The Effects of High Lighting on the Development of Form-Deprivation Myopia in Guinea Pigs

Abstract

Purpose: To investigate the effects of high ambient lighting on refraction and ocular biometry in guinea pig models of form-deprivation myopia (FDM).

Methods: Forty 3-week-old guinea pigs were randomly assigned to groups exposed to either high light (HL, 10,000 lux) or normal light (NL, 500 lux) with normal vision or form deprivation. Throughout the 10-week rearing period, animals were exposed to high light or normal light for 12 hours with a 12-hour light/dark cycle. Refraction, axial length (AL), and radius of corneal curvature (CCR) were measured by cycloplegic retinoscopy, A-scan ultrasonography, and keratometer, respectively.

Results: At the end of treatment, form-deprived eyes under high ambient lighting exhibited more hyperopic refraction and shorter AL than those under normal ambient lighting (2.06 ± 1.68 diopters [D; mean ± SD] vs. −0.59 ± 1.56 D, P < 0.001; 8.36 ± 0.13 mm vs. 8.56 ± 0.16 mm, P < 0.001). Deprived eyes under high ambient lighting were relatively more myopic than their contralateral control eyes at the end of treatment (2.06 ± 1.68 D vs. 5.44 ± 0.66 D, P < 0.001). High lighting induced a significant hyperopic shift in normal eyes after 4 weeks of exposure. There were no significant differences in CCR between eyes exposed to high and normal light, nor between deprived eyes and contralateral eyes.

Conclusions: High ambient lighting could retard, but not fully inhibit, the development of FDM. High light levels contributed to a greater hyperopic shift in normal eyes during the first 4 weeks of treatment. Corneal curvature was unaffected by either high ambient lighting or form deprivation.

As the prevalence of myopia has increased in recent decades^1–3 and a high prevalence has been observed in East Asia,^4–10 myopia has become an important public health problem. Recent epidemiologic studies^11–19 have reported that more time outdoors provided a significant protective effect on the development of myopia in children. On the basis of the assumption that high light intensity may be associated with this protective effect, many researchers have investigated the role of high ambient lighting in experimental myopia. Notably, high light levels have been shown to retard the development of form-deprivation myopia (FDM) in chicks,²⁰ rhesus monkeys,²¹ mice,²² and tree shrews (Siegwart JT Jr, et al. IOVS 2012;53:ARVO E-Abstract 3457). A protective effect of high lighting against lens-induced myopia was also found in guinea pigs.²³ However, there are no studies regarding the effects of high ambient illumination on FDM in guinea pigs. 

In a study by Smith et al.,²¹ form-deprived eyes in rhesus monkeys were more hyperopic than their fellow eyes when exposed to high ambient light. Conversely, the study²² in mice showed that when exposed to high light, form deprivation for 4 weeks still induced greater myopic refraction in deprived eyes than their contralateral eyes. Thus, more studies in mammals are needed to compare refraction between form-deprived and contralateral control eyes that are exposed to high light. 

In this study, we used a guinea pig model to investigate the influence of high light levels on FDM and differences in refraction between form-deprived and contralateral eyes, in order to explore the protective or inhibitory effects of high lighting on the form-deprivation myopia and explore whether form-deprivation (FD) could induce myopia under high illumination of 10,000 lux. 

Methods

Animals

The experiments in this study adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The research in this study was approved by the Animal Care and Ethics Committee at the EYE & ENT Hospital of Fudan University. 

We raised 3-week-old, healthy, pigmented guinea pigs (Cavia porcellus, English short-hair stock, n = 40; obtained from the Animal Experiments Laboratory of Fudan University, Shanghai, China)²⁴ in four independent and specially designed cages (60 × 45 × 40 cm, mesh size: 1.5 × 5.0 cm²) with a 12-hour light/dark cycle in a dark room with a viewing distance of 1.5 m. The indoor temperature was kept at 22 to 26°C, and relative humidity was maintained at 55% to 65%. The animals were provided food and water ad libitum. 

Experimental Design

Forty 3-week-old guinea pigs were randomly assigned to two different groups: HL (high light) and NL (normal light). Each group was further divided into two subgroups based on different treatments (form-deprivation or normal vision). Thus, there were a total of four subgroups: HL+FD group: animals were treated with monocular FD on right eyes under high light conditions (n = 10); HL+NV group: animals with normal vision (NV) were reared under high light conditions (n = 10); NL+FD group: animals were treated with monocular FD under normal light conditions (n = 10); NL+NV group: animals with normal vision were reared under normal light conditions (n = 10). 

Lighting Conditions and Form-Deprivation Treatment

The HL and NL illuminance was approximately 10,000 and 500 lux, respectively, provided by white dimmable light emitting diode (LED) lighting tubes (40W, color temperature 6500K, 365–795 nm, peaking at 450 and 660 nm; RuiGaoXiang Light & Electronic Co. Shanghai, China). The cages were placed in a stainless-steel cabinet, and white LED lighting tubes were installed on the inner walls and inner ceiling of the cabinet. For high illuminance groups, three light tubes were situated 15 cm above each cage and two tubes were situated behind the cage. For normal illuminance groups, one light tube was situated 15 cm above each cage. The animals were exposed to high light or normal light for 12 hours daily. 

Form deprivation was induced by placing translucent eye-masks onto the right eyes of guinea pigs for 10 weeks. Translucent plastic diffusers were fixed onto the eyes using double-sided tape, which was chosen to be thick enough such that a gap was left between the eye and the diffusers for protection of the cornea. The tape was attached far away enough from the eyelid to avoid affecting the eyelid activities. These translucent eye-masks were examined two times daily to ensure that they remained in place. 

Refraction and Ocular Biometry Measurement

All measurements were performed at baseline (i.e., before the start of treatments [week 0]), and at the fourth, eighth, and tenth week, by a research optometrist and an assistant who were blinded to the identity of the four groups. 

Retinoscopy

Cycloplegia was induced by the instillation of one drop of 0.5% tropicamide phenylephrine (Santen) in each eye, 5 minutes apart, four times. After 20 minutes, retinoscopy was performed in a dark room with a streak retinoscope (66 Vision Tech, Jiangsu Province, China) and trial lenses. The refraction results were calculated as the mean values of the horizontal and vertical meridians, recorded as spherical equivalent (SE). We understand that the absolute value of refraction obtained from our experiments would not be as accurate as in the human eye without correction made for the small eye artifact.²⁵ But, in this study, we are focusing on the differences of refraction and relative refraction changes across the different conditions, so the retinoscopy should be a reliable technique for the purpose of our study. 

Ocular Biometry Measurements

Measurements of ocular biometric parameters (axial length [AL] and keratometry) were performed with an A-scan ultrasonography device (KN1800; Kangning, Jiangsu Province, China) and a keratometer (OM-4; Topcon, Tokyo, Japan). Topical anesthesia was administered with one drop of 0.4% oxybuprocaine hydrochloride (Benoxil; Santen, Osaka, Japan) in each eye before ultrasonic measurements. The ultrasound frequency was set at 10 MHz. Ten readings of AL were recorded for each measurement to calculate a mean result. Because the steep corneal curvature of guinea pig eyes exceeded the measurement range of the keratometer, we attached a +8.0 diopter (D) lens onto the anterior surface of the keratometer. The real radius of corneal curvature (CCR) is equal to the reading of radius ×0.451.²⁶ Three readings of CCR were averaged to provide values used for further analysis. 

Statistical Analysis

All data are shown as the mean ± SD. A 1-way ANOVA was used to compare the mean values of SE, AL, and CCR of the same-side eyes among the four subgroups at baseline (week 0). Comparisons of mean values of the parameters of right eyes with left eyes in the four subgroups were checked with paired t-tests. Two-way repeated measures ANOVA was applied to compare the mean SE, AL, and CCR among the high and normal light subgroups across time. Least Significant Difference (LSD) correction was used for post hoc analysis to found differences of the refraction and ocular biometric parameters between HL and NL subgroups at each measurement time-point, and to find the differences among the four time points. There were no differences in SE, AL, or CCR between right and left eyes in HL+NV and NL+NV subgroups among the four time points (paired t-tests, P < 0.05). Therefore, data from the right eyes in HL+NV and NL+NV groups were shown in the current study. All statistical analyses were performed with SPSS 22.0 (SPSS, Inc.; IBM Corp., Chicago, IL, USA). Values of P < 0.05 were considered statistically significant, and values of P < 0.01 were considered highly significant. 

Results

Refraction

Before the start of treatment (week 0), the mean refraction of guinea pigs in all four subgroups was moderate hyperopia, and there were no differences in the mean SE of the same-side eyes among the HL+FD, NL+FD, HL+NV, and NL+NV subgroups (P > 0.05, 1-way ANOVA). 

High light significantly retarded the myopic development induced by form deprivation when compared with normal light at the end of treatment period. (Fig. 1, Table 1, Supplementary Table S1). There was a significant interaction between the treatments (different light conditions or form deprivation) and measurement time (F[11.1, 88.5] = 9.553, P < 0.001, 2-way ANOVA; Table 1) when the SE of all subgroups were compared. The interaction indicated that the effects of treatments on refraction were different by treatment time. The LSD post hoc analysis found that after 8 weeks of illumination, high light shifted refraction toward more hyperopic refractive error in deprived eyes, compared with the normal light group (2.69 ± 1.79 D vs. 0.34 ± 1.03 D, P < 0.001 for LSD post hoc analysis at week 8; 2.06 ± 1.68 D vs. −0.59 ± 1.56 D, P < 0.001 at week 10; Fig. 1A, Table 1). At the end of treatment, form-deprived eyes in monocular FD animals under high lighting exhibited smaller myopic shift than the deprived eyes under normal lighting (−1.89 ± 1.56 D vs. −4.22 ± 1.48 D, P = 0.003, LSD post hoc analysis; Supplementary Table S1). 

Figure 1

 

The spherical equivalent with time in the form-deprived and contralateral control eyes as well as the normal vision eyes under high and normal light in the four subgroups. HL+FD-Deprived, HL+FD-D: deprived eyes in the HL+FD group; HL+FD-Control, HL+FD-C: contralateral control eyes in the HL+FD group; HL+NV-Normal, HL+NV-N: normal vision eyes in the HL+NV group. NL+FD-Deprived, NL+FD-D: deprived eyes in the NL+FD group; NL+FD-Control, NL+FD-C: contralateral control eyes in the NL+FD group; HL+NV-Normal, NL+NV-N: normal vision eyes in the NL+NV group. The data were expressed as the mean ± SD for each time point. *P < 0.05, **P < 0.01, ***P < 0.001 (2-way ANOVA with LSD post hoc tests). (A) In comparison with deprived eyes under normal light, high light shifted refraction toward more hyperopic refraction at weeks 8 and 10. (B) High light shifted refraction toward higher hyperopia in contralateral control eyes compared with normal light group at weeks 4, 8, and 10. (C) High light shifted refraction toward higher hyperopia in normal vision eyes compared with normal eyes under normal light at weeks 8 and 10. (D) The SE values with time in the four subgroups.

Table 1

 

Comparisons of SE (in Diopters) Between HL and NL During the Treatment Period (Mean ± SD)

 

The form-deprived eyes under high lighting were relatively more myopic than their contralateral eyes (paired t = −4.875, P = 0.001; t = −4.912, P = 0.001; t = −6.218, P < 0.001, data not shown) at weeks 4, 8, and 10 (Fig. 1D). FD for 10 weeks induced significant myopic shift in deprived eyes under HL condition compared with contralateral eyes (paired t = −6.572, P < 0.001). 

For the nondeprived eyes, high light significantly shift refraction toward higher hyperopia when compared with NL subgroups (Figs. 1B, 1C; Table 1). After 8 weeks' illumination, the contralateral eyes in monocular FD animals and the normal eyes in normal vision animals under the high lighting manifested greater hyperopic refraction at week 8 (P < 0.001 for contralateral eyes, P = 0.001 for normal vision eyes, LSD post hoc analysis of 2-way ANOVA) and at week 10 (P < 0.001 for contralateral eyes, P = 0.008 for normal vision eyes, LSD post hoc analysis), when compared with the counterparts under the normal lighting (Figs. 1B, 1C; Table 1). 

Under the HL condition or NL condition, the refraction of the deprived eyes, the contralateral control eyes and normal eyes all varied significantly with treatment time (2-way ANOVA, F[2.2, 88.5] = 34.562, P < 0.001; Table 1). 

During the illumination period, high ambient lighting induced significant hyperopic shifts in the normal vision eyes after 4 weeks of illumination compared to eyes in NL group as the eyes at week 4 was more hyperopic than week 0 (P = 0.005, LSD post hoc analysis of 2-way ANOVA, data not shown). From week 4 to 10, the hyperopic refraction of normal eyes decreased with time. At week 10, the SE of eyes in the HL groups exhibited no difference compared with SE at baseline (P > 0.05, LSD post hoc analysis, Fig. 1D, Table 1). Differently, in comparison with baseline, the normal vision eyes of animals reared under normal light condition were more myopic after 10 weeks of illumination (P = 0.019, LSD post hoc analysis of 2-way ANOVA, data not shown). 

Axial Length

Before the start of the treatment (week 0), no differences in the mean AL of right and left eyes were found among the HL+FD, NL+FD, HL+NV, and NL+NV groups. (P = 0.106 for right eyes, P = 0.181 for left eyes, 1-way ANOVA). 

HL retarded the axial elongation induced by form-deprivation when compared with NL group (Fig. 2A, Table 2, Supplementary Table S1). Two-way repeated measures ANOVA found there was a significant interaction between the treatments (different light conditions or form deprivation) and measurement time (F[9.9, 83.2]= 8.247, P < 0.001, Table 2) when the mean AL of deprived eyes, contralateral control eyes, and normal vision eyes in the HL and NL subgroups were compared. At week 4, the mean axial length of deprived eyes under high light condition did not differ from deprived eyes under normal light condition; differently, at weeks 8 and 10, deprived eyes exposed to high light exhibited the shorter AL compared to the deprived eyes under normal lighting (LSD post hoc analysis: 8.32 ± 0.13 mm vs. 8.49 ± 0.16 mm, P = 0.002 at week 8; 8.36 ± 0.13 mm vs. 8.56 ± 0.16 mm, P < 0.001 at week 10; Fig. 2A, Table 2). 

Figure 2

 

The axial length with time in the form deprived and contralateral control eyes as well as the normal eyes with normal vision under high and normal light in the four subgroups. The data were expressed as the mean ± SD for each time point. *P < 0.05, **P < 0.01, ***P < 0.001 (2-way ANOVA with LSD post hoc tests). (A) In comparison with deprived eyes under normal light, high light significantly retarded the axial elongation of deprived eyes at weeks 8 and 10. (B) The axial length of contralateral control eyes under high light was shorter than contralateral eyes under normal light at week 10. (C) High light significantly slowed down the axial elongation in normal vision eyes compared with normal eyes under normal light at week 10. (D) The axial length with time in the four subgroups.

Table 2

 

Comparisons of AL (in mm) Between HL and NL During the Treatment Period (Mean ± SD)

 

The mean AL in deprived eyes exposed to high light elongated more rapidly than AL in their contralateral control eyes during the treatment period (Supplementary Table S2), and therefore was longer than that of contralateral control eyes at weeks 4, 8, and 10 (paired t = 2.537, P = 0.032; t = 3.767, P = 0.005; t = 4.014, P = 0.004; Fig. 2D, Table 2). 

For nondeprived eyes, the normal eyes under high light condition developed shorter mean AL than the counterparts under normal lighting at the end of treatment period (8.14 ± 0.05 mm vs. 8.31 ± 0.05 mm, P = 0.005, Fig. 2C, Table 2). 

Radius of Corneal Curvature

At the start of treatment (week 0), no significant differences in CCR were found in right and left eyes among the HL+FD, NL+FD, HL+NV, and NL+NV groups (P = 0.518, P = 0.851, 1-way ANOVA). 

Two-way ANOVA with repeated measures revealed a significant interaction between the treatments and measurement time (F[12.0, 105.2] = 3.226, P < 0.001, Table 3). There was no significant difference of the average CCR in deprived eyes, contralateral eyes, and the normal vision eyes among the HL and NL subgroups across time (F[5, 44] = 2.401, P = 0.052, Fig. 3A, 3B, 3C; Table 3). What's more, differences of average CCR between deprived and contralateral eyes in HL and NL groups were also not significant at the start and end of treatment (P > 0.05 for paired t-tests, data not shown). 

Table 3

 

Comparisons of Radius of Corneal Curvature (mm) Between HL and NL During the Treatment Period (Mean ± SD)

 

Figure 3

 

The radius of cornea curvature with time in the form-deprived and contralateral control eyes as well as the normal eyes with normal vision under high and normal light in the four subgroups. The data were expressed as the mean ± SD for each time point. (A–D) No significant differences of the average CCR of deprived eyes and contralateral eyes between the HL and NL groups, as well as in the normal vision eyes between the HL and NL groups. The cornea flattened with time in all deprived and contralateral control eyes as well as normal vision eyes of all subgroups.

The cornea flattened with time in all deprived and contralateral eyes as well as normal vision eyes of all subgroups (Fig. 3, Table 3), and the mean CCR were significantly different among the four time points in all subgroups (F[2.4, 105.2] = 1042.39, P < 0.001, 2-way repeated measures ANOVA, Table 3). 

Discussion

The findings in our study demonstrated that high lighting (10,000 lux) reduced the myopic shift induced by form deprivation in guinea pigs. More specifically, exposure to high illumination of 10,000 lux for 10 weeks reduced the degree of form-deprivation myopia by approximately 60%. Similar findings have been reported in other studies^20–22 focused on chicks, mice, tree shrews, and rhesus monkeys. A study conducted by Ashby et al.²⁰ showed that high lighting (15,000 lux) retarded FDM by approximately 60% in chicks over a 4-day treatment. One study²¹ on rhesus monkeys also reported a protective effect of high lighting (18,000–28,000 lux) on the development of FD-induced axial myopia. 

In the current study, high ambient lighting (10,000 lux) did not completely inhibit the development of form-deprivation myopia (approximately 60%), as FD for 10 weeks was still able to induce a significant myopic shift from baseline (−1.89 ± 1.56 D) in guinea pigs, when compared with the contralateral control eyes. That high light was not completely protective of FDM in our study was similar to the results in other studies.^20,22 A study²² on mice (2500–5000 lux) and a study on chicks (15,000 lux) demonstrated that FD still induced myopic tendencies under high light compared with their contralateral eyes. In addition to FDM, incomplete inhibitory effects of high ambient lighting on lens-induced myopia were also found in guinea pigs²³ and chicks.²⁷ These results indicated that with the factors that could induce myopia existing, animals reared under high ambient lighting (2,500–15,000 lux) may continue to exhibit myopic shift. However, the effects of substantially higher light intensity on development of experimental myopia should be investigated to explore whether a higher light intensity could inhibit the development of myopia. 

In contrast to the findings of our study, the study²¹ on rhesus monkeys showed different results; deprived eyes of six monkeys reared under high light were more hyperopic than their contralateral eyes. Smith et al.²¹ suggested that under the high light condition, the natural hyperopic errors of the fellow eyes in monkeys may provide a sufficient stimulus for emmetropization and the diffuser lens may virtually eliminate meaningful signals produced by the hyperopic errors. The differing results between the two studies may be due to different experimental methods, animal models, and different high light levels. It may also be that form deprivation, which produced an abnormal visual stimulus, provides a more significant influence on the refractive development in deprived eyes in our study, even under the high lighting condition. 

In this study, high lighting significantly influenced the course of emmetropization in normal eyes. High lighting induced significant hyperopic shifts in normal eyes for the first 4 weeks of illumination and shifted the refraction toward greater hyperopia compared with eyes exposed to normal light. These findings were similar to those of other studies.^{22,23,28–30} The study in mice²² also showed hyperopic shifts in non-deprived eyes under bright light after 4 weeks. Cohen et al.²⁹ reported that chicks gradually developed severe hyperopia when exposed to continuous light at high illuminance. Li et al.²³ supposed that the flattening of the cornea in nondeprived guinea pigs may have a greater effect than axial elongation, which resulted in the hyperopic shifts. Differently, in our study, there were no changes in corneal curvature in response to light level, it requires more further investigations to explain the obvious hyperopic shift. 

High light level and form deprivation did not influence CCR in the current study. The study on chicks²⁰ and on monkeys²¹ also reported that corneal powers were unaffected by high light. However, there were other studies.^28,29 reporting that continuous high lighting could induce corneal flattening, and that the degree of flattening was significantly associated with light intensity. These different findings may be due to the different lighting methods. In compared with the continuous high lighting, high lighting with a diurnal light-dark rhythm does not affect the corneal curvature, which supports the data of Ashby et al.²⁰ In addition to the effect of high light, form deprivation had no significant influence on CCR during the 10 weeks of treatment, which was similar to the findings in other studies involving marmosets³¹ and guinea pigs,^26,32 suggesting that FD had no impact on corneal curvature in these animals. 

The mechanism of the protective effect of high lighting on FDM remains unclear. Much evidence has been reported suggesting that the protective effect of high lighting may be mediated through the retinal dopamine (DA) system.^29,33–40 Moreover, a DA receptor antagonist promoted the progressive myopia in guinea pigs³⁹ and reversed the inhibitory effects of bright light on myopia in mice.²² However, the precise molecular mechanism by which the DA system mediates the protective effect of high light is unclear, and requires further studies to identify more related molecular signals. 

In summary, our study supports the hypothesis that the strong association of outdoor time with reduction of myopia in children may be related to high ambient lighting. There was a protective but incomplete inhibitory effect of high light levels (10,000 lux) on form-deprivation myopia in guinea pigs. High light also caused a more hyperopic shift in normal eyes compared with normal light, during the first portion of the treatment period. Corneal curvature was unaffected by high light levels and form deprivation. 

Acknowledgments

Supported by a 3-year Action Program of Shanghai Municipality (2011-2013; Grant Nos. 2011-15, 2015-2017, and 2015-13); Shanghai Science Popularization Project (Grant No. 17dz2301400); and Key Laboratory of the Ministry of Health in Myopia Shanghai Scientific Research Plan Project (Grant No. 14411969500). 

Disclosure: L. Zhang, None; X. Qu, None