At the start of the lens-rearing period, the eyes of the control and experimental monkeys were, on average (± SD), moderately hyperopic (right eye control monkeys, +3.65 ± 1.87 D; right eye experimental monkeys, +3.75 ± 1.21 D), and no significant interocular differences were observed in refractive error or vitreous chamber depth in control or experimental groups (paired t-tests; P = 0.12–0.88). There were also no significant differences in the initial refractive errors or vitreous chamber depths between the control group and any of the three experimental groups (two-sample t-tests for right eye data; P = 0.43–0.91).
Emmetropization proceeded rapidly in the control animals (
Fig. 1 , thin lines) with both eyes of each control monkey growing in a coordinated manner toward a low degree of hyperopia. By approximately 18 weeks of age (127 ± 7 days), the mean right eye refractive error for the control monkeys had decreased to +2.49 ± 0.99 D; 17 of the 19 control monkeys exhibited ametropia between +1.25 and +3.69 D.
Continuously wearing of –3.0 D lenses altered the course of emmetropization in a predictable manner
(Fig. 1A) . Toward the end of the rearing period, 5 of the 6 experimental monkeys in the –3 D treatment group exhibited refractive errors that were less hyperopic or more myopic than in any of the control animals. At the end of the treatment period (126 ± 4 days of age), no systematic interocular differences were observed in refractive errors in the –3 D experimental group (paired
t-test;
P = 0.56); the mean right eye refractive error for the –3 D animals was –0.68 ± 1.82 D, which was –3.17 D more myopic than the age-matched control animals. Thus, on average, compared with controls, the animals in the –3 D group had completely compensated for the optically imposed hyperopic errors.
In contrast, four daily 15-minute periods of unrestricted vision largely eliminated the predictable refractive compensation for the –3 D treatment lenses
(Fig. 1B) . Only one of the six animals in the –3 D/plano group showed evidence of compensating for the optically imposed hyperopic defocus. Three of the –3 D/plano animals exhibited refractive error changes that were comparable to those observed for most of the control animals, and two of the –3 D/plano animals showed relative hyperopic shifts in refractive error. Consequently, at the end of the treatment period, one of the –3 D/plano animals exhibited relative myopia that was outside the control range, three of the experimental monkeys exhibited refractive errors within 2 SD of the control mean, and 2 of the –3 D/plano monkeys had hyperopic errors that were more than 2 SD above the control mean.
The four daily 15-minute periods of viewing through +4.5 D lenses had a smaller effect on the refractive compensation for the –3 D treatment lenses
(Fig. 1C) . Although one of the –3 D/+4.5 D animals maintained a moderate degree of hyperopia throughout the observation period, five of the six monkeys in this group exhibited evidence of lens compensation. At the end of the lens-rearing period, two of these five animals exhibited absolute myopic ametropias, and three more of these animals had refractive errors that were less hyperopic or more myopic than in the control animals.
Figure 2summarizes the refractive development for the control and experimental animals. The left panel shows refractive error growth curves for each subject group determined using a locally weighted regression, scatter plot–smoothing algorithm (LOESS plots
26 ), and the right panel includes individual and mean refractive errors for the control and experimental subjects at the end of the treatment period. The pattern of refractive development for the –3 D monkeys showed a clear myopic trajectory compared with that in control animals. At the end of the treatment period, the average ametropia for the –3 D group was significantly more myopic than that for the control monkeys (–0.68 vs. +2.49 D; two-sample
t-test, T = –4.06,
P = 0.01). On the other hand, the pattern of refractive development for the –3 D/plano monkeys was comparable to that for the controls, and the average end-of-treatment refractive error for the –3 D/plano animals was not different from that for the control monkeys (+2.56 vs. +2.49 D; two-sample,
t-test, T = 0.07,
P = 0.95), but it was significantly more hyperopic than that for the –3 D monkeys (+2.56 vs. –0.68 D; two-sample
t-test, T = 2.67,
P = 0.04). The refractive error growth curve for the –3 D/+4.5 D group exhibited an initial myopic growth trajectory that was similar to the that for the –3 D group, but the function leveled off at a more hyperopic level. As a result, over the course of the treatment period, the average changes in refractive error for the –3 D/+4.5 D group (–2.74 D) were intermediate compared with those for the –3 D animals (–4.24 D) and for the –3 D/plano (–1.61 D) and control (–1.14D) groups. At the end of the treatment period, the differences in the average refractive errors for the control and –3 D/+4.5 D groups approached statistical significance (+2.49 vs. +0.77 D; two-sample
t-test, T = –2.11,
P = 0.09) and the median refractive errors for these groups were significantly different (+2.44 vs. +0.47 D; Mann–Whitney
U test,
P = 0.01). However, neither average (+0.77 vs. –0.68 D; two-sample
t-test, T = –1.34,
P = 0.21) nor median (+0.47 vs. –1.19 D; Mann–Whitney
U test,
P = 0.17) refractive errors for the –3 D/+4.5 D monkeys were significantly different from those for the –3 D animals or from those for the –3D/plano animals (average: +0.77 vs. 2.56 D; two-sample
t-test, T = 1.31,
P = 0.23; median: +0.47 vs. 2.32 D; Mann–Whitney
U test,
P = 0.17).
Differential effects of the treatment regimens on refractive development were largely axial in nature and specifically associated with differences in vitreous chamber depth.
Figure 3illustrates the vitreous chamber growth curves for each subject group (left panel) and the relationship between vitreous chamber depth and refractive error for individual subjects at the end of the treatment period (right panel). Compared with control monkeys, the animals in the –3 D and –3 D/+4.5 D groups exhibited faster vitreous chamber growth rates and, on average, longer vitreous chamber depths at the end of the treatment period (–3 D group, 10.04 ± 0.52 mm; –3 D/+4.5 D group, 10.06 ± 0.67 mm; controls, 9.68 ± 0.25 mm). However, these differences were not statistically significant (two-sample
t-tests,
P = 0.16 and 0.23). On the other hand, the vitreous chamber growth curve for the –3 D/plano monkeys was similar to that for the control animals. At the end of the treatment period, the average vitreous chamber depth for the –3 D/plano monkeys (9.72 ± 0.68 mm) was comparable to that for the control animals. Although the variance of the vitreous chamber depths in the experimental groups prevented the average differences from reaching statistical significance, when the data for all subjects were analyzed together, refractive error was significantly correlated with vitreous chamber depth (Pearson
r = –0.65;
P = 0.004).