http://orcid.org/0000-0001-6635-1098Samantha Sze-Yee Lee1,2,3
Since the publication of the landmark ATOM2 (atropine for the treatment of myopia)1 and low-concentration atropine for myopia progression (LAMP) findings,2 low-concentration atropine eyedrops have been a preferred treatment method for myopia control. The ATOM series further inspired several other randomised controlled trials (figure 1), most of which showed that 0.01% atropine eyedrops have some myopia control effect. However, the efficacy of 0.01% atropine varied widely between studies, and this variation is not well explained by baseline spherical equivalent or age (figure 1). The LAMP study2 further reported slightly higher myopia control efficacy with 0.05% concentration, while maintaining a low risk profile. The crossover MOSAIC (myopia outcome study of atropine in children)3 similarly found that, in Irish children, a 0.05% concentration was more effective than a 0.01%, although with smaller effects than those observed in LAMP (figure 1).
Figure 1
Bubble plot showing 12-month treatment effect of 0.01% or 0.05% atropine compared with a placebo, according to baseline refractive error (left) and age (right), reported by randomised placebo-control trials. Studies represented in this figure include trials cited in table 2 of Chia et al’s study in this issue (where baseline age and refractive error information were reported), plus MOSAIC. The size of the bubble (circle or triangle) corresponds to the study sample size. ATOM, atropine for the treatment of myopia (WA, Western Australia; J, Japan; I, Indian); D, dioptre; LAMP, low-concentration atropine for myopia progression; MOSAIC, myopia outcome study of atropine in children; PEDIG, paediatric eye diseases investigator group; SER, spherical equivalent refraction.
The next natural step is to determine whether low-concentration atropine eyedrops can be used for myopia prevention. In 2017, both the ATOM3 and LAMP2 studies commenced their myopia prevention trials.4 5 Despite previously showing that 0.01% atropine eyedrops are moderately effective in slowing progression, both trials4 5 found that the treatment has minimal impact on myopia prevention. A 0.05% concentration, on the other hand, decreased the incidence of myopia by about two-thirds after 1 year and by half after 2 years in the LAMP2 study, relative to a placebo. This represents a major step forward in myopia prevention.
The long recruitment period of the ATOM3 trial,5 relative to its predecessors, is noteworthy. In ATOM3, 217 children were recruited over 4.5 years for the premyopia and low myopia trial, compared with ATOM1, which enrolled 400 children in just 18 months. The slow recruitment rate is not a limitation of ATOM3 per se, but is an important observation in itself. It highlights two contrasting parental attitudes towards active myopia control: nonchalance before myopia onset and heightened concern once their child develops myopia.
As noted by the authors of ATOM3,5 even though eligible children had at least one parent with myopia—a trait expected to increase concern for their child’s sight—parents were reluctant to start their premyopic children on treatment. If parents are unwilling to have their child participate in a trial that provides free treatment and eye care to test a well-tolerated medication for preventing a condition they themselves have, it may be even harder to persuade them to adopt these myopia preventive measures in real-world settings. Nonetheless, studies in China6 7 have shown that parents generally have a positive attitude towards adopting what they presume are myopia prevention measures, such as increasing room lighting, restricting screen time and promoting eye exercises for their child (although there is limited evidence that any of these actions have protective effects). But based on the slow recruitment of the ATOM3 study and my own personal conversations with parents within Australia, parents appear more willing to adopt lifestyle changes, regarded as less invasive, compared with using prescription eyedrops for myopia prevention for their children.
The only lifestyle change that has been proven in randomised controlled trials to be protective against myopia is spending more time outdoors. This preventive strategy is vastly different from atropine eyedrops in regard to impact on lifestyle and implementation strategy. Increasing time spent outdoors is low-cost and has added mental and physical health benefits. However, it may require coordination between the school and the health/public health departments8 and greater commitment from parents. This may not be feasible for families with busy schedules or in unfavourable weather conditions, such as humid summers or wet winters.
Low-concentration atropine eyedrops, on the other hand, are relatively more convenient to implement (a few minutes per day to instil compared with 2 hours per day outdoors), well tolerated and not weather-dependent. However, it incurs financial costs for specialist visits and the product itself, which may not be accessible for lower-income families.
To implement successful public health interventions for myopia control, further studies are necessary to appreciate parents’ willingness and capacity to implement different myopia prevention strategies. Importantly, studies must be region-specific and even individualised, as lifestyles, culture, socioeconomic status and geographical climate are likely to impact myopia control practice and attitudes.
Additionally, as Chia et al 5 suggested, the risk of myopia and unnecessary treatment should be balanced. It may not be wise to immediately start all children with premyopia on treatment, nor would it be practical, as this adds unnecessary burden on the healthcare system. The ATOM3 study was meticulous in its participant selection—enrolling only high-risk children (premyopia plus at least one myopic parent). Based on the placebo group’s outcome, the 2-year myopia incidence in young premyopic children is 40–50%. While this risk is high, this still means that half of the group does not require treatment. Studies are required to improve the risk profiling of children with premyopia to identify those who are most likely to benefit from treatment.
ATOM3 conversely noted that parents whose children had low myopia—a previously neglected refractive status in atropine trials—were eager to have their children started on treatment. With the growing availability of prescription atropine eyedrops, these parents would bypass randomisation to seek treatment directly, slowing down ATOM3’s recruitment progress. Practitioners could leverage this parental motivation to commence myopia control early.
Participant dropout in the low-myopia group was also high, close to 50%, with myopia progression as the most cited reason for withdrawal. Among the remaining children, >80% of children with low myopia receiving a placebo progressed by −1D (dioptre) or more, compared with ∼42% of premyopia children on placebo. Despite diminished study power due to the small sample and high dropout rates, 0.01% atropine showed significant, although modest, myopia control effects in the low-myopia group. This suggests that children with low myopia may derive greater benefit from intervention than those with premyopia. As summarised by Flitcroft,9 there is ‘no safe level of myopia’. Mendelian randomisation studies estimated that for each dioptre reduction in spherical equivalent refraction, the OR of primary open-angle glaucoma10 11 and retinal detachment12 rises by 4–8% and ∼40%, respectively. Early commencement of treatment in children with myopia, no matter how low the refractive error, may be advisable and accepted by parents.
With effective myopia prevention and control options now available, it is time direct some of our attention to the risk of myopia complications as the outcome. It has been assumed that controlling myopia progression would consequently reduce the risk of its complications, though long-term data are limited due to the relative recency of myopia control. The ATLAS13 (atropine treatment long-term assessment study) re-examined the ATOM1 and 2 study participants 15 and 20 years after treatment initiation (placebo or 0.01–1.0% atropine eyedrops). Atropine eyedrops were used only for 2–4 years due to the nature of the original studies, and no difference in rates of myopia-related complications was found between the placebo and treatment groups after 15–20 years. In real-world settings, myopia treatment is likely to last longer than a few years, possibly even into adulthood, and be tapered before cessation (therefore reducing rebound effects). Hence, there is optimism that real-world long-term treatment with atropine eyedrops would yield better outcomes in terms of complication risk. Given that effective, early myopia control is expected to reduce future risk of complications, and that studying long-term outcomes would require decades-long follow-up, initiating experimental trials today would be neither feasible nor ethical. Going forward, studies on long-term complication risks with myopia control treatment will likely need to be observational in nature.
Ethics statements
Patient consent for publication
Not applicable.
Ethics approval
Not applicable.
References: https://bjo.bmj.com/content/110/4/353



