Materials and methods

The CRAYON Pilot Study followed the tenets of the Declaration of Helsinki and was approved by The Ohio State University Biomedical Institutional Review Board. All parents provided informed consent, and child assent was attained from all subjects.

All subjects were 8 to 11 years old at the baseline visit. They had between −0.75 D and −4.00 D spherical component myopia and less than −1.00 D astigmatism by cycloplegic autorefraction. Cycloplegia was achieved by administering one drop of 0.5% proparacaine, followed by two drops of 1% tropicamide administered 5 min apart. All subjects had 20/20 or better visual acuity in each eye and good ocular and systemic health. They had not previously worn gas-permeable contact lenses, they were not taking medications that might affect contact lens wear, they were not participating in other eye or vision studies, and they had no previous eye surgeries.

Eligible subjects were matched by age category (8 or 9 years vs 10 or 11 years) to a historical control subject who was randomly assigned to wear soft contact lenses during the Contact Lens and Myopia Progression (CLAMP) Study.18 After participating in a run-in period of gas-permeable contact lens wear for an average of 2 months, CLAMP Study subjects were randomly assigned to wear gas-permeable or soft contact lenses for the remainder of the study. This randomisation visit was considered the basis for determining the timing of all subsequent visits and served as the baseline for measuring changes during the CLAMP Study. Only subjects randomly assigned to wear soft contact lenses were matched to a corneal reshaping contact lens wearer.

The primary outcome of the CRAYON Pilot Study was the difference in the 2-year change in axial length, measured by a-scan ultrasound, between corneal reshaping and soft contact lens wearers. Secondary outcomes included comparisons of anterior chamber depth, lens thickness and vitreous chamber depth between corneal reshaping and soft contact lens wearers. Refractive error and corneal curvature are temporarily altered by orthokeratology, so they are not compared in this investigation.

A-scan ultrasound

The IOLMaster was not available at the beginning of the CLAMP Study, so no baseline measurements were available for comparison with the IOLMaster. Therefore, A-scan ultrasound measurements were performed and edited until five readings with high, equal lens peaks and a distinct, anterior scleral peak were recorded using identical protocols. The measurements were performed while the subject viewed a distant target under cycloplegia using one drop of 0.5% proparacaine, followed by two drops of 1% tropicamide administered 5 min apart. The examiner lightly touched the cornea with a hand-held probe until a reading was automatically recorded in the automatic mode. The vitreous chamber depth was calculated by subtracting the anterior chamber depth and the lens thickness from the axial length. The baseline measurements were performed at the initial visit for the CRAYON Pilot Study subjects and at the randomisation visit for the CLAMP Study subjects.

Contact lenses

Corneal reshaping contact lens wearers were fitted with Corneal Refractive Therapy (Paragon Vision Sciences, Mesa, Arizona) contact lenses, using HDS-100 materials, according to the manufacturer’s directions. In summary, the spherical component of the manifest refraction and the flat keratometry meridian were used to determine the initial trial lens, which was placed on the eye and evaluated for a proper fit. Subjects were also provided with Unique-pH Multi-Purpose Solution (Alcon, Ft Worth, Texas) and their contact lenses were occasionally cleaned in-office with Progent (Menicon USA, San Mateo, California). Subjects fitted with soft contact lenses wore Focus 2-week disposable contact lenses, and they were given SOLO Care multi-purpose solutions (CIBA Vision Care, Duluth, Georgia). All subjects received free contact lenses, solutions and eye care throughout both studies.

Statistics

The analyses include only the 28 subjects who completed the entire study and the control subjects to which they were initially matched. Repeated measures over time were collected. For the analysis, data were treated as eyes nested within subjects nested within pairs. A subject with no missing data could have six measures of each outcome, one from each eye in each year. We used multilevel modelling, a generalisation of multiple regression that handles clustered observations, to model each outcome.

For each outcome, a linear growth model was fitted. The model used random effects for pair, subject within pair, and eye within subject which adjusted the mean growth curve intercept and the mean growth curve slope. The effects of treatment, visit and their interaction were then evaluated for each of the ocular components of interest. All analyses were conducted using Statistical Analysis Software (SAS) version 9.1 (SAS Institute, Cary, North Carolina).

Results

Forty subjects were enrolled in the CRAYON Study between 30 September 2004 and 2 March 2005. Twenty-eight of the 40 (70%) completed the 2-year study. Subjects dropped out of the study before attending the 1-day (n = 4), 10-day (n = 4), 6-month (n = 2), 1-year (n = 1) and 2-year (n = 1) visits. None of the drop-outs were due to complications; the vast majority was due to lack of interest in contact lens wear after the initial experience. The baseline demographic and ocular characteristics are similar between the subjects who completed the study and the subjects who did not complete the study (table 1), and also between the corneal reshaping who remained in the study for 2 years and the age-matched soft contact lens wearers (table 2).

Table 3 shows the mean (SD) ocular components for each treatment group, and table 4 shows the differences between the groups at each visit. The eyes of both treatment groups grew (both p = 0.0001); however, the annual rate of change in axial length was on average 0.16 mm per year less (p = 0.0004) for corneal reshaping contact lens wearers than soft contact lens wearers (fig 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Mean (SD) axial length (mm) during each year of the study.

Vitreous chamber depth change was similar to that of axial length. There was a statistically significant positive rate of change in vitreous chamber depth in both groups (p<0.0001). Vitreous chamber depth grew 0.10 mm per year faster for soft contact lens wearers than corneal reshaping contact lens wearers (treatment×visit interaction p = 0.006).

The annual rate of change in anterior chamber depth of corneal reshaping contact lens wearers was not statistically significant (mean change = −0.01 mm, p = 0.63), but the rate of change in anterior chamber depth of soft contact lens wearers was statistically significant (mean change = 0.05 mm, p = 0.0005). On average, the anterior chamber depth of the soft contact lens wearers increased 0.06 mm per year more than the anterior chamber depth of the corneal reshaping contact lens wearers (treatment×visit interaction p = 0.004).

Lens thickness did not exhibit any statistically significant annual change for either group (p>0.43); nor was there a statistically significant difference in rate of change between the treatment groups (p = 0.47).

Discussion

Corneal reshaping contact lenses provide short-sighted patients with clear vision without requiring vision correction to be worn during the day.7 8 9 10 11 Results of this study confirm results from prior investigations that indicated that corneal reshaping contact lenses also slow the progression of myopia in children.13 16 17 How these contact lenses may control myopic eye growth is still debatable.

Recent animal studies indicate that the peripheral retina is more responsible for regulation of eye growth than previously thought.19 20 In infant monkeys, form deprivation limited to the peripheral retina produced myopic eye growth, and all monkeys recovered from the induced refractive error, regardless of whether their fovea was ablated with an argon laser.20 Furthermore, ablation of the fovea at an early age did not prevent emmetropisation typically experienced by infant monkeys; nor did it prevent refractive error development induced by form deprivation.19

In humans, myopic eyes experience relative hyperopia in the periphery that hyperopic and emmetropic eyes do not,21 and children who become myopic have more relative hyperopic peripheral blur than emmetropic children 2 years before the onset of myopia.22 Patients with peripheral hyperopic defocus are also more likely to develop myopia.23 Therefore, peripheral hyperopia may act as a signal for increased eye growth.

The focus of light posterior to the peripheral retina may act as a signal for continued myopic eye growth. The current theory of myopia control with corneal reshaping contact lens wear is that the oblate shape of the cornea and the “knee” where the oblate portion of the cornea returns to its original curvature cause peripheral light rays to focus anterior to the peripheral retina. This results in an image shell that provides focused light centrally at the fovea, while the peripheral retina experiences myopic defocus that results in slowed axial growth. There is still much to learn about the role of the peripheral optical profile in regulation of eye growth, but peripheral myopic defocus currently is the leading theory to explain the potential myopia control effect of corneal reshaping contact lenses.

The results of our study indicate that eye growth is slowed by 55%, which is similar to the 46% slowed axial elongation reported by Cho and colleagues.17 Corneal reshaping contact lenses provide clear vision by flattening the cornea, which in theory would shorten the axial length of the eye and could explain the observed treatment effect. In fact, our study found that the anterior chamber depth increased significantly more for soft contact lens wearers than corneal reshaping contact lens wearers. However, when our study and the study by Cho and colleagues eliminated the effects of the anterior chamber by measuring the growth of only the vitreous chamber depth, a significant treatment effect was still present.17 This indicates that some signal must act to slow the myopic eye growth.

Limitations of the current study

Nearly one-third of the subjects withdrew from the study prior to the conclusion. If poor adaptation to corneal reshaping contact lens wear was related to the treatment response (perhaps flatter corneas make it more difficult to adapt and decrease the treatment response), the results would appear to show significant myopia control when in reality the results were skewed because the treatment effects were only measured for the subjects who benefitted most. However, there were no significant differences in baseline demographic or biometric measures between the subjects who were lost to follow-up and those who remained in the study, decreasing the risk of bias.

The examiners were not masked to the treatment group of the subjects because only the corneal reshaping contact lens wearers were actively enrolled in the study. The control subjects were matched to the corneal reshaping subjects following participation in a different study. Although the examiners were not masked by treatment group, it is unlikely that their knowledge could have influenced the outcome dramatically. The same procedures were used for both studies, and the outcome was change over time. In order to affect the outcome, a difference in procedures would have had to taken place after the baseline visit, but the same procedures were used throughout the study.

Some believe that use of soft contact lens wearers as the control group may artificially inflate the treatment effect experienced by corneal reshaping contact lens wearers due to “myopic creep” that has been reported following initiation of soft contact lens wear. However, studies of adolescents with at least 1 year of follow-up have shown that soft contact lens wear does not increase myopia progression.24 25 Soft contact lenses do not increase myopia progression or eye growth compared with spectacles, so soft contact lens wearers are an appropriate control group.

Conclusion

Despite the fact that several studies have investigated myopia control over the past few years, an effective treatment with few side effects is still to be discovered. Corneal reshaping contact lens wear holds promise for myopia control. It has now been shown by two separate controlled trials to slow the axial growth of the eye. However, further investigations must apply the gold standard study design, a randomised clinical trial, to definitively determine whether or not corneal reshaping contact lenses slow the growth of the eye. Investigations must also attempt to determine the mechanism of the treatment effect and why the treatment effect may continue beyond the first year.