The high prevalence of myopia (33% in the United States and as high as 80% in some countries in Asia) and the large number of individuals with uncorrected refractive error in some parts of the world make myopia a significant public health issue [1, 2]. To correct myopia, single vision spectacle lenses and contact lenses are commonly prescribed, with refractive surgery an increasingly popular option in adults. While these treatments correct the myopic refractive error, they do not slow the accompanying eye growth and associated changes in the eye [3]. The high prevalence of myopia and its prominence as a public health problem emphasize the importance of understanding the mechanisms of eye growth and of finding effective treatments that slow myopia progression and axial elongation.

COMET

Different types of spectacle lenses, such as bifocals and progressive addition lenses (PALs), have been evaluated for slowing progression in several recent clinical trials [4]. The largest of the treatment trials using this type of lens was the Correction of Myopia Evaluation Trial (COMET), a multi-center, randomized, double-masked clinical trial supported by the National Eye Institute of the National Institutes of Health in the United States to evaluate whether PALs (Varilux Comfort with a +2.0 D addition) slowed the rate of progression of myopia compared to conventional single vision lenses (SVLs) [5]. COMET enrolled 469 children aged 6 to < 12 years who were ethnically diverse (46% white, 26% African-American, 14% Hispanic, and 8% Asian) and had baseline myopia between -1.25 D and -4.50 D. The main outcome measure was progression of myopia determined by autorefraction after cycloplegia with two drops of 1% Tropicamide. Axial length, a secondary outcome measure, was assessed by ultrasonography, as shown in Figure 1. Ocular components were measured in COMET children to determine if any slowing of myopia progression was accompanied by reduced axial elongation. Based on results using various animal models of myopia and eye growth, we would expect such an association [6].

Fig. 1: Axial length measurement of a COMET subject by ultrasonography.

With respect to the 3-year outcome data of COMET, retention was excellent, with 462/469 (98.5%) of the children completing the three-year visit. Three-year increases in myopia were -1.26 ± 0.06 D for the PAL group and -1.46 ± 0.07 D for the SVL group. The overall adjusted 3-year treatment effect of 0.20 ± 0.08D was statistically significant (p = 0.004) but not clinically meaningful. The mean increases in the axial length of eyes of children in the PAL and SVL groups, respectively, were: 0.64 ± 0.02 mm and 0.74 ± 0.02 mm, with a statistically significant difference of 0.10 ± 0.03 mm (p = 0.0009).

Cross-sectional relationship between myopia and axial length

At baseline, the Pearson correlation coefficient between axial length and spherical equivalent myopia was 0.32 (p< 0.001) [7]. This value is lower than what has been reported in other studies, in part because of the limited range of myopia used as an inclusion criterion for enrollment in COMET. To further illustrate this point, after five years of follow-up in COMET children, the correlation between axial length and refractive error was 0.50, presumably because of the increased range of myopia after that period of time. The relatively low correlations are also due to the influence of the cornea, as illustrated by the gender differences seen in COMET data at baseline. At the start of COMET girls had significantly shorter eyes than boys (OD: 23.92 mm versus 24.36 mm, p < 0.0001), even though they had a similar amount of baseline myopia (-2.40 D) as boys (-2.35 D). Keratometry showed that girls had significantly steeper corneas than boys, which could account for having the same amount of myopia even with shorter eyes.

Longitudinal results for myopia and axial length

Unlike the relatively low correlation between the amount of baseline myopia and axial length in COMET children and the somewhat higher 5-year correlation, the changes in myopia and axial length were found to be highly correlated [5]. At three years the correlations between the changes in myopia and axial length were 0.86 for the children wearing PALs and 0.89 for the children in SVLs. After five years the correlation was also high (r = 0.89), as shown in Figure 2. These high correlations show that the progression of myopia was mainly axial in nature.

Fig. 2: Scatter plot showing change in myopia vs. change in axial length after 5 years.

Ocular components other than vitreous chamber depth did not change very much over three years, and similar results were found in the PAL and SVL groups. Mean three-year changes in the eyes of COMET children in the PAL and SVL groups, respectively, were 0.06 ± 0.11 mm, 0.07 ± 0.09 mm (anterior chamber); -0.01 ± 0.10 mm, - 0.01 ± 0.08 mm (lens thickness); and 0.56 ± 0.33 mm, 0.65 ± 0.34 mm (vitreous chamber). The three-year difference between groups was significant only for vitreous chamber depth (difference = -0.09 ± 0.03 mm, 95% CI: -0.15 to -0.03; p = 0.002). Lens thinning cannot account for the differential progression of myopia in the two treatment groups, since little evidence of lens thinning was found during the three years of follow up. Mean changes in corneal radii were 0.03 ± 0.03 D for the PAL group and 0.03 ± 0.07 D for the SVL group in the horizontal meridian, and -0.01 ± 0.05 D for the PAL group and -0.01 ± 0.05 D for the SVL group in the vertical meridian. These values did not differ by treatment group.

At baseline in COMET, factors that were independently related to 3-year myopia progression and axial elongation included age and ethnicity [8]. Younger age at baseline was associated with faster progression, with myopia in 6-7 year-old children progressing 1.3 D more than myopia in 11-year olds over 3 years. African-American children had slower progression and eye growth than children of Asian, Hispanic, mixed, and white ethnicity. Gender was found to be related to myopia progression but not axial elongation, with slightly faster progression in girls. Among these myopic children, a 0.5 mm increase in axial length was associated with 1.0 D of myopia progression, a finding that differs from the commonly reported association of a 1.0 mm change in axial length corresponding to 3.0 D of progression.

Extending the benefit of treatments for myopia

In COMET the treatment benefit of PALs for both refractive error and axial length was observed at one year, and sustained at that level over the next two years. Similar to these results, other clinical trials of spectacle lenses, contact lenses, and pharmaceutical treatments have shown that most therapies for myopia have relatively small treatment benefits in the first few months that do not grow over time and thus cannot be considered clinically meaningful [4]. Pos-sible solutions to this problem, which could be evaluated in future studies, include combining treatments (e.g., lenses and eye drops), switching from one treatment to another when the first one no longer slows myopia progression, or introducing periods of time without any treatment.

One reason why treatments may show limited benefits is because inclusion criteria for clinical trials typically are quite broad in order to achieve generalizability, yet specific treatments are not likely to work for all myopic children. It may be helpful to take into account factors such as the amount of myopia, the progression of myopia in the past year, oculomotor characteristics (e.g., accommodation and phoria), and parental refractive state when considering treatment options for a particular child. For example, subgroup analyses in COMET showed that PALs were more effective than SVLs in children with low myopia, large accommodative lags, near esophoria, and two myopic parents [9, 10].

If a treatment were found to slow progression over many years and with minimal side effects, then it would be beneficial to consider using that therapy for a child at risk for the development of myopia. However, at this time we cannot identify with certainty an at-risk child likely to develop myopia. Having two myopic parents and an emmetropic refractive error at age 5 years are strongly associated with later myopia development in a child. However, the predictive potential of these risk factors is not high enough to justify the cost and possible side effects of treatment for a child who may never develop myopia. Another concern is whether a non-myopic child would comply with any treatment, such as wearing lenses that are not needed to see clearly at distance. On the other hand, individuals might be more motivated to comply with any treatment that had been shown to prevent the development of myopia.

Future treatment options

Two possible treatment options that might be recommended in the future are correction of peripheral refractive errors, and many hours of outdoor activity for children. Recent animal work has suggested that visual signals from the fovea may not be essential for normal eye growth since the peripheral retina appears to be able to regulate emmetropization and induce myopia in response to abnormal visual input [11]. Correction of peripheral refractive errors might be achieved by specially designed contact lenses worn during the day or by corneal refractive therapy with lenses worn during the night. One of the simplest therapies for retarding myopia could turn out to be providing children with substantial hours of outdoor activity each week, though the mechanisms by which outdoor activity might slow eye growth need additional study [12].

In summary, in order for treatments to be effective over long periods of time and without producing significant side effects, more research is needed to understand the underlying mechanisms of myopia development and eye growth.