Continuous Ambient Lighting and Eye Growth in Primates

Abstract

purpose. To determine the effect of continuous light exposure on ocular growth and emmetropization in infant monkeys.

methods. Nine infant rhesus monkeys were reared with the normal vivarium lights on continuously. The 24-hour light cycle was initiated between 1 and 4 weeks of age and maintained for 6 months. The ocular effects of continuous light were assessed by cycloplegic retinoscopy, keratometry, and A-scan ultrasonography. Longitudinal control data were obtained from 23 normal infants that were reared with an illumination cycle that included defined light and dark phases (either 12-hour light:12-hour dark or 8.5-hour light:15.5 hour dark).

results. In contrast to previous studies involving light-reared chickens, no monkeys exhibited exaggerated ocular growth. There were no significant differences between treated and control monkeys in corneal radius, overall eye size, or the axial dimensions of individual ocular components. At the end of the treatment period, eight of the nine experimental monkeys also exhibited the moderate hyperopic errors (range, +1.5 to +3.4 D) that are typically found in normal animals. Aspects of emmetropization were, however, unusual for three monkeys. One monkey manifested a −0.50 D myopic error that was associated with an abnormally steep cornea but had normal axial lengths. Two additional monkeys developed persistent axial anisometropias.

conclusions. In infant primates constant light exposure does not promote the constellation of ocular changes (in particular corneal flattening, a decrease in anterior chamber depth, and an increase in vitreous chamber depth) that has been observed in light-reared chickens. The slight variations from the expected developmental sequence observed in three infants may reflect individual differences. However, it is also possible that aspects of the emmetropization process may not operate as effectively under constant light as they do under an ordinary light/dark cycle.

For optimal vision there must be a very precise match between the length of the eye and the eye’s optical power. Why some individuals achieve the remarkable degree of precision required for optimal vision whereas others develop refractive errors like myopia is not well understood. Visual experience, however, may play an important role in both normal and abnormal refractive development. 

Evidence from a wide range of animal species, including higher primates, indicates that most eyes achieve the optimal refractive state because the axial growth of the eye is regulated by visual feedback in a manner that eliminates refractive errors. ^{1 2 3 4 5} However, visual experience can also interfere with the normal emmetropization process. It is well recognized that conditions like cataracts that prevent clear vision and interrupt normal visual feedback produce abnormally long, myopic eyes in both animals and humans. ^{6 7 8} However, environmental viewing conditions that do not necessarily preclude the formation of clear retinal images can also influence early ocular growth and result in anomalous refractive errors. For example, evidence obtained from chickens indicates that the duration of the daily light period can have a profound effect on many aspects of early ocular growth. ^{9 10 11 12} In particular, rearing young chicks in continuous light results in an extreme enlargement of the globe that is associated with dramatic increases in corneal radius of curvature and vitreous chamber depth. ^{10 12} Depending on the breed of chicken and the duration of the period of continuous light, the resulting refractive errors can range from high hyperopia to high myopia. ^{10 13}  

Because many aspects of visual experience have been shown to affect ocular growth in chickens and monkeys in a qualitatively similar way, ^{14 15 16} it is possible that continuous light exposure could also alter refractive development in primates. Interestingly, a recent survey of a patient-based population found a strong association between childhood myopia and continuous nighttime lighting before the age of 2 years; specifically, children who slept with room lights on were five times more likely to be myopic than those who slept in the dark. ¹⁷ This relationship, however, was not apparent in more representative samples of the general population. ^{18 19}  

Thus, the possible role of nighttime light exposure in the genesis of abnormal refractive errors in primates remains unclear. To address this issue, we have examined the effects of continuous light exposure on early eye growth and emmetropization in infant macaque monkeys. Macaque monkeys are ideal subjects because ocular and refractive development are very similar in humans and macaques, ²⁰ but with monkey subjects the lighting cycle can be controlled precisely. 

Materials and Methods

Subjects

The effects of continuous exposure to light were investigated in nine infant rhesus monkeys (Macaca mulatta). Three of the infants were born and raised at the Yerkes Regional Primate Research Center of Emory University. The other six infants were reared at the University of Houston. At both sites, the subjects were housed in rooms with the normal fluorescent ceiling lights on continuously for 24 hours each day. The three Emory monkeys were housed in a single-level, multiple-animal caging system. The effective illumination at the top of the cage was 640 lux. The six Houston monkeys were housed in a two-level, stacked caging system. Four of the Houston monkeys (monkeys PAR, PEG, XAV, and XYL) were reared entirely in the upper cages. Two Houston monkeys (monkeys ABE and ADA) spent the first 3 months of the treatment period in the lower cages before being moved to upper cages for the remainder of the light-rearing period. The illumination at the tops of the upper and lower cages was 630 and 230 lux, respectively. The lowest illumination levels were 210, 150, and 15 lux for the Emory cages and the upper and lower Houston cages, respectively. The 24-hour light cycle was initiated between 1 and 4 weeks of age and maintained until the monkeys were at least 6 months of age. At 6 months of age, four of the six Houston monkeys were returned to a normal 12-hour:12-hour light:dark cycle. The three Emory monkeys and the other two Houston monkeys were maintained on the 24-hour lighting cycle throughout the observation period. Based on the relative rates of human and monkey development, our light-rearing period was equivalent to about the first 2 years of life for a human infant. ^{20 21}  

Longitudinal data for normal ocular growth were obtained from 23 infant monkeys that were raised in identical laboratory settings, which were maintained on a normal diurnal lighting cycle (Houston: n = 16, 12-hour:12-hour light:dark cycle; Emory: n = 7, 8.5-hour:15.5-hour light:dark cycle). Some aspects of refractive development for these animals, primarily refractive error, have been previously reported. ^{5 20 22} Additional cross-sectional data were also available from 232 normally reared monkeys. ²⁰  

All the rearing and experimental procedures were approved by the Institutional Animal Care and Use Committees at the University of Houston and Emory University and were in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Optical and Biometric Measurements

Our assessment of eye growth concentrated on the ocular characteristics that are dramatically altered in young chickens by continuous light exposure, specifically, the eye’s refractive status, the corneal radius of curvature, and the eye’s axial dimensions, especially anterior chamber and vitreous chamber depths. To make these measurements, cycloplegia was induced (Houston, 1% tropicamide; Emory, 1% cyclopentolate), and the animals were anesthetized with ketamine hydrochloride (10–20 mg/kg) and acepromazine maleate (0.1–0.2 mg/kg). The spherical-equivalent, spectacle-plane refractive corrections were determined by retinoscopy. The radius of curvature of the cornea along the eye’s pupillary-axis was determined with a hand-held keratometer (Alcon Auto-keratometer; Fort Worth, TX, Houston and Emory) and a video-topographer (EyeSys 2000; Houston). A-scan ultrasonography was used to measure the eye’s axial dimensions. General ocular health was examined via measures of intraocular pressure (Tonopen), ophthalmoscopy, and fundus photography. The animals were first examined at the onset of the continuous-light-rearing period and typically every 2 to 4 weeks for the remainder of the observation period (for more details, see Refs. ⁵ , ²⁰ , and ²³ ). 

At 6 months of age, the end of the rearing period for some monkeys, there were no systematic differences in refractive error, axial dimensions, or corneal power between the right and left eyes of the treated subjects (paired t-test, P = 0.63–0.86). Consequently, comparisons between control and treated monkeys have been primarily restricted to data from the right eyes. 

Results

In some species, continuous exposure to light levels that are not intense enough to burn the retina can nevertheless produce alterations in ocular physiology that could influence emmetropization. In particular, young chicks exposed continuously to moderate light levels for several months develop substantial elevations in intraocular pressure, ¹³ ophthalmoscopically visible fundus lesions, ¹⁰ and histologically verified retinal degeneration. ^{10 11} However, over the period of continuous light, none of the experimental monkeys exhibited elevated intraocular pressures (IOPs). The IOPs of our light-reared monkeys varied from 4 to 15 mm Hg (mean ± SD, 8.2 ± 3.1 mm Hg) and were always well within the range for normal monkeys. ²⁴ Moreover, at the end of the treatment period no visible retinal lesions were observed in any of our experimental subjects. 

There was no evidence that exposure to constant light produced exaggerated ocular growth in any of the treated monkeys. At the start of the continuous-light-rearing period, the axial lengths (Fig. 1) and corneal radii (Fig. 2) for all the treated monkeys were representative of those of normal infant monkeys. As the treated infants matured, their axial lengths increased systematically, but at every age the longitudinal data for each treated infant fell within the range of values for the normal control monkeys. Likewise, although corneal radius of curvature increased systematically with age, the radius of curvature did not increase more or faster in the treated than in the normal monkeys. 

Figure 3 compares the major ocular dimensions for the light-reared and normal monkeys at 6 months of age (i.e., the end of the treatment period for 4 of the Houston monkeys). Based on the results of two-sample t-tests, there were no statistically significant differences between the treated and control groups in vitreous chamber depth (T = 1.21, P = 0.25), corneal radius (T = 0.83, P = 0.42), corneal thickness (T = −1.34, P = 0.23), anterior chamber depth (T = 0.27, P = 0.80), or lens thickness (T = −0.69, P = 0.50). In comparison, previous studies have demonstrated that in young chickens only a few weeks of exposure to continuous light is sufficient to produce significant increases in axial length, vitreous chamber depth, corneal radius, and corneal thickness, combined with significant decreases in anterior chamber depth and lens thickness. ^{9 10 11 12 25}  

As illustrated in Figures 4 and 5 , which show refractive error plotted as a function of age for the Houston and Emory monkeys, respectively, emmetropization was largely unaffected by exposure to constant light. At the start of the continuous-light-rearing period, all the treated infants exhibited hyperopic refractive errors that were within the range for normal monkeys. Over the next 2 to 3 months, eight of the nine light-reared monkeys either maintained a low, stable level of hyperopia or exhibited a reduction in hyperopia down to the low hyperopic values that are typically observed in normal, 4- to 5-month-old monkeys. Thereafter, refractive errors were relatively stable for the remainder of the observation period. At 6 months of age there were no significant differences between the refractive errors of the treated and control animals (Fig. 3A ; two-sample t-test, T = 0.24, P = 0.82). 

Although we did not find any evidence for the macrophthalmous commonly found in light-reared chickens, several observations in individual monkeys suggest that continuous light exposure may have affected the efficiency of emmetropization. First, one light-reared monkey (ROG7, Fig. 5 ) briefly manifested a low myopic refractive error (−0.5 D at 3 months of age) and thereafter exhibited essentially emmetropic refractive errors that were clearly below the levels of hyperopia found in normal monkeys. This is potentially significant because myopia is very rare in young monkeys. ²⁰ In rhesus monkeys, the emmetropization process is largely complete by 4 or 5 months of age. ^{5 20} At these ages none of the 23 normal monkeys that we followed longitudinally had hyperopic errors below+ 0.87 D of hyperopia. However, it is important to note that this myopic animal did not exhibit an increase in either overall axial length or vitreous chamber depth, the most common ocular changes associated with myopia. Instead, this animal’s refractive error was associated with a relatively steep cornea (filled triangles, Fig. 2 ). For example, at 124 days of age, when this animal showed the largest relative myopic error, its corneal radius of curvature was shorter than that for any of the age-matched, control monkeys. Second, beginning at approximately 100 days of age, monkeys XAV and PAR developed persistent axial anisometropias. XAV’s anisometropia, which was larger than any interocular refractive-error difference observed in normal monkeys (up to 1.62 D, see Figs. 4 and 6 ), was maintained for the remainder of the treatment period but eventually did resolve several months after the animal was returned to a normal lighting cycle. PAR’s anisometropia was relatively small (between 94 and 171 days; average, 0.74 D; range, 0.5–0.94 D), but it also persisted for the rest of the light-rearing period. The significance of these findings is not clear cut. Control animals occasionally exhibit anisometropias of approximately 1 D; however, as shown in Figure 6 these interocular differences in refractive error are typically transient and regress between measurement sessions. It is also important to note that of the nine light-reared monkeys, XAV had the highest degree of hyperopia at the start of the treatment period. In this respect, we have previously observed that infant monkeys frequently develop significant anisometropias in response to high degrees of optically imposed hyperopia or when they are recovering from high degrees of experimentally induced hyperopia. ^{5 26} Consequently, it is possible that XAV’s anisometropia was due to its high initial hyperopic refractive error rather than exposure to continuous light. 

Discussion

Our results show that continuous light exposure does not alter overall eye size or the dimensions of individual ocular components in infant monkeys. The failure of constant light exposure to produce macrophthalmous in monkeys is somewhat surprising because chickens exhibit dramatic ocular changes when exposed to constant light, ^{9 10 11 12 25} and the eyes of chickens and monkeys have been shown to respond in similar ways to a variety of environmental manipulations. ^{1 2 3 5 6 22 23 27 28 29 30} There are a number of possible explanations for the qualitatively different responses to constant light found in chickens and monkeys. Infant monkeys, through behavioral strategies, may be able to avoid constant light exposure more effectively than young chicks. For example, although both chicks and monkeys sleep with their eyelids closed, infant monkeys could exclude more light from reaching their retinas by covering their eyes with their arms when they sleep. Although we cannot rule out the possibility that infant monkeys somehow reestablished an effective light/dark lighting cycle by shielding their eyes for a significant part of each day, we did not observe any obvious behavioral differences between continuous-light-reared infants and normal control monkeys. Regardless any such behavioral differences between chicks and infant monkeys would not restrict the extrapolation of our results to human infants. 

In the chick, the increase in vitreous chamber depth produced by constant light appears to be mediated by different mechanisms than those that underlie the anterior segment changes. ^{31 32} Although the exact mechanisms that are responsible for either of these changes are not well understood, the processes that have been implicated in the posterior segment changes in chickens also provide a plausible explanation for the absence of vitreous chamber elongation in our light-reared monkeys. For example, because many vision-dependent changes in eye growth appear to be mediated by local retinal mechanisms, ¹⁵ the obvious and consistent pathologic retinal alterations produced by constant light in chicks ^{10 11} could play a role in the exaggerated posterior segment growth. It has also been suggested that the light-induced posterior segment changes could come about as a result of the light-induced anterior segment changes found in chicks. For instance some evidence suggests that the elevation in IOP that is associated with the anterior segment changes in chicks plays a central role in vitreous chamber enlargement, ³³ although it has been shown that eye enlargement can occur before significant IOP changes can be measured. ¹³ It has also been hypothesized that the increase in vitreous chamber depth found in light-reared chicks represents emmetropizing growth stimulated by the light-induced decrease in corneal power. ¹⁰ If any or all these ideas are correct, it is not surprising that we did not observe significant increases in the vitreous chamber depth of our light-reared monkeys. None of our monkeys demonstrated ophthalmoscopically detectable retinal anomalies, elevated IOPs, or the exaggerated corneal flattening commonly found in light-reared chicks. 

The failure of constant light to produce anterior segment changes in infant monkeys comparable to those in light-reared chickens may be related to interspecies differences in the way in which light can influence pineal melatonin synthesis. In the chicken, the constant-light induced changes in the anterior segment, in particular corneal curvature, anterior chamber depth and lens thickness, are closely associated with a light-induced reduction in the circulating levels of pineal derived melatonin. ^{32 34} However, the light signal that suppresses plasma melatonin levels in chickens does not necessarily originate in the retina. Eliminating all neural signals from the retina does not prevent ocular enlargement in chickens maintained on a 24-hour light cycle. ^{32 35} In fact, the anterior segment alterations produced by constant light in chicks are not vision-dependent because they occur even when the photoreceptors are destroyed with neurotoxins. ³¹ Extraocular light exposure is critical, however, because hoods that shield the pineal gland from extraocular light can prevent the anterior segment changes in chicks reared in constant light. ^{36 37} In contrast, extraocular light exposure does not suppress plasma melatonin levels or entrain circadian rhythms in mammals. ^{38 39 40} It is possible that the thicker cranium of monkeys prevents sufficient direct photoreception by the pineal gland, at least at bright room illumination levels similar to those used in this study, and thus prevents the corneal flattening and the reduction in anterior chamber depth in infant monkeys exposed to continuous light. 

Our results have important implications for the ongoing debate over whether ambient light at night promotes the development of myopia in humans. Because myopia is normally due to excessive axial elongation, the eye enlargement observed in light-reared chickens has been taken as support for the idea that nighttime lighting is a risk factor for myopia in children. ^{17 41} However, given the differences in the effects of constant light on monkeys and chickens and the close similarities between the visual systems of macaque monkeys and humans, ^{20 21} extrapolating the constant light results from chickens to humans is risky. In particular, it seems unlikely that mechanisms similar to those responsible for the constellation of macrophthalmic changes in constant-light-reared chicks somehow promote the development of common myopia in children many years after an early exposure to night-time lighting. 

Our investigation does not, however, rule out early exposure to constant light as a potential risk factor for common refractive errors like myopia. Although monkeys do not develop light-induced macrophthalmous, some of our monkeys demonstrated atypical emmetropization patterns. It is possible that these animals represent the extremes of normal emmetropization; however, the refractive anomalies observed in these monkeys suggest that exposure to constant light reduced the efficiency of the emmetropization process. In this respect, several previous observations in primates support the idea that light-driven circadian rhythms may influence eye growth. Specifically infant marmosets exhibit diurnal rhythms in axial elongation and choroidal thickness ⁴² and infant monkeys reared in constant darkness typically maintain abnormally high levels of hyperopia. ⁴³ It has also been reported that in addition to the macrophthalmic changes described above, young chicks maintained in continuous light show smaller than normal emmetropizing responses to lens-imposed defocus. ^{44 45} Given that chicks and infant monkeys exhibit qualitatively similar compensating growth responses to optical defocus, these results, together with those noted above, suggest that in constant light the vision-dependent mechanisms that influence eye growth in primates may not respond to visual feedback as effectively as they would in more normal light cycles. If this idea is correct, it could provide an explanation for the recent controversy over the association between myopia and early exposure to light at night. A high association between myopia and early night-time light exposure was found in a sample of clinical patients in a tertiary eye-care setting, ¹⁷ whereas no association was found in nonclinical, school-based populations. ^{18 19} It is reasonable to propose that the clinical patients may have exhibited this association because they had a higher than normal prevalence of large refractive errors during infancy, and a compromised emmetropization process could not overcome these errors. Consequently as children, the clinical population exhibited a larger range of refractive errors than did the infants from the general population who were also exposed to ambient light at night. In this respect, it will be important to directly test the idea that primate emmetropization does not operate as effectively in continuous light and consequently that the effective operating limits of the emmetropization process are constrained by constant exposure to high light levels. It is also important to keep in mind that thus far we have only followed our light-reared monkeys for a relatively short period that is equivalent to about the first 2 to 3 years in a human infant. In humans, myopic errors typically develop much later in childhood, and it is possible that continuous light exposure early in life somehow makes the eye more susceptible to myopiagenic factors that are normally encountered later in life. If this is the case, the prevalence of unusual refractive errors in our population of light-reared monkeys should increase with age. 

 Supported by Grants RO1 EY03611, RR07146, RO1 EY05975, and T32-EY07092 from the National Eye Institute (National Institutes of Health, Bethesda, Maryland), Grant RR00165 from National Center for Research Resources (National Institutes of Health, Bethesda, MD), and funds from the Greeman-Petty Professorship, University of Houston Foundation.

 Submitted for publication August 24, 2000; revised January 10, 2001; accepted January 26, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Axial length plotted as a function of age for the right eyes of the individual light-reared monkeys. Axial length was defined as the distance from the pole of the cornea to the vitreal-retinal interface. Open and filled symbols: monkeys reared at the University of Houston and Emory University, respectively. For each subject, the first data point represents the start of the treatment period. For the Houston monkeys the first symbol that includes a central black dot indicates the age at which the animals were returned to a normal 12-hour:12-hour light:dark cycle. The small crosses represent the data for the normal monkeys.

Figure 2.

 

Corneal radius of curvature plotted as a function of age for the right eyes of individual light-reared monkeys. Thin lines: longitudinal data for the normal monkeys. See Figure 1 for other details.

Figure 3.

 

Comparisons between lightreared and normal monkeys at 6 months of age. The spherical-equivalent refractive error (A), vitreous chamber depth (B), corneal radius (C), corneal thickness (D), anterior chamber depth (E), and lens thickness (F) are shown for the right eyes of individual normal (⋄) and light-reared monkeys (♦, •). ♦ and⋄ , the Emory and Houston monkeys, respectively. The group means are shown by the larger symbols with error bars (± SD).

Figure 4.

 

Spherical equivalent spectacle-plane refractive corrections plotted as a function of age for the right (filled symbols) and left eyes (open symbols) of the six light-reared infants treated at the University of Houston. Thin lines: normal monkeys.

Figure 5.

 

Spherical equivalent spectacle-plane refractive corrections plotted as a function of age for the right (filled symbols) and left eyes (open symbols) of the three light-reared infants treated at Emory University. The first data point represents the start of the treatment period. Thin lines: normal monkeys.

Figure 6.

 

Interocular differences in refractive error (right eye − left eye) plotted as a function of age for the light-reared (open and filled symbols) and normal control monkeys (thin lines) that were reared in Houston. The first data point represents the start of the treatment period. Arrow: age at which monkey XAV (▪) was returned to a normal 12-hour:12-hour light:dark cycle.