbit model with a normal dopaminergic system alternating rapid myopization and found that changes in ChT and choroidal structure were similar to those seen in myopia in humans. Choroidal circulation is crucial for supplying retinal and scleral oxygen, and a decline in choroidal blood perfusion can cause a relatively hypoxic environment, promoting axial elongation and myopia progression. We believe that the pigmented rabbit will serve as a decent mid-sized animal model to explore the pathophysiology and treatment of myopia, and the modified choroidal structural imaging method can promote the development of animal choroidal imaging research. 

Acknowledgments

Supported by the Medical Scientific Research Foundation of Guangdong Province, China (A2017243), Science and Technology Program of Guangzhou, China(20210210145) and National Natural Science Foundation of Guangdong, China (2020A1515010829). 

Disclosure: Z. Jiang, None; W. Yan, None; H. Fang, None; C. Liu, None; Z. Zhang, None; Z. Li, None; Y. Wei, None; Y. Ni, None