Abstract
purpose. Early in life, the optical demand associated with the eye’s effective refractive state regulates emmetropization in many species, including primates. However, the potential role of optical demand and/or defocus in the genesis of common refractive errors, like myopia, that normally develop much later in life is not known. The purpose of this study was to determine whether chronic optical defocus alters refractive development in monkeys at ages corresponding to when myopia typically develops in children.
methods. A hyperopic anisometropia was produced in seven adolescent rhesus monkeys by photorefractive keratectomy (PRK) with an excimer laser. Standard treatment algorithms for correcting myopia in humans were used to selectively flatten the central cornea of one eye thereby producing relative hyperopic refractive errors in the treated eyes. The laser ablation zones were 5.0 mm in diameter and centered on the monkeys’ pupils. The laser procedures were performed when the monkeys were 2 to 2.5 years old, which corresponded to onset ages between approximately 8 and 10 human years. The ocular effects of the induced anisometropia were assessed by corneal topography, retinoscopy, and A-scan ultrasonography.
results. By approximately 30 days after PRK, the experimentally induced refractive errors had stabilized and the treated eyes were between +0.75 and +2.25 D more hyperopic than their fellow eyes. Subsequently, over the next 300 to 400 days, six of the seven monkeys showed systematic reductions in the degree of anisometropia. Although some regression in corneal power occurred, the compensating refractive changes were primarily due to relative interocular differences in vitreous chamber growth.
conclusions. Vision-dependent mechanisms that are sensitive to refractive error are still active in adolescent primates and probably play a role in maintaining stable refractive errors in the two eyes. Consequently, conditions that result in consistent hyperopic defocus could potentially contribute to the development of juvenile onset myopia in children.
This study addressed the question of whether optical defocus could feasibly play a role in the genesis of common refractive errors such as juvenile-onset myopia. Early in life when the eye grows very rapidly, visual experience can have a dramatic effect on eye growth and refractive development (for reviews, see Refs.
1 2 3 4 ). In many species, including humans, depriving the eye of form vision during early infancy accelerates axial growth, resulting in substantial amounts of myopia.
5 6 7 8 9 10 Similarly, imposing relative hyperopic refractive errors with a negative lens produces compensating myopic growth in many species (McFadden S, et al.
IOVS 1995;36:ARVO Abstract 3504),
11 12 13 14 including primates.
15 16 17 18 In many respects, the experimentally induced myopia in infant monkeys is similar to juvenile-onset myopia in humans. For example, in both cases the ametropias are characterized by an increase in vitreous chamber depth
15 16 17 18 and a comparatively smaller decrease in corneal radius (Qiao Y, et al.
IOVS 2002;43:ARVO E-Abstract 2927).
2 The fact that vision-dependent mechanisms can transform a normal primate eye into a conventionally shaped myopic eye raises the possibility that the common myopic errors found in many humans are caused, at least in part, by visual experience. However, the onset of myopia in humans typically occurs after approximately 8 years of age, during a much slower phase of ocular growth.
19
Because the effects of visual experience on the eye’s refractive state decline with age,
20 21 22 it is possible that the “sensitive period” for refractive development ends too early for visual experience to play a role in the development of juvenile-onset myopia in children.
23 However, recent investigations in the chicken,
24 tree shrew,
22 marmoset (Troilo D, et al.
IOVS 1999;40:ARVO Abstract 5081), and rhesus monkey
25 have shown that form deprivation, even when it is initiated after ocular growth is nearly complete, can produce axial myopia. Thus, some degree of residual plasticity, at least to abnormal visual experience, may extend into adulthood in primates.
Although form deprivation produces myopia in adolescent monkeys, there are many gaps in our knowledge regarding vision-dependent mechanisms during the slower, juvenile phase of ocular growth. In particular, we do not know whether visual feedback associated with the eye’s refractive state influences ocular growth during the stage of eye development when myopia typically emerges in children. It is critical to evaluate the effects of optical defocus, because the mechanisms that mediate form-deprivation myopia and those responsible for the compensating refractive changes in response to altered optical demands are not identical.
26 27 28 In this respect, evidence from tree shrews suggests that optical defocus can produce compensating myopic growth during adolescence.
14 29 However, it has been reported that adolescent rhesus monkeys with form-deprivation myopia show no signs of recovery after the restoration of unrestricted vision.
30 Although there are several possible explanations for the failure of adolescent monkeys to recover from experimentally induced myopia, one possibility is that optical defocus does not produce compensating growth in the eyes of juvenile primates. Thus, the purpose of this study was to determine whether chronic optical defocus could predictably alter refractive development in adolescent monkeys at ages corresponding to when myopia typically develops in human children.
Seven normal rhesus monkeys (Macaca mulatta) that were obtained as adolescents from a government-sponsored breeding colony were used as subjects. The animals were housed in the animal care facilities of Sun Yat-sen University of Medical Sciences, where they were maintained on a 12-hour light–dark lighting cycle. All the rearing and experimental procedures were approved by the Institutional Animal Care and Use Committees at the University of Houston and the Sun Yat-sen University of Medical Sciences and were in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.
When the monkeys were between 2 and 2.5 years of age, hyperopic anisometropia was produced through photorefractive keratectomy (PRK). Based on comparisons of ocular growth, rhesus monkey eyes mature four times faster than human eyes.
31 Consequently, in human years, our monkeys can be considered to have been between 8 and 10 years of age, the age range when myopia frequently emerges in children.
32 The PRK procedure was performed while the animals were anesthetized (ketamine hydrochloride, 15 mg/kg intramuscular; acepromazine maleate, 0.15 mg/kg; 1% tetracaine, topical) using an excimer laser (Technolas 217; Bausch & Lomb, Tampa, FL) with flying spot and eye-tracking technologies. The ablation zone was 5.0 mm in diameter and centered on the animal’s pupillary axis. The corneal epithelium was intact at the onset of the procedures, and the eyes were stabilized using forceps. The ablations were performed using the laser parameters that are normally used to correct myopia in adult humans. Our goal was to produce between 1 and 2 D of relative hyperopia in the treated eyes, which required using laser parameters that were ordinarily necessary to correct 3 to 6 D of myopia in adult humans.
Immediately after the PRK procedure, 1 drop each of 2% homatropine hydrobromide and 0.3% ofloxacin (Tarivid; Aventis Pharma Ltd., Strasbourg, Germany) was instilled together with 0.3% tobramycin and 0.1% dexamethasone ophthalmic ointment (Dobradex; Alcon Laboratories, Fort Worth, TX). Postoperative analgesia was maintained by intramuscular injections of buprenorphine (0.01 mg/kg every 8 hours), and treatment with the topical antibiotic drops (0.3% ofloxacin; four times daily) was continued until epithelialization was complete. The topical steroid (3% tobramycin and 0.1% dexamethasone) was tapered over a period of 2 weeks. Slit lamp examinations were performed daily until re-epithelialization was complete and periodically throughout the observation period, to assess corneal clarity.
Each subject’s refractive status, corneal curvatures, and eyes’ axial dimensions were measured before the PRK procedure. Subsequent measures were obtained at 3, 14, and 30 days after PRK and then typically at 1-month intervals thereafter. To make these measurements, cycloplegia was induced with 3 drops of 0.5% tropicamide instilled 10 minutes apart, 45 minutes before retinoscopy was performed, and the monkeys were anesthetized with an intramuscular injection of ketamine hydrochloride (15–20 mg/kg) and acepromazine maleate (0.15–0.2 mg/kg) and the topical instillation of 0.5% tetracaine hydrochloride. Each eye’s refractive status, which is specified as the spherical equivalent spectacle-plane refractive correction, was determined along the pupillary axis by two independent observers using a streak retinoscope and handheld trial lenses. Corneal topography was assessed using a handheld videotopographer (Vista; EyeSys, Houston, TX). Corneal power was calculated using a refractive index of 1.3375 and defined as the average spherical equivalent refracting power for three consecutive readings of the central corneal radius of curvature (simulated K). The eye’s axial dimensions were measured by A-scan ultrasonography (AXIS-II; Quantel Medical Inc., Clermont-Ferrand, France). The 11-MHz transducer was placed in direct contact with the cornea with care being taken not to indent the cornea. The intraocular distances were determined using ultrasound velocities for the normal human eye and the reported data represent the average of 10 readings. To be considered an acceptable measure, the SEM of the 10 vitreous chamber readings had to be 8 μm or less.
An intact corneal epithelium was re-established within 36 to 48 hours after surgery. Trace amounts of corneal haze were observed in some of the monkeys several days after the PRK procedure. However, these slight amounts of haze disappeared by 7 days after surgery, and thereafter the corneas remained clear throughout the observation period.
Before the PRK procedures, all the experimental subjects exhibited low refractive errors that were typical of normal adolescent monkeys. The refractive errors in the right eyes varied between −0.38 and +1.25 D, with six of the seven monkeys showing low degrees of hyperopia (average = +0.53 ± 0.49 D). The right and left eyes were also well matched. Initially, there were no significant interocular differences (paired t-test, P > 0.05) in refractive error (mean anisometropia, 0.21 ± 0.29 D), vitreous chamber depth (mean difference, 0.09 ± 0.06 mm), or corneal power (mean difference, 0.20 ± 0.19 D).
Figure 1 shows the changes in spherical equivalent refractive error, vitreous chamber depth, and corneal refracting power that took place after the PRK procedures for two selected subjects. As illustrated in the bottom panels, the PRK procedures decreased the central corneal radius, resulting in a reduction in refractive power. In most animals (e.g., monkey 063,
Fig. 1 , left column), the reduction in power was greatest 3 to 7 days after the laser procedures. Subsequently, corneal power typically increased by approximately 1.0 D reaching a relatively stable value by approximately 30 days after the treatment. Between 30 and 120 days after the laser procedure, the corneas in all the treated eyes were on average 1.16 ± 0.43 D (range, 0.54–1.73) flatter than those in the fellow eyes. This reduction in corneal power produced a relative hyperopic shift in the refractive errors of the treated eyes creating a hyperopic anisometropia (
Fig. 1 , top). The magnitude of the induced anisometropia during this same period was on average +1.50 ± 0.40 D, which was 0.34 D larger than the measured amount of laser-induced corneal flattening. Based on previous observations of monkeys with imposed anisometropia,
16 it is likely that the treated animals consistently fixated with the less hyperopic, nontreated eyes and as a consequence the treated eyes of our experimental monkeys chronically experienced an amount of hyperopic defocus equivalent to the degree of anisometropia.
The data for monkeys 063 and 038 were included in
Figure 1 because, in response to the imposed anisometropia, these animals demonstrated the greatest and least amounts of compensating ocular growth, respectively. After the PRK procedure, monkey 038’s treated eye exhibited approximately 3.0 D of hyperopia and was approximately 2.0 D more hyperopic than its nontreated eye, which was the largest imposed anisometropia in our treated animals. However, despite experiencing chronic monocular defocus for the remainder of the almost 600-day observation period, the vitreous chamber depths in both eyes increased slowly at the same rate and the corneas of both eyes showed small systematic reductions in power (as represented by the negative slopes of the regression lines fit to corneal power data). Consequently, the absolute refractive errors in both eyes and the degree of imposed anisometropia were stable throughout the observation period (i.e., this monkey showed no evidence for compensating ocular growth.) On the other hand, monkey 063 showed clear evidence for differential interocular growth that compensated for the imposed anisometropia. The PRK procedure resulted in approximately 1.5 D of hyperopic anisometropia in monkey 063. The refractive errors in both eyes and the degree of imposed anisometropia were relatively stable from approximately 30 to 300 days after the laser treatment. However, beginning at approximately 300 days, there was a relative acceleration of the vitreous chamber growth in the treated eye and a concomitant systematic reduction in the treated eye’s hyperopia. Once the balance in refractive errors between the two eyes was restored, both eyes then demonstrated similar vitreous chamber elongation rates, and the refractive error balance between the two eyes was stable for the remainder of the observation period. Although the corneal power of the treated eye was essentially stable throughout the observation period, the recovery observed in monkey 063 was facilitated in part by a small relative reduction (∼0.5 D) in the corneal power of the nontreated eye.
Figure 2 illustrates the changes in the interocular differences in refractive error, vitreous chamber depth, and corneal power that took place in all the treated monkeys after the PRK procedures. To facilitate comparisons, the interocular differences in vitreous chamber depth were expressed in diopters by assuming that at the start of the experiment the optical powers of the two eyes were identical, that each eye had a single principal plane, that the distance between the retina and the principal plane was scaled to overall axial length (0.92 of the total axial length),
33 and that the refractive index of the vitreous chamber was 4/3: diopters = [1.333/(
AL · 0.92)] −{1.333/[(
AL · 0.92) +
VC]} where
AL is axial length in meters and
VC is interocular difference in vitreous chamber depth in meters. The diamond symbols and solid lines representing the interocular differences in corneal power were calculated by taking the difference between the linear regression functions that were fit to the corneal power data for the treated and nontreated eyes (see the bottom panels in
Fig. 1 ). Positive values indicate that in comparison to the nontreated eye, the treated eye was more hyperopic, had a longer vitreous chamber, and/or a steeper cornea. The data for individual monkeys have been arranged according to the magnitude of the compensating refractive error changes.
Six of the seven treated monkeys showed systematic reductions in the degree of imposed anisometropia during the observation period. For several of these monkeys, as noted in
Figure 1 , the initial degree of imposed anisometropia was stable for the first 200 to 300 days of the recovery period. Subsequently, there were systematic decreases in the degree of anisometropia that were synchronized with somewhat abrupt relative increases in the treated eye’s vitreous chamber depth (e.g.,
Figs. 2A 2B 2F ). For other animals, the decrease in the degree of anisometropia was more gradual and was associated with similarly gradual changes in the interocular differences in vitreous chamber depth (e.g.,
Figs. 2C 2D 2E ). There was a high correlation between the degree of anisometropic compensation and the interocular differences in vitreous chamber depth that developed after the PRK procedure (Pearson correlation = 0.74;
P = 0.059). In five of the six monkeys that exhibited compensating anisometropic growth (
Fig. 2 , subjects A–F), the changes in the interocular differences in vitreous chamber accounted for a greater amount of the compensating anisometropic changes than did the interocular changes in corneal power. The magnitude of the interocular corneal changes was larger in monkey 077
(Fig. 2C) , however, the vitreous chamber alterations that occurred in this animal were appropriate to account for the remainder of the compensating anisometropic changes. As illustrated by monkey 038
(Fig. 2G) , no compensating anisometropic changes were found when the axial growth rates for the two eyes were similar.
In some animals such as monkey 063 in
Figure 1 , it was evident that the interocular differences in vitreous chamber depth emerged as a consequence of an increase in the treated eye’s growth rate. There were no obvious treatment-related departures in the growth trajectories of the nontreated eyes in any of the experimental monkeys; however, subtle decreases in the vitreous chamber growth rates of the nontreated eye cannot be ruled out and could have also contributed to the observed reductions in anisometropia.
Figure 3 compares the amount of anisometropic compensation with the initial amount of imposed anisometropia for individual monkeys, where the amount of imposed anisometropia represents the average anisometropia between 30 and 120 days after PRK, and the anisometropic compensation represents the difference between these initial amounts of imposed anisometropia and the average anisometropia for the last four measurements at the end of the observation period. In six of the seven treated monkeys, (
Fig. 3 , open symbols), there was a good correspondence between the initial degree of imposed anisometropia and the degree of anisometropic change that took place during the observation period. For this subgroup, a paired
t-test indicated that there were no significant differences between the initial anisometropia and the amount of anisometropic compensation (
t = 1.52,
P = 0.19). However, as discussed earlier, one animal (monkey 038), which is represented by the filled diamond in
Figure 3 , was an obvious outlier. In this respect, it should be noted that in comparison to the other treated animals, monkey 038 also had the highest absolute degree of hyperopia in its treated eye after the PRK procedures and it manifested the largest initial degree of anisometropia.