Childhood Myopia Part II: Treatment Mechanisms, Emerging Options, and Considerations

Abstract

The rapidly increasing prevalence of myopia and its association with vision-threatening complications in later life has intensified efforts to identify treatments that can prevent the onset and slow the progression of myopia in children. These efforts have led to the development of several optical, environmental, and pharmacological myopia control treatments. Investigations of additional treatment options are ongoing and have shown great promise for clinical translation. Although significant progress has been made in identifying potential pathways and modes of action of these treatments, a complete mechanistic understanding is still lacking. This critical review article explores the potential mechanisms underlying the myopia control effects of contemporary treatments for childhood myopia. It also provides an evidence-based overview of emerging and experimental myopia control therapies. Additionally, the article addresses key considerations, including patient-related factors, treatment limitations, clinical trial designs, efficacy, and cost-effectiveness. This article is the second part of the commissioned paper “Treatment of Childhood Myopia” submitted to the National Academies of Sciences, Engineering, and Medicine for the consensus study Focus on Myopia–Pathogenesis and Rising Incidence. For a comprehensive overview of contemporary treatments for childhood myopia in and outside the United States, readers are referred to Childhood Myopia Part I: Contemporary Treatment Options. The findings in this review article underscore the need for continued research to deepen our understanding of the mechanisms underlying myopia treatments, evaluate the long-term safety and efficacy of current and emerging therapies, and identify patient- and treatment-related factors to optimize treatment outcomes in the clinical care of children with myopia.

Efforts to develop effective strategies to control childhood myopia have intensified owing to the rapid increase in myopia prevalence globally and the potential risk of vision loss from myopia-associated ocular pathologies in later life.¹ These efforts have resulted in the development of several treatment strategies that have been found to be safe and effective in randomized clinical trials.^2,3 Contemporary treatments for childhood myopia control include atropine eye drops, dual-focus, bifocal, and multifocal contact lenses, orthokeratology lenses, specialized spectacle lenses, and time outdoors.^2,4,5 For a comprehensive review of these treatments, readers are referred to “Childhood Myopia Part I: Contemporary Treatment Options.” Several alternative myopia control therapies are on the horizon, including phototherapies like blue-light stimulation of the optic nerve head and violet light. Although these emerging therapies have shown promising results in preclinical or clinical studies, several issues remain unresolved, including mechanisms of action, treatment safety, and long-term effectiveness. 

The objective of this critical review article is to summarize current knowledge regarding the mechanisms of contemporary treatments and development of new treatments for myopia control in children. First, the article explores the potential mechanisms of currently available treatments for childhood myopia with synthesis of evidence from basic, clinical, and translational studies. Next, it provides an overview of emerging and experimental treatment options with a focus on potential efficacy and mechanisms from animal and human studies. Then, it discusses treatment trajectories, limitations, and cost-effectiveness. Finally, the article concludes with some key clinical and patient-related considerations regarding the treatment of myopia in children. This paper is the second part of the commissioned paper “Treatment of Childhood Myopia”⁶ submitted to the National Academies of Sciences, Engineering, and Medicine for the consensus study “Focus on Myopia–Pathogenesis and Rising Incidence.”¹ 

Mechanisms of Contemporary Myopia Treatments

Pharmacological Treatment - Atropine

Despite the long history of atropine eye drops in myopia control, the site and mode of atropine's anti-myopia action remains unclear. The earlier belief was that excessive or prolonged accommodation is the primary cause of myopia,⁷ so atropine was thought to be acting by paralyzing accommodation via its strong cycloplegic effects. However, the observation that atropine is also effective in preventing experimental myopia in chicks without blocking accommodation (as they possess striated muscles with nicotinic receptors)⁸ suggested a non-accommodative mechanism for atropine's anti-myopia effects. Although muscarinic receptors are widespread in the eye,⁹ insights from subsequent animal and human studies have provided compelling evidence for the current hypothesis that atropine presumably acts at one or more sites along the retina-choroid-sclera defocus signaling pathway. In chicks, intravitreal injection of atropine was found to increase retinal dopamine levels, which are normally reduced in experimentally induced myopia, and subsequently trigger spreading depression effects in the retina.¹⁰ In mice, the levels of GABA transporter 1, normally elevated in the myopic retina, were reduced after atropine administration.¹¹ It has also been proposed that the light-adaptive signaling molecule nitric oxide could mediate the inhibition of form-deprivation myopia by atropine in chicks, most likely in the retina.¹² Findings from these animal studies support a retinal site of atropine's action, which presumably begins in M1/4 receptors to influence the signaling cascade (Table 1). 

Table 1.

 

Contemporary Treatments for Childhood Myopia: Potential Mechanisms, Evidence of Efficacy, and Limitations

 

However, the exact retinal location and mechanism remain elusive, as atropine can still block myopia development in the chick model of myopia despite the destruction of retinal amacrine cells containing muscarinic receptors.¹³ Furthermore, atropine causes changes in muscarinic receptor density in the iris and ciliary body but not in the retina,¹⁴ and there is support for non-muscarinic targets for atropine, including α2A-adrenoceptors, at least in vitro.¹⁵ These findings suggest, perhaps, that atropine's action to inhibit axial elongation involves non-muscarinic receptors directly at the level of the choroid or the sclera. In humans, atropine eye drops cause thickening of the choroid¹⁶ and also abolish the choroidal thinning typically associated with exposing the retina to hyperopic defocus.¹⁷ Atropine also alters protein expression levels in the sclera¹⁸ and induces morphological changes, such as thickening of the fibrous layer and thinning of the cartilaginous layer, which are typically known to occur in the opposite direction in experimental myopia in chicks.¹⁹ Furthermore, muscarinic receptor antagonists inhibit the synthesis of DNA and glycosaminoglycan in scleral chondrocytes despite chicks lacking any apparent source of acetylcholine.²⁰ Non-retinal sites of action also seem likely because myopia prevention in chicks typically requires high nano-molar concentrations of atropine.¹³ 

Whether atropine inhibits the generation of the defocus signal or influences the transmission or reception of the signal is unclear. In humans, atropine selectively enhances the inner retinal responses induced by myopic defocus in the peripheral retina,²¹ affects neural activity of photoreceptors,²² alters diurnal rhythms of ocular components,²³ and reduces relative peripheral hyperopia.²⁴ These findings suggest that atropine directly influences one or more components of the retina-choroid-sclera signaling pathway that regulates the growth and refractive state of the eye. 

Repeated “Laser” Red Light Therapy

The strong myopia control efficacy of repeated laser red light (RLRL; also defined as repeated low-level red light) therapy, as demonstrated in several randomized controlled trials (see the reviews^25,26), is not surprising, considering prior evidence from wavelength-rearing studies in animal models. It has been shown that exposure to limited-bandwidth long-wavelength lights (red/amber, illumination approximately 500 lux) not only produces substantial hyperopia but also inhibits induced myopia (produced by imposing negative spherical lenses) in tree shrews and macaque monkeys.^27–30 Even 1 hour of exposure produces significant effects in animals, showing a temporal nonlinearity of the red light effect.³¹ The red light hyperopia effect is thought to originate from the difference in image contrast between longer-wavelength sensitive and short-wavelength sensitive cones arrays, although the precise physiological and molecular mechanisms are elusive.³² An opponent dual-detector spectral drive model of emmetropization has been proposed whereby the emmetropization mechanism modulates eye growth by comparing image contrast across the cone arrays. A sharper image contrast on the longer-wavelength cones (“red” image contrast) could signal that the eye is too long and should slow its growth; conversely, a sharper image contrast on the short-wavelength cones (“blue” image contrast) could signal that the eye is too short and should accelerate its growth.^33,34 Whether this model can explain the beneficial effect of RLRL in slowing myopia progression is yet to be understood. A critical difference between the preclinical and clinical studies has been the light source: animal studies have used light-emitting diodes (LEDs), whereas RLRL studies have used semiconductor laser diodes. It is also uncertain whether the treatment duration (3 minutes twice a day) used in the clinical trials is adequate for the emmetropization mechanism to initiate and integrate growth signals based on comparisons of image brightness across the cone arrays, as proposed for the red-light effect in animal models that involve much longer treatment periods.³⁴ A recent study showed that a 10-minute exposure to LED-generated incoherent narrow-band red light induced axial shortening but exposure to LED-generated near-infrared light had no effect in humans.³⁵ This finding suggests that the effects of red light is likely through visible stimulation and not thermal energy, raising question over the choice of coherent laser red light as an anti-myopia therapy for children. 

An alternative mechanism for the myopia control effect of RLRL is also likely given multiple reports of axial length shortening after treatment in children.^36,37 Physiologically, axial shrinkage of the eye globe seems inconceivable. One possibility is massive choroidal swelling, but changes in choroidal thickness were reported to explain 28.3% observed axial length changes,³⁷ suggesting other posterior segment events (e.g. inflammation) could be responsible. Explaining these observations without an understanding of the mechanism of treatment is challenging. Unfortunately, little information is available on the mechanism by which the RLRL treatment slows myopia in children. Several mechanisms have been proposed (Table 2), including improved choroid blood flow, cytochrome and nitric oxide signaling, mitochondrial effects, and reduction of scleral hypoxia.³⁸ However, these mechanisms remain speculative and require further investigations. 

Table 2.

 

A Summary of Emerging and Experimental Interventions for the Treatment of Childhood Myopia

 

Optical Treatments

Experiments in a variety of animal models have consistently shown that altering the visual input with the application of lenses can modulate eye growth and refractive state.^39–45 Imposing a negative lens on the eye induces hyperopic defocus (image plane behind the retina) and causes the eye to elongate and become myopic; conversely, imposing a positive lens on the eye induces myopic defocus (image plane in front of the retina) and causes the eye to slow the growth and become hyperopic.^39,46,47 Smith et al. found that applying form deprivation and hyperopic defocus only in the periphery with clear central vision led to myopic eye growth in rhesus monkeys. Without the peripheral form deprivation, the animal eyes continued to emmetropize.^48,49 Ablation of the fovea and parafovea did not affect myopic eye growth or emmetropization.⁵⁰ In humans, lenses designed to reduce peripheral hyperopic defocus have also been found to slow myopia progression and axial elongation in children.^51,52 These results suggest that the peripheral retina likely plays a critical role in controlling eye growth. 

The peripheral defocus theory is the prevailing hypothesis as to how optical treatments, such as dual-focus and multifocal soft contact lenses, orthokeratology lenses, and defocus-incorporating spectacle lenses slow myopia progression.⁵³ This theory was first proposed in 1971 by Hoogerheide and Rempt, who evaluated myopia development in young pilots,⁵⁴ although their data do not provide evidence for a predictive role as measurements of refraction were not made at the baseline but after some participants developed myopia.⁵⁵ Individuals with myopia generally have a relative hyperopia in the retinal periphery with or without correction with single-vision spectacle lenses.⁵⁶ The design of dual-focus and multifocal soft contact lenses and specialized spectacle lenses create simultaneous myopia defocus in the retinal periphery, which could signal the emmetropization mechanism to slow eye growth. Measurements of peripheral refraction through dual-focus contact lenses and multifocal spectacle and contact lenses have found relative myopic shifts in peripheral refraction compared with single-vision lenses.^57–59 Children wearing defocus-incorporated multiple segment (DIMS) lenses showed no change in relative peripheral refraction during treatment for 2 years, but those wearing single-vision lenses continued to show hyperopic shifts.⁶⁰ When children were switched from single-vision lenses to DIMS lenses, a reduction in relative peripheral hyperopia was found, although changes were asymmetrical across the retina regions.⁶¹ Other studies have demonstrated similar changes in peripheral refraction after wearing orthokeratology lenses,^62,63 which likely generate peripheral myopic defocus by steepening the mid-peripheral cornea.⁶⁴ These results support the peripheral refraction theory of optical myopia treatments. However, the optimal magnitude and retinal eccentricity for the maximal efficacy of inducing peripheral myopia defocus remain unclear. The magnitude of peripheral retinal defocus could have a dose-related relationship with treatment efficacy, producing a greater reduction in myopia progression with relatively stronger power treatment zones. For example, higher-add power distance-centered multifocal lenses that produce stronger relative peripheral myopic defocus was found to have a greater treatment effect than +1.50 add multifocal lenses.⁵² 

However, the peripheral retinal defocus theory of myopia is not universally accepted, and the link between peripheral refraction and myopia development remains a matter of great debate.^55,65,66 Several longitudinal studies have shown no association between peripheral refraction and myopia,^65,67–69 suggesting that relative peripheral refraction may not predict the development of myopia. It has also been shown that single-vision soft contact lenses induce relative peripheral myopia compared with single-vision spectacles⁵⁶ despite both being ineffective in slowing myopia progression.⁷⁰ In the Collaborative Longitudinal Evaluation of Ethnicity and Refractive Error (CLEERE) study, children who became myopic had more hyperopic relative peripheral refractive errors than did children with emmetropia from 2 years before onset through 5 years after onset of myopia.⁷¹ These findings suggest that peripheral refraction changes could be a consequence of myopia rather than a cause. More recently, the Bifocal Lens for Nearsighted Kids (BLINK) study found that peripheral refraction changes accounted for only 30% of the slowing of axial growth in children wearing +2.50 add multifocal contact lenses,⁷² providing evidence that peripheral defocus was not responsible for the changes in axial elongation with distance-centered soft multifocal contact lenses. Moreover, the relative positive power of the treatment zones in optical interventions may not translate to the same magnitude of defocus in the retinal periphery.⁷³ 

Apart from peripheral defocus, several other mechanisms have been proposed to explain the efficacy of myopia treatments. The accommodative theory is based on the relationship between accommodative lag and myopia development. A greater accommodative lag has been found in myopes than in non-myopes.^74,75 This increase in the lag of accommodation could produce an increase in on-axis hyperopic defocus, which is a stimulus for eye growth.⁷⁶ However, the role of accommodation in myopia development and progression is unclear.⁷⁷ In children, accommodative lag is not associated with myopia progression,⁷⁸ and accommodative amplitude, lag, or facility remains unchanged after 5 years of treatment with multifocal contact lenses.⁷⁹ Moreover, strategies aimed at reducing the inaccuracy of accommodative responses have yielded minor treatment effects.^80,81 

Another potential mechanism is the direct effect of simultaneous retinal defocus on retinal signals of eye growth. It has been demonstrated that dual-focus lenses modify electrophysiology responses, suggesting a direct effect on retinal signals of eye growth.^82,83 Higher-order aberrations (HOAs) have also been proposed as a possible mechanism for slowing myopia progression. HOAs are imperfections in the optical system that cannot be corrected by sphero-cylindrical lenses.⁶⁴ Individuals with myopia tend to be less sensitive to negative defocus, which is interpreted as a larger difference in peripheral sensitivity due to HOAs causing differences in the depth of field.⁸⁴ The reshaping of the cornea from the orthokeratology lens produces an increase in HOAs, specifically spherical aberration.^85–87 The amount of HOAs induced by orthokeratology lenses continues to increase until around 1 month of wear, at which time the aberrations tend to stabilize.⁸⁸ The amount of HOAs also increases with increased pupil size,⁸⁹ again implying that pupil size may alter the amount of myopia control effect. Specific terms, like spherical aberration and coma (horizontal and vertical), are linked to slowing myopia progression.^90–92 Spherical aberration may result from the eye taking on a more oblate shape, and coma is likely related to lens decentration.^93,94 Additionally, altering the orthokeratology lens design can induce greater HOAs for myopia control, specifically parameters like treatment zone size and compression factor.⁹⁵ 

Recently, a seemingly paradoxical theory emerged that high retinal contrast is a stimulus for eye growth.⁹⁶ This theory contends that intense retinal contrast leads to excessive stimulation of the retina, and reducing retinal contrast could provide a signal to slow eye growth and myopia progression (see Table 1). The hypothesized retinal contrast theory was based on the previously observed relationship among retinal cone expressions, genetics, and myopia.^97,98 Abnormalities in long wavelength-sensitive and middle wavelength-sensitive cone photoreceptors were found in people with a specific type of inherited high myopia, leading to reduced functional opsin and signals in damaged cones. It has been suggested that this reduction in signals in the affected cones could cause aberrant contrast signaling between adjacent cones, triggering axial elongation and myopia.^97,98 The contrast theory is supported by the positive results from a clinical trial in which the diffusion optics technology (DOT) lenses, designed to reduce retinal contrast by dispersing light, are partly effective in slowing axial elongation and myopia progression in children.^99,100 Other specialized myopia control spectacle lenses, which are presumed to work through peripheral retinal defocus signals, may also share the same common mechanism of reduced retinal contrast as a signal to slow eye growth.^101–105 The findings from a recent randomized controlled trial, which showed significantly slower myopia progression and axial elongation in children wearing spectacle lenses with either positive-power or negative-power lenslets, provide further support for the existence of a common mechanism.¹⁰⁵ 

Time Outdoors

Daylight, comprising direct, dispersed, and reflected sunshine, varies in irradiance and spectral power based on latitude, time, year, and environment.¹⁰⁶ Human evolution over millions of years occurred within natural day/night cycles, and the introduction of artificial light disrupts this cycle, potentially affecting not only health¹⁰⁷ but also ocular development.¹⁰⁸ The beneficial effect of time outdoors in preventing myopia may potentially be due to ambient light characteristics of an outdoor environment (e.g. intensity and spectral composition), which can work synergistically and are absent in artificial lighting.¹⁰⁹ Other potential mechanisms for the effects of time outdoors include circadian rhythms, reduced peripheral defocus, higher vitamin D levels, physical activity, near work, and a high spatial frequency environment (see Table 1). One or a combination of these mechanisms may be at play.¹¹⁰ 

Higher Illuminance Levels

Higher intensity of light is thought to prevent myopia development by increasing retinal dopamine levels.¹¹¹ In animal models, elevated light levels (approximately 15,000 lux) have been found to slow the rate of myopia development in response to form deprivation and minus lens wear,¹¹² with higher levels of dopamine metabolites in the retina.¹¹³ These findings suggest that higher light levels are protective against myopia development. However, because emmetropization is a beneficial adaptation phenomenon,¹¹⁴ higher light levels could also be simply compromising the efficiency of the emmetropization mechanism. 

In an outdoor environment, the illuminance levels can exceed 130,000 lux on a sunny day and measure 50,000 lux on a hazy sunny day; on an overcast day, it is approximately 15,000 lux.¹¹⁵ Although weather conditions, altitude, and latitude influence the level of outdoor illumination, these levels significantly depart from the usual indoor illumination levels, ranging from 50 to 500 lux.^115,116 Even in the shade from trees, outdoor light levels range from 5556 to 7876 lux, and wearing a hat (4112–8156 lux) still provides sufficient illumination, exceeding 1000 lux. Although sunglass wear reduces light levels to 1792 to 6800 lux, these levels are still 11 to 43 times brighter than illumination levels in indoor lighting environments.¹¹⁵ 

Direct evidence for the protective effects of higher illuminance levels against childhood myopia comes from cluster-randomized trials that used light sensors to objectively monitor light levels in schoolchildren. In the Shanghai Time Outside to Reduce Myopia (STORM) trial of 6295 Chinese students of ages 6 to 9 years, the 2-year adjusted incidence of myopia was 11% to 16% lower in the experimental groups of children who spent 40 to 80 minutes of additional time outdoors than the control group with habitual outdoor time.¹¹⁷ On average, the experimental groups were exposed to significantly higher light intensity (2984 ± 806 lux/minute) than the control group (106 ± 27 minutes/day) such that the cumulative light exposure (intensity × duration) of 600,000 to 750,000 lux was found to reduce the risk of myopia onset by 15% to 24%. In a follow-up study, the secondary analysis of the data showed that the beneficial effect of reducing myopia onset occurred only with continued outdoor exposure of at least 15 minutes to no less than 2000 lux intensity.¹¹⁸ In the Recess Outside Classroom Trial 711 (ROCT711) of 693 grade 1 Taiwanese schoolchildren, the intervention group of students who spent at least 11 hours per week outdoors (in addition to 200 scheduled recess hours) for 1 year showed significantly less myopic shift (by 0.12 diopter [D]), axial elongation (by 0.05 mm), and incident myopia (by 14.5%) than the control group.¹¹⁹ Although significantly less myopia shift was observed in children who spent 200 minutes of outdoor time weekly in 1000 lux or more, 3000 lux or more, and 10,000 lux or more light intensity environments, in the group spending 125 to 199 minutes of outdoor time weekly, only those exposed to bright light (≥10,000 lux) had significantly lower myopic shift, suggesting that shorter durations of outdoor time would require exposure to higher light levels to achieve a beneficial effect. Interestingly, this trial found myopia-protective effects of outdoor time in both myopic and non-myopic children, unlike the STORM trial which showed positive effects only in children without myopia.¹¹⁷ 

Circadian Rhythms

Daylight is a natural zeitgeber or time cue for synchronizing internal circadian rhythms due to its temporal intensity and spectral distribution changes.¹⁰⁶ Circadian rhythms in ocular structures have been observed,¹²⁰ with corresponding variations in ocular measurements throughout the day. Visual cues, such as optical defocus, significantly alter the phase relationship of these ocular components, including axial length and choroidal thickness.^121–124 Animal studies have found downregulation of circadian rhythm genes in the form-deprived eye compared to the fellow control eye, and acceleration of eye growth leading to myopia phenotype after the knockout of a critical circadian gene, indicating a potential role of circadian rhythms in eye growth regulation.^125,126 The potential eye growth-modulatory effects of circadian rhythms are possibly mediated by intrinsically photosensitive retinal ganglion cells through the activation of photopigment melanopsin.^127,128 Melanopsin sensitivity peaks between 460 and 500 nm, corresponding to blue light, whereas the visual system is most responsive to mid-wavelengths at 555 nm.^129,130 

Melatonin peaks during the night and decreases during the day and can be a marker for optimal regulation of circadian rhythms. Although administration of melatonin reduces induced myopia in animal models, findings on the association between melatonin and myopia in humans are mixed, with reports of both higher levels of melatonin in patients with myopia¹³¹ and no association.¹³² Myopia is also linked to shorter sleep duration,¹³³ although studies have shown mixed results.^134,135 

Understanding the dynamics among light, circadian rhythms, and myopia is essential for addressing the unique light-related needs of children and teenagers in educational settings.¹³⁶ Children, particularly prepubescent children,¹³⁷ are more sensitive to environmental light than adults, leading to a more noticeable decrease in melatonin levels. This decrease is twice as significant as in their adult parents.¹³⁸ This suppression of melatonin occurs even under typical home illumination conditions.¹³⁸ Due to the combination of larger pupils and more transparent ocular media, children may have a higher light transmission through the crystalline lens and greater retinal illumination, especially for shorter visual wavelengths.¹³⁹ This could result in a stronger signal to the suprachiasmatic nucleus in children than in adults,¹³⁹ affecting the regulation of circadian rhythms. 

Reduced Peripheral Defocus

In an outdoor environment, most objects are far away, which creates a uniform dioptric environment with little impact on peripheral retinal defocus. Unlike outdoors, a typical indoor environment contains objects at varying distances, resulting in high dioptric variation in the environment¹⁴⁰ and greater magnitudes of peripheral hyperopic defocus. Despite the dioptric differences between indoors and outdoors, whether environmentally induced peripheral hyperopic defocus is involved in the development or progression of myopia remains unclear. Large-scale longitudinal studies showed no association between peripheral hyperopia and myopia development.^65,67 However, a cross-sectional study of Hong Kong schoolchildren found that children with higher axial length to corneal radius ratio lived in smaller homes, suggesting that the size of living space was related with central and peripheral refraction profiles.¹⁴¹ Wearable devices capable of measuring object distances could aid in our understanding of defocus patterns in the visual environment, potentially providing valuable insights into the role of environmentally induced peripheral defocus in human myopia.¹⁴² 

Spectral Composition of Light

In vertebrate species, including humans, the inherent longitudinal chromatic aberration of the eye causes shorter wavelengths of light to focus closer to the cornea than the longer wavelengths of light.¹⁴³ Animal studies have provided compelling evidence that broad-band light is essential for normal eye growth and refractive development; in narrow-band light, animals cannot achieve or maintain emmetropia,^45,144,145 although exception exists.¹⁴⁶ It has been suggested that the emmetropization mechanism uses longitudinal chromatic aberration-derived wavelength cues to compare short-wavelength-sensitive and longer-wavelength-sensitive cone image contrasts to guide eye growth and refractive development.³⁴ An outdoor environment contains a relatively high proportion of shorter wavelengths than an indoor environment with tungsten and fluorescent lights. Differences in the spectral composition of indoor and outdoor environments may affect the input provided by wavelength cues to the emmetropization mechanism, thus potentially affecting eye growth and myopia development. 

Higher Vitamin D Levels

Vitamin D could be related to time outdoors, as exposure to ultraviolet radiation stimulates its production.¹⁴⁷ However, the connection between vitamin D and myopia appears inconsistent, possibly due to challenges in accurately measuring outdoor time or controlling confounding effects in cross-sectional studies. A recent meta-analysis reported an association of myopia with levels of serum 25-hydroxyvitamin D (25(OH)D), a marker of vitamin D levels, showing reduced levels of this marker in myopic individuals compared with non-myopes.¹⁴⁸ However, a prospective birth cohort study found no association between myopia and vitamin D marker levels after adjusting for time spent outdoors.¹⁴⁹ Moreover, a genetic study using Mendelian randomization and four genetic markers (single-nucleotide polymorphisms [SNPs]) found no link between a genetic predisposition to lower vitamin D levels and an increased risk of myopia.¹⁵⁰ The Mendelian randomization study, which minimizes the impact of outdoor time, may not have detected a threshold effect due to the small influence of genetic markers on vitamin D levels. Despite this, the strongest evidence from longitudinal and Mendelian randomization studies suggests that vitamin D is unlikely to be causally related to myopia.¹¹⁰ 

Increased Physical Activity

Studies have shown an association between physical activity and myopia, suggesting that physical activity may prevent myopia through biochemical changes.^151,152 However, these study findings are limited by the lack of adjustment for outdoor time. It has been shown that outdoor sports and activities reduce myopia risk, whereas indoor sports and activities do not.¹⁵³ Moreover, the protective effect of time outdoors occurs independently of physical activity levels,¹⁵⁴ arguing against the role of physical activity in protecting against myopia. 

Reduced Near Work

Myopia is often linked with activities involving close-up work and higher levels of education.^155,156 Although the exact mechanism through which visual experiences during extended hours of near tasks increase the risk of myopia in children is not fully understood, the role of accommodation may not be as strong,⁷⁷ as accommodation does not seem necessary for emmetropization.⁴⁵ Other mechanisms include optical and biomechanical factors, the timing of breaks,¹⁵⁷ and contrast polarity¹⁵⁸ and colors used in reading materials.¹⁵⁹ Hence, the association between outdoor time and myopia might be attributed to individuals engaging in less near work while outside.¹¹⁰ Evidence suggests that, in practice, time spent on near work and time spent outdoors are mostly unrelated. For instance, in Australian children aged 6 to 12 years, the correlation between time spent outdoors and near work was weak (r = 0.20 and r = 0.03, respectively).¹⁵³ In addition, the effects of time outdoors and near work on the likelihood of myopia incidence have been observed to be independent.^160,161 

High Spatial Frequency

The spatial frequency content of an outdoor environment has been proposed as a protective factor for myopia development.^140,162,163 Comparisons of outdoor and indoor scenes showed that outdoor scenes contain a greater amount of mid to high spatial frequencies.¹⁶² Bright environments with higher mid and/or high spatial frequencies can prevent myopic shifts in chicks.^164,165 Myopiagenic stimuli (occluders and minus lenses) used to induce myopia in animal models reduce contrast primarily at high spatial frequencies.^166,167 Temporal flicker generated by the movement of spatial frequency stimuli also seems to influence eye growth,¹⁶⁸ inducing myopia in low-frequency flicker environments compared with high-frequency ones. Further studies are needed to explore a causal link between spatial frequency and myopia in humans. 

Emerging and Experimental Treatments

7-Methylxanthine

7-methylxanthine (7-MX), a nonselective adenosine receptor antagonist and metabolite of caffeine and theobromine, is licensed in Denmark as a treatment for childhood myopia and has been in use since 2009.¹⁷² The potential clinical application of 7-MX as myopia treatment is based on the premise that 7-MX has a nontoxic safety profile,^173–175 and the oral administration of 7-MX appears to inhibit induced myopia in animal models and small cohort studies in children. However, findings from animal studies appear inconsistent. In guinea pigs, rabbits, and monkeys, experimental myopia paradigms (form-deprivation and minus lens treatment) produced significantly less myopia and axial eye growth in 7-MX treated animals than the controls, albeit with varying effects.^176–178 On the contrary, no effect was found in chicks.^179,180 The lack of effect was likely due to partly cartilaginous sclera in chicks, but oral administration of 7-MX also did not influence induced myopia in tree shrews,¹⁸¹ small primate-like mammals with fibrous sclera as in humans.¹⁸² 

In a placebo-controlled trial, Trier et al.¹⁷⁵ randomized 83 Danish children aged 8 to 13 years with an average axial growth rate of 0.075 to 0.39 mm per 6 months to receive a 400 mg 7-MX tablet or placebo for 1 year, following which children were allowed to continue (7-MX group) or start treatment (placebo group) with 7-MX tablets once or twice per day for another year. After 2 years, all participants discontinued treatment and had final measurements after a 1-year wash-out period. The authors evaluated the efficacy in two subgroups (moderate baseline axial growth = 0.075–0.10 mm/6 months, and high baseline axial growth = 0.200–0.390 mm/6 months) because the rate of axial elongation in the placebo group was found to be proportional to the baseline level. After 12 months of treatment, the average myopia progression and axial elongation in 7-MX-treated children was 0.40 D and 0.19 mm in the moderate axial growth group (n = 27) and 0.67 D and 0.35 mm in the high axial growth group (n = 16), respectively. The corresponding values for placebo-treated children were 0.51 D and 0.25 mm for the moderate axial growth group and 0.76 D and 0.38 mm for the high axial growth group. The treatment effects were small (0.11 D and 0.06 mm for the moderate-rate group and 0.09 D and 0.03 mm for the high-rate group) and not statistically significant. 

Unfortunately, the lack of a placebo group in the second year of the trial prevented the assessment of the true efficacy of 7-MX. Comparisons of 2-year axial elongation between children treated with 7-MX for 24 months and children treated with placebo for the first 12 months and then 7-MX for the next 12 months showed a small but statistically significant axial length control effect (0.01 mm) only in the moderate baseline axial growth group. Still, there was no corresponding effect on myopia progression, suggesting that any potential axial length control efficacy of oral 7-MX may be limited to children with an axial growth rate of 0.08 to 0.19 mm per 6 months. When the treatment was discontinued in the third year, no evidence of rebound effect was found, and myopia progression and axial elongation were similar to those observed in the previous period. Recently, the same group analyzed outcomes in children treated with varying doses of oral 7-MX and reported that treatment with oral 7-MX was associated with a reduced rate of myopia progression and axial elongation,¹⁷² suggesting that 7-MX could be a potential treatment option for childhood myopia. However, inconsistencies in interpretation and small effect sizes observed in the placebo-controlled clinical trial mean that the potential efficacy of 7-MX in slowing the progression of childhood myopia remains equivocal. 

The potential efficacy of 7-MX in myopia treatment is thought to be related to its ability to block adenosine receptors and affect scleral remodeling (see Table 2). However, the specific subtype and target tissue remains unknown. The adenosine isoforms ADORA1, ADORA2a, ADORA2b, and ADOR3 are expressed in several tissues, including the neural retina, retinal pigment epithelium, choroid, and sclera.^183–185 Studies in mice have shown that ADORA2a knockout produces a myopia phenotype and reduces scleral collagen fibril diameters, whereas induced myopia reduces ADORA1 expression and increases ADORA2b expression in the retina.^183,186 Myopic axial elongation is typically accompanied by thinning and a reduction in collagen content and fibril diameter in the posterior sclera.¹⁸⁷ Therefore, 7-MX could affect these changes and reduce the biomechanical alterations that make the sclera susceptible to extension and deformation. In animal models, oral administration of 7-MX was found to increase collagen-related amino acid content, collagen diameter, and overall thickness of the posterior sclera.^177,178,188 Cui et al. observed the presence of all known subtypes of adenosine receptors in the human scleral fibroblasts, lending weight to the scleral remodeling theory for the anti-myopia action of oral 7-MX.¹⁸⁹ However, it is also possible that 7-MX could be acting via non-adenosine receptors via off-target actions or by influencing other processes, such as dopamine or acetylcholine transmission^190,191 or circadian rhythms, implicated in eye growth regulation. The limited ability of 7-MX to cross the blood-barriers,¹⁹² however, supports a direct influence on growth signaling mechanisms of the retina, choroid, or sclera for myopia control.¹⁷⁶ 

Anti-Glaucoma Agents

Intraocular pressure is thought to modulate eye growth due to its tangential stretching force on the scleral walls.¹²⁵ During the development of myopia, the sclera undergoes thinning and creeps at a higher rate than the non-myopic eyes.^193,194 Therefore, the biomechanically weaker and thinner sclera of myopic eyes could be more vulnerable to the tangential stretching force of intraocular pressure, leading to accelerated eye growth. Consequently, it has been suggested that reducing intraocular pressure may slow eye growth and myopia progression by lowering the stretching force on the sclera walls.¹⁹⁵ 

Studies have linked elevated intraocular pressure and alterations in diurnal intraocular pressure rhythms with myopia,^196,197 although the Singapore Cohort Study of the Risk Factors for Myopia (SCORM) study found no correlation between intraocular pressure and refractive error or axial length in children.¹⁹⁸ Nonetheless, intraocular pressure-lowering drugs have received interest as a potential myopia treatment, largely due to positive findings from animal studies. These experiments found that treatment with alpha-adrenergic agonist¹⁹⁹ and prostaglandin F2-alpha analog^195,200 reduced induced myopia, although treatment with beta-blockers was ineffective.²⁰¹ 

In a randomized clinical trial, Jensen randomized children to single-vision spectacles, bifocal spectacles, and single-vision spectacles with twice daily use of a beta-blocker, 0.25% timolol maleate.²⁰² Despite the reduction of intraocular pressures in the treated group by 3 millimeters of mercury (mm Hg) on average, no difference was found in 2-year myopia progression between the control and treatment groups (1.14 vs. 1.18 D), suggesting that timolol was ineffective in slowing myopia in children, although a positive association between intraocular pressure and progression rates was observed in the control group. Further studies are necessary to determine if other intraocular-pressure agents can be an effective treatment strategy and translated to clinics. 

Pirenzepine

The search for a pharmaceutical agent that minimized the side effects of high concentrations (0.5%–1%) of atropine led to evaluation of pirenzepine, a selective M1 muscarinic receptor antagonist. In animal experiments, pirenzepine was found to reduce induced myopia and axial elongation across multiple species.^14,203–205 The efficacy of pirenzepine in treating myopia has since been demonstrated in randomized clinical trials of both Asian and non-Asian children. Tan et al. assigned 353 children aged 6 to 12 years to receive 2% pirenzepine gel twice daily (gel/gel) or daily (gel/vehicle) or vehicle twice daily (vehicle/vehicle).²⁰⁶ After 1 year of treatment, myopia progression in the gel/gel group was slower by −0.47 D than the vehicle/vehicle group and by −0.31 D in the gel/vehicle group. Only 29% of children who received pirenzepine twice daily showed a progression of at least −0.75 D compared with 57% of children who received vehicle only. Consistent results were found in a 2-year multicentric US trial where 174 children aged 8 to 12 years were randomized to receive twice daily application of 2% pirenzepine gel or placebo.²⁰⁷ Among the 145 participants who completed the 1-year follow-up, average myopia progression was over 50% slower in the treated group (−0.26 D) compared with the placebo group (−0.53 D).²⁰⁷ At the end of 2 years, the treated and control groups progressed by −0.58 D and −0.99 D in spherical equivalent refraction, respectively, producing a 2-year treatment effect of −0.41 D in myopia progression and 0.20 mm in axial elongation, although the latter was not significant. However, the study was limited by a significantly high dropout rate, with only 57.9% of participants continuing the study in the second year. 

Despite the impressive efficacy of pirenzepine in these trials, enthusiasm was limited due to safety concerns. The Asian study reported a high rate of numerous adverse events, many of which were more frequent in the treated group compared with the placebo group (e.g. frequency of papillae/follicles: 51%–59% in the treated group vs. 14% in the control group). Although the US study found a better safety profile, pirenzepine has since received little attention as a myopia treatment due to perhaps the twice-daily treatment regimen, relatively higher efficacy of low-concentration atropine eye drops, and the difficulty in sourcing the drug commercially.²⁰⁸ 

Caffeine

The efficacy of 7-MX, an adenosine receptor antagonist and a caffeine metabolite, in reducing induced myopia in mammalian animal models has led to an interest in investigating caffeine as a potential myopia treatment. In rhesus monkeys, topical administration of caffeine was shown to reduce lens-induced myopia and promote hyperopic shifts in refraction in both lens-treated and fellow control eyes.²⁰⁹ Although short-term experiments in adults showed that oral intake of caffeine increases electrical activity of the inner retina,²¹⁰ results from a recent randomized clinical trial of 96 Vietnamese children were disappointing, with no difference in myopia progression between the groups of children treated with 2% caffeine and those wearing single-vision spectacle lenses over a year (−0.70 vs. −0.76 D).²¹¹ 

Blue Light Stimulation of the Optic Nerve Head

Increasing evidence for the role of melanopsin-expressing intrinsically photosensitive retinal ganglion cells (ipRGCs) and the neurotransmitter dopamine in eye growth regulation and myopia development^111,212 has stimulated interest in enhancing retinal dopaminergic activity to increase dopamine levels as a potential treatment for myopia (see Table 2). In animal models, ablation of ipRGCs was found to induce myopia, whereas their activation resulted in hyperopic refractive shifts.²¹² Dopaminergic amacrine cells, which receive input from both rod and cone photoreceptors and ipRGCs, are the primary source of dopamine.²¹³ Because the axon pathway of ipRGCs involves the optic nerve head, selective stimulation of the blind spot with a blue light wavelength that corresponds to the peak sensitivity of melanopsin (approximately 480 nm) has been proposed as a method to stimulate the retrograde effect on dopaminergic amacrine cells and has been shown to increase retinal dopamine levels.²¹⁴ In a short-term human study, exposure of the optic nerve head to blue light for only 1 minute increased b-wave amplitudes in flash and pattern electroretinograms.²¹⁵ Surprisingly, the corresponding peak time responses after blue light stimulation were delayed, suggesting a potential impairment of retinal adaptive circuitry. In this study, the possibility of cone photoreceptor stimulation due to scattering of light from the blind spot cannot be ruled out, considering that exposure to ambient blue light by itself (that included wavelengths overlapping with melanopsin) appears to produce a reduction in human axial length after just 1 hour of exposure.²¹⁶ However, blue light has also been shown to disrupt the emmetropization process, producing highly variable responses and longer-term myopia in some animals.²¹⁷ Nevertheless, the findings from short-term studies suggest that blue light stimulation of the optic nerve head could be an effective myopia treatment. Results from the ongoing randomized clinical trial will likely provide more definitive answers soon.²¹⁸ 

Narrow-Band Long-Wavelength Light

Narrow-band long-wavelength light has emerged as a potential myopia treatment strategy. Studies in tree shrews (small dichromatic mammals close to primates) have shown that exposure to ambient narrow-band red light (peak wavelength = 636 ± 10 nm) slows eye growth and produces hyperopia,²¹⁹ as well as reduces myopia induced by minus lens wear.²⁸ The hyperopic effect of narrow-band red light occurs in infantile animals and juvenile and adolescent animals after they have completed the initial emmetropization process.²²⁰ Consistent with the results in tree shrews, narrow-band red light has been found to induce hyperopia and reduce myopia induced by form-deprivation or minus lens wear in trichromatic macaque monkeys.^29,221 Like other anti-myopia stimuli,^222–225 narrow-band red light appears to have a nonlinear effect, with brief periods (2 hours/day) of exposure producing significant effects.³¹ Recently, exposure to ambient amber light (wavelengths > 500 nm) has been shown to produce a robust hyperopia-inducing effect of similar potency to red light.³⁰ The hyperopic effect of long-wavelength light seems to occur even in low illuminance levels (50–100 lux) but may be affected by white light contamination.²⁷ 

The mechanism by which narrow-band long-wavelength lights produce hyperopia is unclear. Light-rearing experiments suggest that wavelength cues have substantial effects on emmetropization: animals emmetropize normally in broad-band light but deviate away from emmetropia when the spectrum is limited to narrow-band lights.^45,145 These emmetropization responses to wavelength cues are probably not caused by potential wavelength-induced alterations in circadian rhythms.²²⁶ As stated previously, in narrow-band long-wavelength light, the emmetropization mechanism could be using the sharper “longer-wavelength cone contrast” as a sign to slow eye growth, producing hyperopia.^33,34 When tree shrews were raised in small cages with chromatically simulated myopic defocus (blurring the blue channel), they developed significantly less proximity-induced myopia than those raised in the same cages without simulated defocus.²²⁷ However, the effect of long-wavelength light appears inconsistent across species. In species distant from humans with differing numbers of cones (chick), nocturnal habitat, and ultraviolet sensitivity (mice), or crepuscular activity patterns (guinea pigs), opposite results have been found, with long-wavelength lights producing myopia.^228–233 Despite inter-species differences in emmetropization responses to wavelength cues, findings from two diurnal, closest-to-human models with good vision and well-characterized emmetropization, suggest that the anti-myopia effects of narrow-band long-wavelength light are likely to translate to humans.¹⁴⁵ Recent clinical studies provide preliminary evidence that long-wavelength differentially affects neural activity of the inner retina²³⁴ and elicits axial length shortening in the short term.³⁵ Further research is necessary in this area to confirm the potential effects of narrow-band long-wavelength light in slowing myopia progression in children. 

Violet Light

Violet light has been recently proposed as a potential treatment strategy for childhood myopia.²³⁵ In chicks and mice, exposure to violet light (360–400 nm wavelengths) was found to suppress myopia induced by form-derivation or minus lens wear,^235,236 although study limitations warranted caution about the interpretation and potential significance of the findings.²³⁷ The underlying mechanism of the myopia-protective effect of violet light remains unclear, but a violet light-sensitive atypical opsin called OPN5 or neuropsin present in the retinal ganglion cells is thought to play a role. This hypothesis is supported by recent findings that the myopia-protective effect of violet light is dependent on time of day and OPN5 expression levels.²³⁸ 

Despite the encouraging results in animal studies, the translation potential of violet light as a myopia treatment appears low. Unlike chickens and mice, humans lack an ultraviolet receptor and vision,²³⁹ and the human crystalline lens effectively blocks wavelengths below 400 nm.²⁴⁰ In a retrospective study, children wearing violet light-transmitting soft contact lenses had a 0.08 mm reduction in axial elongation compared with children wearing violet light-blocking eyeglasses and a 0.05 mm reduction than children wearing partial violet light-blocking soft contact lenses.²³⁵ However, a subsequent 2-year randomized clinical trial found no difference in myopia progression or axial elongation between groups of children wearing violet light-transmitting or conventional eyeglasses (−1.53 vs. −1.42 D and 0.03 vs. 0.11 mm, respectively).²⁴¹ These findings suggested a lack of efficacy of violet light in slowing myopia progression in children. However, post hoc data analysis indicated some benefits in children engaged in less than 180 minutes of near work or those who had never worn eyeglasses. Further studies are needed to evaluate these potential benefits. 

Ring-Focus Contact Lens

The recent development of soft contact lenses with novel non-co-axial ring focus design (Johnson and Johnson Vision Inc.) has provided a potential alternative optical treatment strategy for myopia. These lenses were designed to enhance myopia treatment efficacy while mitigating the trade-off between efficacy and vision of peripheral plus power lens designs, which improve the efficacy with higher adds⁵² but carry the potential to significantly affect the vision due to an increase in optical aberrations. The ring focus design has concentric zones of increased plus power compared to the power required for refractive correction at a distance like other myopia treatment contact lenses, but the peripheral plus power zones have higher relative power (+7 D) and do not create a co-axial point focus, rather form a ring focus in front of the retina. Additionally, the ring focus design has a central +10 D coaxial treatment zone to increase the treatment efficacy. 

The safety and efficacy of the novel ring focus lens design were tested in a multisite, prospective, randomized, controlled, double-masked, four-arm, parallel-group clinical trial conducted across different sites in the United States, Canada, and China.²⁴² In this trial, 199 children were randomized to wear a prototype lens designed to enhance efficacy (EE), another prototype lens designed to enhance vision (EV; similar to enhanced efficacy design but with treatment zones closer to the lens center and no +10 D co-axial treatment zone), a dual focus lens with +2.5 D co-axial peripheral plus power, or a single-vision contact lens. After 6 months of treatment, the average axial elongation was 0.08 mm in the EE lens group, 0.12 mm in the EV lens group, 0.14 mm in the dual-focus lens group, and 0.20 mm in the single-vision lens group. The corresponding values for myopia progression were −0.12 D, −0.26 D, −0.25 D, and −0.35 D, respectively. Compared with the spectacle lens group, the EE and EV lenses showed axial elongation by −0.11 mm and −0.06 mm, respectively, but only the EE lens had a statistically significant effect on myopia progression, slowing it by 0.22 D over 6 months. Moreover, a greater slowing of axial elongation was observed with the EE lens than with the dual-focus lens, indicating a superior efficacy. Visual performance was similar between the EV and single-vision lenses. Although the EE and dual-focus lenses produced worse distance visual acuity up to 0.07 log MAR than the single-vision lenses, the reduction was not statistically significant at any time point. The results from this trial indicate that the ring-focus design contact lenses are safe and effective for the treatment of myopia in children. These lenses have received breakthrough device designation from the US Food and Drug Administration (FDA) and approval from Health Canada for myopia management. 

Surgical Treatments

Myopia is caused by the expansion of the sclera wall, resulting in scleral thinning and the formation of posterior staphyloma in some extreme cases. Animal studies have shown that, despite having a similar elasticity, the viscoelastic properties of myopic sclera are considerably different than non-myopic sclera, causing them to creep (pressure-induced deformation) over time.^194,243 Surgical treatments for myopia are based on the premise that stabilizing the sclera may reduce posterior scleral deformation, which could slow axial elongation and myopia progression and reduce the risks of myopia-related complications.³ 

Surgical approaches for stabilizing the sclera and reducing scleral expansion include scleral reinforcement surgeries, injection-based scleral strengthening, and collagen cross-linking scleral strengthening. Posterior scleral reinforcement has a long history of use in myopia²⁴⁴ and is currently implemented in several countries, including China and Russia, in adults and children. Over the years, studies reporting the safety and efficacy of this surgical treatment in slowing myopia progression and axial elongation have generally shown some benefits.³ More recent studies report the efficacy of posterior sclera reinforcement in slowing axial elongation in children and adults with high myopia, albeit with widely varying treatment effects, from 0.18 mm to 1.05 mm.^245,246 Although progress has been made in posterior sclera reinforcement technique with the use of donor sclera tissue²⁴⁷ and specially designed polymeric devices, particularly in high myopia,²⁴⁸ significant drawbacks, such as invasiveness of the procedure, required use of general anesthesia, and reliance on the availability of health donor sclera tissue, limited their application and advancement.²⁴⁹ 

Injection-based scleral strengthening involves the application of chemical reagents via injection under Tenon's capsule to stabilize the scleral extracellular matrix. There is a paucity of evidence regarding the effectiveness of this strategy in treating myopia, with no data from randomized clinical trials. However, limited available data, mostly from case series and case-control studies, suggests that injection-based scleral strengthening may offer reasonable success by stabilizing myopia in a large proportion of treated participants.^250,251 In these studies, the frequency of cases showing myopia stabilization was found to decrease after 4 to 9 years of treatment,²⁵¹ posing questions about the long-term effectiveness of injection-based scleral strengthening as a treatment for myopia. 

Observation of an association of scleral creep and reduction in intermolecular collagen crosslinks in animal studies¹⁹³ suggest that techniques to improve biomechanical stiffness and viscoelastic properties of the sclera by increasing crosslinking between collagen fibers may make it less vulnerable to deformation, and thereby, reduce the likelihood of myopia progression and myopia-related complications. Consequently, interest is growing in identifying new techniques or translating techniques previously used successfully in corneal collagen crosslinking,²⁵² such as the combined use of riboflavin and ultraviolet A radiation, to scleral collagen crosslinking. Although rare, complications have been reported with corneal collagen crosslinking, requiring penetrating keratoplasty to resolve the condition. However, resolving a failed scleral collagen crosslinking is complicated and risky. Moreover, the use of riboflavin and ultraviolet A radiation has the potential to damage cellular components in tissue²⁵³ and may result in the loss of photoreceptors and retinal pigment epithelium.²⁵⁴ These cytotoxic risks, along with the need to expose the sclera to implement these techniques, make their use in scleral collagen crosslinking challenging. Regardless of cellular toxicity, collagen crosslinking techniques are also likely to crosslink collagen in blood vessels and capillaries and produce permanent alterations in tissue biomechanics, including potential effects on scleral myofibroblasts. Therefore, challenges exist in implementing collagen crosslinking an opaque and vascular tissue such as the sclera.²⁴⁹ Although alternative techniques like chemical crosslinking using glyceraldehyde have been suggested as being potentially effective and safe,²⁴⁹ evidence regarding the optimal method of chemical crosslinking, ocular effects, and long-term effectiveness is significantly lacking. Recent data suggest that scleral crosslinking with low cytotoxic agents like genipin can reduce the cyclic softening response in tree shrews with induced myopia.²⁵⁵ However, experimental application of scleral collagen crosslinking for myopia treatment is currently limited to animal models. 

Artificial Lights

Whether time outdoors in daylight is protective against myopia development due to characteristics of the outdoor environment or simply not spending less time indoors or on screens exposed to artificial lighting is unclear. The impact of artificial light at night on babies and infants has long attracted attention. Night lighting is associated with myopia.²⁶¹ Gwiazda et al. reported that myopic parents were more likely to use night-time lighting aids for their children.²⁶² In contrast, Zadnik et al.²⁶³ and Jiang et al.²⁶⁴ found no significant link between night-time lighting and myopia development in children. 

In recent years, the planned withdrawal of incandescent lamps (approximately 9% blue) from the lighting market has led to a sharp increase in LED lighting systems. The LED technology is the most efficient way to produce light; however, the ANSES report proposed the additional blue light (up to 47% blue light) in the LED spectrum raises concerns about the harmful effects LEDs may have on children,²⁶⁵ as their underdeveloped optical system allows blue light to reach the retina.²⁶⁶ The authors concluded that retinal exposure to artificial sources is far from negligible: for a 14-year-old, the phototoxic dose is close to 17% of the natural phototoxic dose in winter and close to 8% over the entire year.²⁶⁵ The European Union Scientific Committee on Health, Environmental and Emerging Risks (SCHEER) and the European Environmental Agency published reports on the potential risks associated with LED systems on human health, particularly those relating to blue light.²⁶⁷ 

Several studies have explored the relationship between LED lighting and myopia. Pan et al.²⁶⁸ found that using LED lamps for homework was associated with a higher prevalence of myopia in Chinese schoolchildren. Conversely, animal studies have shown that exposure to narrow-band, long-wavelength LED lighting produced hyperopia and reduced lens-induced myopia in infant rhesus monkeys and tree shrews.^28–30,220 Recent reports have highlighted the potential of light exposure, including LED, in preventing myopia, with some recommending the use of light therapies.^109,269 Recently, repeated twice-daily exposure to long-wavelength red light has been proposed as a potential treatment strategy for myopia. However, the devices have generally used semiconductor laser diodes to administer light therapy.²⁷⁰ 

Could exposure to LEDs affect eye health? In a narrative review,²⁶⁶ Cougnard-Gregoire et al. found no evidence for the detrimental effects of screen use and LEDs in everyday use on the human retina. Although blue light was found to cause photochemical reactions in ocular tissues and temporary or permanent damage to the retina in in vitro or in vivo studies, there was no evidence of deleterious effects on the human retina, which was attributed to the natural protective mechanism offered by macular pigments capable of filtering blue light. The authors also reported no evidence for the beneficial impact of blue-blocking lenses in preventing eye diseases, but this finding is inconsistent with two other studies that have found positive effects. Results from these studies showed that, compared to the pseudophakic eyes with non-blue filter intraocular lenses, the progression of geographic atrophy was significantly slower,²⁷¹ and the development of abnormal fundus autofluorescence and incidence of age-related macular degeneration were significantly lower in eyes with blue light-filtering intraocular lenses.²⁷² Further research seems necessary to conclusively establish potential ocular health benefits of specific protective measures against blue light. 

Myopes may be more susceptible to LED lighting. For instance, Rao et al.²⁷³ examined marker location task failures on a computer screen under nine LED lighting conditions: three different illuminances (100, 300, and 500 lux) and three different color temperatures (3000, 4500, and 6500 K). Morning task mistakes were minimal and consistent across the nine lighting conditions. In the afternoon, task errors were considerably more significant and varied by lighting. Color temperature 4500 K and illuminance 500 lux caused most afternoon errors. Task errors differed significantly between morning and afternoon sessions, and no significant sex difference was found. Under the same lighting, high-myopia students made considerably more task errors than low-myopia students. LED lighting did not affect office task accuracy by gender, but it did by the time of day and myopia.²⁷³ 

Hessling et al. assessed the impact of LED lighting on human health, compared the health risks associated with LED lighting with those from long-established light sources, and discussed the potential risks to children from LED lighting.²⁷⁴ The primary outcome measured in the study was the relative blue light retinal endangerment and relative melatonin inhibition of different light sources, including the iPad Mini 2 in day mode. The potential risk to children due to the transmission of short-wave spectral range light to the retina was also discussed. The authors concluded that although LED illuminants may pose a new health risk, they are comparable to fluorescent tubes, which have been used for decades and have evolved. Warm white LEDs offer better blue light hazard and melatonin inhibition than halogen light sources. 

As handheld digital devices are now integral to childhood (in educational and recreational settings), chronic exposure (enhanced by LED illumination) over a lifespan may have a cumulative phototoxic effect on the retina.²⁶⁶ Recognized adverse health effects linked to LEDs come from factors like glare, disturbance of circadian rhythms, and flicker. For more details, please refer to the report “Light-Emitting Diodes (LEDS): Implications for Safety.”²⁷⁵ Glare is an indirect hazard resulting from surrounding interference, especially in poorly designed LED fixtures. This can cause discomfort glare, where the light source is much brighter than nearby objects, and disability glare, where intense brightness scatters within the eye. 

Circadian rhythm is regulated by light on a 24-hour cycle. Exposure to intense “cool white”/blue-rich radiation may disrupt circadian rhythm, with the effect mainly occurring during evening or night-time exposure.²⁷⁵ For circadian regulation, the eye is most sensitive to the blue wavelength, emitted strongly by white LEDs²⁷⁶ because this non-visual effect requires blue-light-sensitive melanopsin retinal ganglion cells (spectral sensitivity = approximately 480 nm). Synchronizing the central circadian clock requires bright blue light during the day and complete darkness at night.²⁷⁶ The growing use of LED systems has raised concerns about blue light's effect on the circadian rhythm, which is crucial for a healthy lifestyle, as disruption in the circadian cycle has been linked with sleep disturbances and emotional issues.²⁷⁷ Modern lifestyles interrupt the natural light/dark cycle, especially in urban environments. Artificial blue light at night and a lack of bright outside light in the morning may affect circadian and sleep rhythms and, consequently, eye growth and myopia development.²⁷⁸ 

Flicker is a common issue in LEDs powered directly by electricity. Most A/C-driven LEDs show flicker effects. Animal studies indicate that exposure to flickering narrow-band blue light (464 nm wavelength) emitted by LEDs can lead to elongation of the vitreous chamber and myopia.²¹⁷ Two other critical considerations in lighting design for myopia prevention involve window materials and electric illumination. First, window materials play a crucial role, with options like low-e coatings altering the spectral distribution of transmitted light, reflective coatings and tints varying based on color and chemistry, and dynamic glazing systems, such as electrochromic glazing, capable of changing the spectral properties, timing, and relative intensity of daylight illumination.²⁷⁹ Electric illumination often lacks the shortest visible wavelengths (350 to 480 nm). To address this, the spectral power distributions of typical LEDs may need adjustment. 

Typically, indoor lighting levels fall within the range of 10 to 1000 lux, whereas outdoor light levels can vary between 10,000 and 30,000 lux on a cloudy day or under shade.¹¹⁵ On a sunny day, outdoor light levels can exceed 100,000 lux. Increased exposure to natural light is associated with positive social-emotional, cognitive, and physical health outcomes for children,¹³⁶ including a reduction in the prevalence and incidence of myopia.²⁸⁰ On the contrary, artificial lighting exposure affects refractive error development and circadian dysregulation, which may further exacerbate myopia.²⁶⁹ 

The potential risk to children from increased exposure to short-wavelength light, which includes more blue, violet, and ultraviolet light reaching the retina, is unclear. This raises concerns about possible harm to the eyes from artificial light sources. To prevent myopia, cautioning against excessive shielding from daylight may lead to overexposure to UV rays. There is a critical need to understand not only the current levels of light exposure in children but also the quantity and quality of children's light exposure, including exposure to various lighting sources. It seems crucial to establish the long-term effects of artificial light and the impact of natural daylight on preventing myopia in children. To accurately understand child light exposure, objective measurements are necessary. Profiling the existing exposure of children across different ages and demographics may help create evidence-based public health messages for healthy eye development. Although moving more of the school day outdoors is a practical idea, it requires changes to educational infrastructure, public support, and political will, making it a challenging solution.²⁸¹ Taking this into consideration, adjustments may be needed in building design and school facilities. 

Treatment Limitations

Myopia treatments are not without limitations, but the benefits of treating myopia are substantial. Bullimore et al. studied the prevalence of myopic maculopathy and cumulative risks of visual impairment as a function of myopia levels and found that slowing myopia by 1 D can reduce the risk of myopic maculopathy by 37% and visual impairment by 20% and can save 0.5 to 1 year of visual impairment.²⁸² When the final level of myopia was plotted as a function of age, delaying the onset of myopia by 1 year was predicted to reduce myopia by 0.75 D or more, particularly in East Asians.²⁸³ The risk-benefit model by Bullimore et al. suggested that only 4 to 7 individuals need to be treated to prevent 5 years of visual impairment, but less than 1 in 38 would experience vision loss with myopia treatments.²⁸² For the implementation of myopia treatments, the benefits appear to outweigh the risks considerably. 

While treatments such as orthokeratology, dual-focus, and multifocal contact lenses, RLRL therapy, atropine, and myopia control spectacles are effective in slowing the progression of myopia, considerable variation in effectiveness exists at individual levels, suggesting that they may not be equally effective for everyone. Different individuals may respond differently to each treatment, and therefore, it may be challenging to decide which treatment option to use, how closely to monitor, or when to switch treatments. Some children are “fast progressors” and continue to show myopia progression despite being on treatment²⁸⁴; however, they may not necessarily be nonresponders but progressing at a slightly higher rate than the average progression rate and at a lower rate than that would be without the treatment.²⁸⁵ These children may benefit from combination therapy with multiple treatments, such as atropine and orthokeratology^286–291 or atropine and specialized spectacles,²⁹² although the combination therapy of atropine and bifocal contact lenses appears to offer no added benefit.²⁹³ The cost of myopia treatments varies depending on several factors, including treatment modality, geographic area, and length of time on treatment. Treatment costs can restrict access to care.²⁹⁴ 

One of the unintended consequences of myopia treatments is the effect they have on foveal vision, clinically measured as best-corrected visual acuity. Both orthokeratology and peripheral defocus soft contact lenses are designed such that the center portion of the lens corrects myopia, and the lens periphery induces myopic defocus, which can influence the quality of vision or best-corrected visual acuity. Often, high-contrast visual acuity is not affected, but low-contrast visual acuity can be up to 0.12 logMAR worse after orthokeratology lens wear compared to single-vision spectacles.⁸⁵ Contrast sensitivity also decreases after treatment with orthokeratology lenses.⁹³ However, low concentrations of atropine, less than 0.1%, have minimal impact on best-corrected visual acuity. The LAMP study found a 0.02 logMAR reduction in distance visual acuity and a 0.03 logMAR reduction in near visual acuity in both 0.05% atropine and placebo groups.²⁹⁵ 

In general, spectacles pose minimal physical risks associated with wear; however, myopia-control spectacles have been shown to affect peripheral vision.²⁹⁶ The highly aspherical lenslets myopia control spectacle lenses displayed a slight reduction in low contrast acuity when viewed through the periphery of the lens and affected the reading performance of low contrast words.²⁹⁷ Defocus-incorporated soft contact lenses (DISC) demonstrated worse central contrast sensitivity compared with single-vision spectacles and HALT lenses.²⁹⁸ Papadogiannis et al.²⁹⁹ found that the highly aspheric lenslet target (HALT) and DIMS lenses had a small effect on relative peripheral refraction but a greater reduction in retinal contrast. This reduction in peripheral retinal contrast has been postulated to contribute to the slowing of myopia progression. 

Treatment with orthokeratology lenses is affected by diurnal variation based on when the lenses were removed and how long they were worn. Ideal correction is achieved by wearing the lenses for at least 8 hours every night.²⁹⁴ With peripheral defocus soft contact lenses, the stimulus for treatment is only present during lens wear, so it is encouraged that the lens be worn during school and at home for schoolwork at a minimum,³⁰⁰ as full-time wear seems to improve the efficacy of some optical interventions.^300,301 Children appear to tolerate wearing myopia treatment lenses full time without any issues. On average, the wearing time for peripheral defocus soft lenses in clinical trials ranged from 12 to 14 hours per day.³⁰² 

Although dual-focus soft contact lens is the only FDA-approved myopia treatment, this modality has some limitations. A potential adverse event with contact lens wear is dryness or allergic conjunctivitis.³⁰³ Allergies are common in pediatric patients and can be exacerbated by contact lens wear.³⁰⁴ Additionally, the amount of myopia and astigmatism that each lens design can correct varies, and some patients may be limited in their treatment options to certain lens modalities, materials, or replacement schedules. Soft contact lenses can also increase the likelihood of corneal staining, abrasions, infiltrative events, or microbial keratitis, although the risk appears low.³⁰⁵ The rate of corneal infiltrative events with soft contact lenses was lower in 8- to 12-year-olds (97 per 10,000 patient-years) compared to 13- to 17-year-olds (335 per 10,000 patient-years), and the same was true for incidences of microbial keratitis (0 vs. 15 per 10,000 patient-years).³⁰⁶ Moreover, dual-focus soft contact lens wear for 6 years resulted in no contact lens-related adverse events and no significant changes in biomicroscopy, suggesting that soft contact lenses have minimal impact on the ocular health of children over the long term. Compared with adults, the incidence of corneal infiltrative events associated with soft contact lens wear in children is substantially lower (136 vs. 318 to 720 per 10,000 patient-years),³⁰⁵ likely reflecting differences in patient behavior, treatment adherence, and parental supervision between children and adults wearing soft contact lenses. Children must be able to comply with the prescribed treatment and adhere to recommended hygiene habits to minimize the risk of infection. 

However, the reported incidence of microbial keratitis with orthokeratology lens wear is higher in children compared with adults (13.9 vs 0 per 10,000 patient-years), suggesting an increased risk.³⁰⁷ Risks associated with spectacle lens wear appear minimal. In one study, the incidence of adverse events in children was less with spectacle lens wear than with daily disposable hydrogel lens wear (1.8 vs. 4.5 per 100 patient-years).³⁰⁸ Lam et al. explored the safety of the DIMS spectacle lenses and showed that long-term wear did not result in any adverse events.³⁰⁹ Soft lenses move with eye movements, whereas spectacle lenses do not, so the patient experiences distinct parts of the lens in different positions of gaze.²⁹⁷ Dual-focus soft contact lenses and orthokeratology do not significantly affect high-contrast visual acuity, but low-contrast visual acuity measured with dilated pupils or at low brightness to induce larger pupils reveals minor but potentially significant impairments.^85,310,311 Generally, there are no uniform guidelines for reporting adverse events, and many studies do not include this information, making it challenging for clinicians to know the true incidence in pediatric patients.³¹² 

Treatment with atropine eye drops carries potential ocular and systemic side effects. Although high concentrations of atropine (0.5% and 1%) are highly effective in slowing myopia progression, there is pronounced photosensitivity due to pupil dilation and near blur due to lack of accommodation, resulting in poor treatment compliance and high dropout.^313,314 While these effects can be mitigated and treatment compliance can be improved with photochromatic and progressive addition lenses (e.g. 13.5% dropout in ATOM1), strong rebound effects occur upon discontinuation of treatment.^315,316 These side effects and rebound effects of atropine are markedly reduced at lower concentrations (0.01%–0.1%).^316,317 The frequency of photochromatic glass use in ATOM studies was significantly lower for 0.01% (7%) than 0.1% and 0.5% (60%–70%). ATOM2 study also found a significant dose-related reduction of effect on accommodative amplitude, near visual acuity, and pupil size.³¹⁸ Accommodation was only slightly reduced to 11.3 D with 0.01% atropine but more significantly reduced with 0.1% and 0.5% to 3.8 D and 2.2 D, respectively. Both photopic and mesopic pupil size were increased after atropine use by approximately 3 mm with 0.1% and 0.5% concentrations, but the increment in pupil size was markedly reduced (1 mm) with 0.01%. 

Consistent results were found in the LAMP study,²⁹⁵ in which low concentrations of atropine had minimal effects on accommodative amplitude (2 D reduction with 0.05%), pupil size (1 mm reduction with 0.05%), and decreases in the distance and near visual acuity (0.02–0.03 logMAR) after 1 year of treatment. The frequency of use of photochromatic glasses was similar between the low-concentration atropine and placebo groups (30%–34% vs. 40%). These results suggest that low-concentration atropine eye drops produce markedly reduced ocular side effects, although allergic reactions were common. Possible systemic side effects of atropine include dry skin, mouth, and throat, drowsiness, restlessness, irritability, delirium, tachycardia, and flushing of the face or neck,^3,319 and some children may develop hypersensitivity.³²⁰ These systemic effects are unlikely to occur with low concentrations, although there is a published case report of 0.05% atropine-induced systemic hypertension that resolved with discontinuation.³²¹ The optimal concentration of atropine eye drops for myopia control remains controversial.^322–324 

Although RLRL treatment has been shown to be effective in slowing myopia progression, significant concerns remain over safety, long-term efficacy, and potential benefits in children other than those of Chinese descent, as highlighted by multiple groups in recent commentaries.^26,171 There is at least one published case report describing retinal damage secondary to RLRL treatment. After 5 months of RLRL treatment, a 12-year-old female patient presented with macular damage leading to vision loss in both eyes and exhibited bilaterally darkened fovea with hypoautofluorescent plague, disruption of foveal ellipsoid zone, discontinuity of interdigitation zone, and moderate reduction in retinal response at the macular region, all of which appeared to recover to normal levels following treatment discontinuation.¹⁷⁰ Recent evaluation of RLRL treatment devices has also noted that these devices could reach or exceed maximum permissible exposure limits after 3 minutes of treatment and may potentially cause photochemical and thermal damage to the retina.¹⁶⁹ Moreover, the rebound effect of RLRL treatments appear strong^325,326 and significantly greater than pharmacological and optical myopia control treatments,³¹⁶ raising questions about its long-term treatment efficacy. 

Treatment Trajectories

Myopia is generally diagnosed by the age of 10 years, but the onset could be as early as 3 to 4 years or as late as teenage or early adulthood.^327–329 Progression occurs rapidly in the early years, generally ages 7 to 12 years.^330,331 Early age of onset of myopia appears strongly associated with a higher final level of myopia,^332,333 so treatment should start as early as possible³³⁴ to reduce myopia-associated risks.²⁸² Some studies have looked at initiating treatment in children who have not yet developed myopia but are at risk due to less optimal age-appropriate refractive state and other risk factors.^335,336 The incidence of myopia was significantly lower in non-myopic children aged 4 to 9 years after a 2-year treatment with 0.05% atropine relative to placebo.³³⁷ However, starting treatment in children without myopia remains challenging. The CLEERE study showed that a hyperopic buffer was needed to prevent a child from developing myopia. Children with refractive errors < +0.75 at age 6 years, < +0.50 at age 7 to 8 years, < +0.25 at age 9 to 10 years, and less than plano at age 11 years were likely going to develop myopia.³³⁶ In addition, axial length was found to increase in the 2 to 3 years preceding myopia onset.⁷¹ Observing refractive status and changes in axial length may help practitioners predict if and when a child will develop myopia. 

Once a child becomes myopic, treatment can and should be initiated. To estimate if a treatment is effective, comparisons can be made to estimates of natural myopia progression. Normative growth curves are one way to assess progression or treatment efficacy in an individual child. However, these are average values and may not always factor in other variables, such as age and ethnicity, that can impact progression. The comparison of axial length percentile curves generated from European children³³⁸ and those from Chinese children³³⁹ shows that children of Chinese ethnicity have longer axial lengths, perhaps representative of a higher prevalence of myopia in that population. These differences emphasize the need to use appropriate charts for comparison, as using a European growth curve for a Chinese child would overestimate myopia progression.³³⁹ 

Myopia treatments may offer the greatest benefits when initiated at a younger age when progression is fastest. Treatment response in younger children may sometimes be regarded as poor,³⁴⁰ but this may reflect the underlying faster progression. Regardless, younger children may need more aggressive treatment. Treatment efficacy is greatest in the first 6 months to 1 year of treatment but is significantly reduced thereafter.^{80,99,284,315,341–343} This makes the assessment of long-term treatment efficacy challenging, as most placebo-controlled myopia treatment studies are limited to 1 to 2 years duration.³⁴⁴ 

In the evaluation of the dual-focus contact lens, comparisons were made between a group that was treated for 6 years and a group that received a placebo for the first 3 years and then switched to treatment for another 3 years. The mean difference in myopia progression between the first 3 years and the second 3 years for the continued treatment group was 0.01 D. In comparison, the group that had placebo for the first 3 years and treatment for the second 3 years showed a mean difference of 0.98 D. This finding demonstrates that treatment efficacy with dual-focus soft contact lens remains relatively sustained over 6 years of treatment.³⁴⁵ The 6-year DIMS study observed a similar sustained treatment effect.³⁰⁹ 

Evidence regarding the appropriate duration of myopia treatments is lacking. The Correction of Myopia Evaluation Trial (COMET) found that the mean age of stabilization of myopia (progression ≤ 0.50 D) was 15.6 years (COMET group, 2013). However, myopia continued to progress into mid-teens and early adulthood in many individuals, with the proportion of participants with myopia progression of 50% at 15 years of age, 25% at 18 years of age, 10% at 21 years of age, and 5% by 24 years of age (COMET group, 2013). The growing evidence of myopia onset and continued progression in young adults^346,347 suggests a continuation of treatment for as long as feasible. Practitioners should closely monitor patients when deciding to discontinue treatment, as any increase in progression will warrant reinstating the treatment.²⁹⁴ Additionally, children who do not seem to respond to myopia control treatments (sometimes called “nonresponders”) and those who exhibit excessive growth (“fast progressors”) may be hard to differentiate.²⁸⁵ Despite the lack of evidence that switching treatment modalities is an effective approach, children deemed “nonresponders” or “fast progressors” may require an alternative treatment strategy. A rebound effect has been shown to occur with atropine eye drops as low as 0.01% concentration,^{284,317,348–350} repeated red light therapy,^325,326 and orthokeratology lenses,^351,352 but myopia progression does not seem to accelerate after discontinuation of soft contact lenses^341,353,354 or specialized spectacle lenses.^309,355,356 It is pertinent to note that proper evaluation of rebound effects requires a comparison of progression between a previously treated group and a concurrent control group during the same period after stopping the treatment³¹⁶ rather than a comparison of progression between the treatment and washout periods in the same group as is the case in recent systematic reviews.^357,358 

Cost-Effectiveness of Treatments

Economic assessments of healthcare interventions aid evidence-based advocacy, policymaking, and patient care.³⁵⁹ Myopia is growing as a substantial economic burden. In 2015, uncorrected myopia cost US $244 billion, and myopic macular degeneration accounted for US $6 billion in productivity.³⁶⁰ In Singapore, the annual direct cost of treating myopia for teens was SGD $25 million³⁶¹ and SGD $755 million for adults.³⁶² These costs include refractive surgery, glasses, contact lenses, solutions, and associated myopia visual impairment issues. Due to the increasing prevalence of myopia and the strain on healthcare resources, a reliable economic evaluation of myopia treatments is necessary to maximize benefits. Data on the economic benefits of myopia treatments are sparse. Lack of economic evaluation evidence may hamper myopia intervention uptake.³⁶³ 

Myopia treatment costs vary significantly worldwide, but direct costs are the major contributor. In the United States, the annual direct costs range from US $14 to US $26 per capita, with contact lenses being the most expensive option.³⁶⁴ In Singapore, the mean annual direct cost of myopia is approximately SGD $900 (US $709) per person. The major drivers of this cost are spectacles, contact lenses, and optometry services.³⁶² For Singaporean schoolchildren, the mean annual direct cost of myopia is SGD $221.68 (US $148), with higher costs associated with higher family income and parental education levels.³⁶¹ Myopia treatment costs are a significant financial burden for patients and their families, and effective control methods are needed to alleviate these costs.³⁶⁵ Myopia treatment is now an evidence-based standard of care globally.^{294,366–369} The 2023 Myopia Consensus Statement released by The World Society of Pediatric Ophthalmology and Strabismus states, “There is sufficient evidence to warrant the adoption of myopia prevention and control measures in clinical practice in children with progressive myopia of childhood.”³⁷⁰ Despite the increasing recognition of myopia as a high-priority problem and the establishment of a task force by the American Academy of Ophthalmology,³⁷¹ there are no formalized insurance or vision plan services that cover the costs of myopia treatments in the United States. 

The cost-effectiveness of various myopia treatment options varies. Previously, Gwiazda and Chang pointed out limitations in available treatments, including short-term benefits and side effects.^372,373 However, in 2021, Foo et al. suggested that myopia treatment costs are justified, considering the high direct costs of myopia correction.³⁶⁴ Recently, discussions of the cost-effectiveness of myopia treatments have gained increasing interest.^374,375 

One systematic review revealed that the annual direct expenditures for contact lenses ranged from $198.30 to $378.10, whereas the costs for spectacles and refractive procedures were $342.50 and $19.10, respectively.³⁶⁴ The annual prevalence-based direct expenditures for myopia were between $14 and $26 per capita in the United States, $56 per capita in Iran, and $199 per capita in Singapore. The respective populations were 274.63 million, 75.15 million, and 3.79 million. In New Zealand, Hong et al. examined the cost-effectiveness of photorefractive myopia screening at age 11 years, administering atropine 0.01% eye drops for positive cases.³⁷⁶ The authors found that screening plus atropine eye drops saved 7 lifelong blindness cases per 100,000 children and had an incremental cost-effectiveness ratio of NZ $1590 (95% confidence interval [CI] = 1390–1791) per quality-adjusted life-years gained. 

In a systematic review, Agyekum et al. evaluated the cost-effectiveness of interventions for myopia and its complications, including those for preventing myopia progression, correcting refractive error, and treating pathologic myopia using costs, quality-adjusted life-years, and incremental cost-effectiveness ratio as outcome measures.³⁷⁷ Low-concentration atropine (0.01%) and corneal refractive surgery were the most cost-effective treatments for myopia treatment, and ranibizumab and conbercept were found to be the most affordable treatments for pathological myopia. Preventing myopia progression was reported to be more cost-effective than treating pathological myopia. For instance, using 0.01% atropine for myopia progression produced an incremental cost-effectiveness ratio of $1001 per quality-adjusted life-years versus $12,852 to $246,486 per quality-adjusted life-years for treating pathologic myopia. Although the cost-effectiveness of refractive surgery was low, but 0.01% atropine was found to be a more cost-effective treatment due to lower treatment cost and additional benefits of preventing myopia complications and related visual impairment. 

More recently, a Markov model was used to perform an economic evaluation on the cost-effectiveness of 13 interventions for preventing myopia progression in children.³⁷⁸ Over 5 years, 0.05% atropine and time outdoors were found to be the most cost-effective interventions. An amount of 0.05% of atropine had an incremental cost-effectiveness ratio of US $220 per spherical equivalent reduction, whereas outdoor activity resulted in a cost savings of US $5 per spherical equivalent reduction and US $8 per axial length reduction. Additionally, highly aspherical lenslets and repeated “laser” red light therapy were other cost-effective options with an incremental cost-effectiveness ratio of US $448 and US $846 per spherical equivalent reduction, respectively. Orthokeratology was also found to be cost-effective, albeit at a slightly higher incremental cost-effectiveness ratio of US $2376 per axial length reduction. 

Agyekum et al.’s study has attempted to fill the gap in understanding the cost-effectiveness of myopia treatments,³⁷⁸ but significant methodological flaws, such as inconsistent treatment effect sizes, linear extrapolation of effect to 5 years, and an inexplicable control measure, limit its utility. For instance, approximately −2.50 mm reduction in axial length and −5.00 D reduction in spherical equivalent refraction for single-vision spectacle lenses were included as baseline values of a comparator for cost-effectiveness analysis of interventions. In addition, the treatment effect sizes used for the model-based analysis seemed inconsistent with the literature. For example, the axial length control effect size for orthokeratology lenses (0.36 mm) was significantly larger than the remarkably consistent pooled effect size (0.26–0.28 mm over 2 years) reported in several meta-analyses.^379–381 The effect size of 0.17 mm reduction in axial elongation for 0.01% atropine eye drops is significantly greater than the observed overall effect size of 0.06 mm over 1 year across clinical trials. These amplified effect sizes appear to be derived from a review³⁸² that included many inaccuracies treated and control group effects in individual studies. As an example, change in axial length was reported as 0.03 mm/year in the orthokeratology group, 2.08 mm/year in the control group of 1 study, and 0.06 mm/year in the orthokeratology group and 0.59 mm/year in the control group in another study. The model also used a 5-year time horizon, so the effect sizes were further multiplied by 5 to obtain a 5-year effect without accounting for the well-established reduction in the efficacy of myopia treatments over time, with the greatest efficacy in the first year of treatment.³⁸³ Considering these limitations, the potential significance of the findings is low. 

Apart from the evidence of effectiveness, evidence regarding the value proposition of interventions seems critical. To address this gap, Fricke et al. developed and modeled a system for assessing and comparing the lifetime financial expenses of active myopia management (i.e. use of treatments) against traditional myopia management (i.e. use of corrective lenses).³⁶³ In contrast to Agyekum et al.,³⁷⁸ who modeled cost-effectiveness options for a 10-year-old child with varying levels of myopia, Fricke et al. reported estimates for an 8-year-old child from urban regions of Australia and China presenting with symptomatic myopic of −0.75 D in both eyes and was given the choice of active or traditional myopia management. Options for these managements were available in both nations, and the costs of all product kinds, essential appointment fees, and other relevant expenditures were calculated using data provided by important sources in both countries. The authors found that the lowest lifetime cost option was anti-myopia spectacles in Australia and low-dose atropine in China. The lifetime cost for traditional myopia management with a 3% discount rate was US $7437 (95% CI = US $4953 to US $10,740) in Australia and US $8006 (95% CI = US $3026 to US $13,707) in China. The final level of myopia had the greatest impact on the lifetime costs of myopia in both countries. 

In summary, lifetime savings from reduced myopia often offset the upfront costs of myopia treatments in childhood. The use of low concentrations of atropine is cost-effective, but it requires additional cost of correction with spectacles or contact lenses. A dose of 0.05% of atropine can reduce myopia progression in children with acceptable side effects and can minimize adult myopia treatment costs.³⁷⁷ Economic reviews of myopia treatments and evidence on cost-effectiveness are lacking. As childhood myopia is increasing in prevalence and severity worldwide, and multiple myopia treatments are now available, a robust and comprehensive analysis of the cost-effectiveness of myopia treatments will provide critical data for health policy decisions, thereby maximizing health outcomes with limited resources. 

Treatment Efficacy, Approval, and Clinical Trial Considerations

Inconsistency in reporting treatment efficacy impedes comparisons of treatment effects among various interventions and impacts researchers’ and eye practitioners’ assessment of optimal treatments. The efficacy of myopia is commonly reported as a percentage change in myopia progression or axial elongation or annualized outcomes of these measures between the treated and control groups.^{3,208,384–387} However, reporting efficacy in relative terms has significant limitations due to its inherent assumption of a constant efficacy across varying progression rates and treatment duration.³⁸³ Consequently, it has been proposed that efficacy is best reported as a cumulative absolute reduction in axial elongation, reflecting true treatment effects over a particular period, regardless of critical determinants of efficacy, such as age, progression rate, and ethnicity.³⁸³ To date, the maximum reported absolute efficacy is limited to 0.44 mm reduction in axial elongation or 1 D slowing of myopia progression over 3 to 5 years.³⁸³ 

As mentioned previously, dual-focus soft contact lenses remain the only treatment approved by the FDA for slowing myopia progression in children. Although treatments can still be prescribed “off-label,” the designation of FDA-approved treatment often carries an indication of a high level of efficacy and safety of a treatment. One of the biggest barriers to interventions being submitted for FDA approval appears to be the difference in position regarding the endpoints among the two centers overseeing the approval process of drugs and devices. Whereas approval of drugs, such as atropine eye drops, falls under the purview of the FDA Center for Drug Evaluation and Research, devices such as spectacles and contact lenses are within the purview of the FDA Center for Devices and Radiological Health. The endpoint threshold for approval of products appears to be different between the two centers. For instance, the FDA Center for Devices and Radiological Health approved dual-focus lenses, which showed treatment effects of −0.73 D and 0.32 mm over 3 years. The approval process was likely influenced by an earlier FDA co-sponsored workshop, which had determined a treatment effect of −0.75 D reduction or 30% to 40% reduction in spherical equivalent refraction as a clinically significant treatment effect.³⁸⁸ If approval of dual-focus lenses were regarded as the precedent an acknowledgment of myopia as a global public health problem, withholding other treatments that have demonstrated comparable efficacy in lesser duration (−0.80 D and 0.35 mm over 2 years for HALT lenses and −0.44 D and 0.34 mm over 2 years for DIMS lenses)^301,389 appears ethically inappropriate. Unlike the devices, approval of drugs by the Center for Drug Evaluation and Research seems to require less stringent criteria based on responder analysis, that is, the proportion of participants or eyes meeting a set threshold, as evident from the recent CHAMP trial, which used the proportion of 0.02% atropine-treated eyes showing less than 0.50 D progression over 3 years as the primary endpoint.³⁹⁰ Although this endpoint was not met in the CHAMP trial, it seems a relatively low bar considering the earlier precedent set during the approval of dual-focus lenses. Notably, the absolute 3-year treatment effects in the CHAMP trial (slowing of myopia progression = 0.10 D with 0.02% atropine and 0.24 D with 0.01% atropine; slowing of axial elongation = 0.08 mm with 0.02% atropine and 0.13 mm with 0.01% atropine) were small, clinically insignificant, and remarkably similar to the effect sizes of progressive addition lenses in the COMET trial.⁸⁰ It remains to be seen whether the community would embrace a treatment with questionable clinical meaningfulness if it were to receive FDA approval for myopia treatment.³²² Moreover, the limitations associated with the responder analysis strategy that requires dichotomization of a continuous measure are well documented and likely to yield an overestimation of benefits.^383,391,392 

Challenges related to the design of clinical trials have the potential to impede the future development of novel, safe, and effective treatments for childhood myopia.³⁹³ Research in myopia treatment space faces several challenges, including the ethical dilemma of withholding treatment to include a true placebo group, recruitment difficulties, and the likelihood of higher attrition in the placebo group due to the availability of proven treatments, selective withdrawal of fast progressors in the placebo group, thereby affecting efficacy outcomes, and the potential of participants seeking treatments outside of the trial.³⁹⁴ Considering these challenges, a conventional placebo-controlled trial may no longer be feasible. As an alternative to the conventional randomized, controlled clinical trial, a recent commentary³⁹⁴ outlined several strategies, such as a non-inferiority design with an approved drug or device as the control,³⁹⁵ prediction of longer-term efficacy from short-term studies using initial control data as an input to progression models based on previous trials, the inclusion of age, sex, and ethnicity-matched virtual control groups based on previous data,³⁹⁶ extrapolation of short-term control data to subsequent years based on population-specific proportional reduction in axial elongation,³⁹⁷ and, finally, survival analysis to evaluate time-to-treatment-failure allowing participants to exit the trial after a certain threshold of progression.³⁹⁸ Some of these strategies are not without limitations, however. For instance, determining efficacy using a virtual control is only feasible if progression can be reliably predicted based on prior data in a specific population. This is not always the case, as demonstrated by the progression rates in the control groups of DIMS and HALT lens trials. Despite similar age at enrollment (mean age = approximately 10 years) and ethnicity, these two control groups progressed at contrasting rates, with a considerably faster progression in the HALT lens trial (−1.46 D and 0.69 mm over 2 years) than in the DIMS trial (−0.85 D and 0.55 mm over 2 years). 

For an issue of global public health significance, such as myopia, timely adoption of proven treatments can make a substantial difference in limiting future consequences, as evidenced by the rapid rise of myopic maculopathy as the leading cause of new blindness in several East Asian countries with high prevalence of myopia.^399,400 Moreover, the axial length control endpoint warrants greater emphasis as a clinical trial endpoint because it best relates to the primary justification of administering myopia treatments, which is to slow axial eye growth and reduce the risk of myopia-related complications in later life. 

Overall Summary

Significant progress has been made in understanding the mechanisms behind contemporary treatments for childhood myopia. Several new treatments, such as light therapies, are on the horizon, and evidence regarding their safety and efficacy is evolving. However, important questions remain about the clinical application of current and emerging myopia control treatments, including dosage, duration, frequency, and cost-effectiveness, long-term safety and efficacy, and their benefits in preventing myopia-related complications. Several knowledge gaps also exist in terms of patient-level treatment considerations, such as the age of treatment initiation and termination, duration of treatment, evaluation and prediction of individualized treatment success, identification of nonresponders, and compliance and safety. Further research is warranted to better understand the mechanisms of myopia treatments, evaluate the long-term safety and efficacy of existing and emerging treatments, and identify patient- and treatment-related considerations for maximizing treatment benefits in the clinical care of children with myopia. 

Acknowledgments

Supported by a grant from the NIH/NEI (R21EY036536 to SK). 

This manuscript is a part of the commissioned paper, “Treatment of Childhood Myopia” submitted by the authors to the study committee for the National Academies of Sciences, Engineering, and Medicine (NASEM) consensus study, “Focus on Myopia: Pathogenesis and Rising Incidence.” This consensus study was supported by the American Academy of Optometry; the American Optometric Association; the Health Care Alliance for Patient Safety; the Herbert Wertheim School of Optometry & Vision Science, University of California, Berkeley; Johnson & Johnson Vision; the National Eye Institute; Reality Labs Research; Research to Prevent Blindness; and the Warby Parker Impact Foundation. 

Disclosure: S. Khanal, Meta (F, R); E.S. Tomiyama, CooperVision (R); S.C. Harrington, Thea Pharmaceuticals (R), CooperVision (R)