There has been an increase in the number of published reports dealing with myopia. In fact, the questionable scientific value of some of the early writings, as well as the controversy as to whether myopia is environmentally induced or genetically determined, may have inhibited myopic research during the middle decades of the twentieth century. As an example, ‘Investigative Ophthalmology and Vision Science’ published only one article on myopic research during the period 1961 to 1977, while in more recent years, it is rare to note an issue of IOVS without at least one article on the topic of myopia and often there are several. The same may be said of conference presentations at international eye research meetings.
Animal models of the development of refractive error in the late 1970s are, at least in part, responsible for the increase in attention directed to research on myopia. In his book, ‘The Myopias’, published in 1985, Curtin notes that while it is recognised that heredity and environment are the two factors involved in the production of myopia, ‘There is almost unanimous acceptance of the former but major reservations continue to surround the latter.’ In a 1996 review on the role of genetics versus the environment in relation to myopia, Mutti, Zadnik and Adams note that: ‘This debate was at something of a standstill until the revolutionary discovery in the late 1970s that depriving a young animal eye of clear vision could result in myopia.’
During the past 35 years, the avian eye, particularly the chicken eye, has become a common animal model of ocular refractive development. This is a result of the discovery by Wiesel and Raviola in 1977 and by Wallman, Turkel and Trachtman in 1978, that it is possible to induce myopia in macaque monkeys and chickens by depriving the eye of clear form vision during early development. In fact, earlier work, such as that by Lauber, Macginnis and Boyd in 1966 showed that it is possible to alter eye size and eye development in chickens by manipulating their exposure to light and through the use of occluding goggles; however, the emphasis of this study was on a glaucoma model rather than on a model of refractive development and this meant that these earlier examples of environmental manipulation of ocular development did not attract the attention of those of the late 1970s. The 1977 Wiesel and Raviola report in Nature is often the starting point of reviews and introductions to this line of research. Curtin felt it necessary to validate his description of the late 1970s and early 1980s period of animal research on myopia by pointing out that the initial monkey work involved ‘the Nobel laureates Hubel and Wiesel …’ In fact, while Hubel was an author on the study of the effect of monocular deprivation on the development of the visual cortex, which led to the discovery of the effects of deprivation on the eye, he is not an author on the latter work dealing with myopia.
As a precocial bird that undergoes a rapid phase of early ocular growth, the chicken has become the predominant species used in this research because the manipulation of the visual environment can commence at hatching and the effects on ocular development take place within days or hours. Form deprivation usually results in the induction of myopia. In 1988, Schaeffel, Glasser and Howland demonstrated that it is possible to induce either myopia or hyperopia by using concave or convex lenses to alter the retinal image, and Irving, Callender and Sivak[48, 49] and others have demonstrated the wide range of refractive states that can be induced using lightweight goggles and lenses. Essentially, a concave lens simulates hyperopia, causing a compensatory increase in ocular growth with the result that the eye is myopic when the lens is removed. The opposite, a slowing of ocular growth, occurs with the application of convex lenses. Although attention has focused on change in axial length in studies of visually induced refractive error, the roles of the cornea and lens are less clear. Zhu and colleagues have shown that the imposition of positive or negative lenses causes ocular changes in chick eyes within one or two hours.
While experiments dealing with the development of refractive error are frequently based on animal models such as the chick, various mammals such as monkeys and tree shrews have also been studied. Other vertebrates that have been used to induce form deprivation myopia include: grey squirrel, mouse, guinea pig,[55, 56] cat and another bird, the American kestrel. It is assumed that mammal models are more comparable to the human eye in terms of ocular structure and function, as well as genetic profile; however, when it comes to diurnal vision and the existence of significant accommodative ability, the avian eye is a reasonable model as well.
All of these are higher vertebrates. Typically, experimental myopia is induced in young animals, not adults, and the younger the animal the higher the induced refractive error.[46, 47, 60] Moreover, it has been shown that unilateral congenital cataracts can result in axial eye elongation in humans.
Experiments, such as those in which the chick optic nerve is cut, show that refractive error development is determined by a local retinal mechanism. Somehow the eye distinguishes whether the input visual signal is over- or under-focused and in chicks at least, rapidly adjusts retinal position and focal length of the eye, by thinning or thickening of the choroid. Ocular growth then accelerates or slows down for a more permanent change in refractive state, resulting in an eye that is either too long or too short.
Form deprivation and positive or negative lenses can be applied to the fish eye, with results that are very similar to those obtained with higher vertebrates, such as chicks. This suggests that a universal mechanism controls the refractive development of the eye, regardless of differences in morphology, physiology and habitats and the fact that fish continue to grow through life.[64, 65]
Contemporary efforts to prevent or reduce myopia
In a review of treatment options for myopia, Gwiazda specifies that to be considered seriously, myopic treatment studies should include a concurrent control group, random assignment to the treatment and control groups, masking of investigators who collect the data, standardised measurements, a large enough sample and a small loss to follow-up. In general, results published on myopia during the last quarter of the twentieth and into the twenty-first century are based on experiments carried out on larger numbers of subjects, with better controls, standardised measurements and appropriate masking.
Visual Training and Biofeedback
Efforts to improve training through the use of auditory biofeedback were made in the 1970s and 1980s.[67, 68] This involved the development of an instrument, the Accommotrac, which is essentially an optometer that measures the vergence of light reflected by the retina and also provides a tone that varies in pitch with accommodation. Subjects are trained to use the sound to accommodate and unaccommodate, a form of accommodative facility training; however, a careful study that included the use of ‘blind’ examiners to measure refractive state after training concluded that positive findings are based on a learned effect on visual acuity, not a reduction of myopia.
A long-standing method used by clinicians to slow down the progression of myopia is to undercorrect the myopic eye by as much as a dioptre, as an effort to avoid excessive accommodation.[3, 28] In fact, virtually all texts dealing with refractive procedures, going back at least to the time of Landolt in the 1880s, instruct the clinician to record as a subjective endpoint the least minus correction that provides maximum visual acuity to minimise accommodation. It is tempting to suggest that the original reasons for withholding some of the minus correction were superstitious in nature, although no evidence has been found in the literature to support this notion. Mutti, Zadnik and Adams note in their nature versus nurture review that the study of myopia is characterised more by deeply held belief than by research.
The rationale behind undercorrection is not clear, although recent models of emmetropisation suggest that explanations based on a closed feedback loop model of ocular refractive development can provide an explanation.[71, 72] Grosvenor noted that there were no studies of the effects of undercorrection on myopia; however, Goss refers to two poorly controlled and unconvincing studies involving small numbers of subjects in his review. Later, Ong and colleagues showed that spectacle intervention, in terms of whether young myopes wore their corrections full-time, part-time or not at all, had no effect on the progression of myopia, while Chung, Mohidin and O'Leary found that an undercorrection of 0.75 D resulted in a more rapid progression of myopia. Finally, Goss demonstrated that an overcorrection of the same amount (0.75 D) had no effect on myopic progression.
The idea behind the use of bifocals is based on the supposition that excessive accommodation can affect the eye mechanically and structurally or to the possibility that a change in the relationship between accommodation and convergence can affect refractive development. Early investigations concerning the value of prescribing bifocals to children were described above in the section on early efforts to prevent or reduce myopia. Subsequent research into the utility of bifocals in controlling myopic development were carried out by Oakley and Young, Goss and Grosvenor and colleagues (also known as the Houston Study). Neither the Goss report nor the Houston Study, the largest and best-controlled study, showed any retardation in progression of myopia with bifocal wear. While the Oakley and Young study did show a positive effect, it has been criticised on the basis of lack of adequate controls for experimenter bias, as the authors themselves note. Moreover, the Houston Study was a three-year prospective approach, whereas the other two efforts were retrospective ones. A re-examination of the data of this study and two others by Goss and Grosvenor showed a small reduction in the rate of myopic progression (about a fifth of a dioptre less per year) in subjects with esophoria. This has also been shown more recently by Fulk, Cyert and Parker, who followed 82 children with near-point esophoria for 30 months and found a slight reduction (0.25 D) for those wearing bifocals.
In 1999 Leung and Brown published the results of a two-year longitudinal study in which the rate of myopic progression was compared in Hong Kong Chinese children wearing single vision correction versus those wearing progressive addition lenses. The progressive lenses were used to improve subject compliance in wearing the correction and to provide clear vision at all working distances. The results indicated that the progression of myopia is significantly reduced in those subjects wearing progressive lenses; however, a later study, in which the examiners were masked (unlike the Leung and Brown work) and which involved a larger number of children from the same population produced results indicating no significant difference between the single vision and progressive lens groups. Finally, a large multi-centre study, the COMET study, involving 469 ethnically diverse children, who were followed for three years, concluded that there was a small and significant difference in myopic progression between the single vision group and the group wearing progressive lenses; however, the difference (0.20 D) does not indicate that there should be a change in clinical practice. A similar result was reported from a Japanese clinical trial involving 92 children.
Earlier, Gwiazda, Grice and Thorn found that myopic children with esophoria also demonstrate large lags of accommodation at the near point. The question of whether progressive lenses can be effective at retarding the progression of myopia in children with near point esophoria and large lags of accommodation has been addressed in a couple of recent reports published in 2011 and 2012. One large 120 collaborator multi-centre double masked study involving 118 children found a statistically small but clinically unimportant effect on slowing myopia. A similar result was found for 85 children with high accommodative lags, that is, a small statistically significant effect but one with little or no clinical relevance. A few years earlier, a report from the CLEERE (Collaborative Longitudinal Evaluation of Ethnicity and Refractive Error) Study Group involving over 1,100 children showed that the amount of accommodative lag is not a valid predictor of which children will become myopic, although increased accommodative lag was noted in children after they became myopic.
Contact Lenses and Orthokeratology
The potential for using contact lenses to more than just correct refractive errors was noted during the period of corneal contact lens development in the early 1950s and 1960s, particularly since the early fitting strategies involved using base curvatures that were flatter than the curvature of the cornea to maximise tear flow under the lens.[89, 90] The resultant flattening of the cornea meant that myopia was reduced, at least for a time. This led practitioners to deliberately flatten the cornea through a regimen referred to as orthokeratology; however, in 1960, Morrison noted anecdotally that myopic progression in children seems to diminish, even when the contact lenses are prescribed with base curvatures that parallel the corneal surface. The basic conundrum that has bedevilled the approach involving the use of contact lenses to treat myopic progression was, and still is, whether the sole effect of the lenses is to flatten the cornea, at least temporarily, or whether the lenses also have a slowing effect on the growth of the eye.
Stone[91, 92] carried out a substantial five-year study of myopic progression involving 124 children, both with and without contact lenses. The initial study concluded that two years of contact lens wear stabilised the progression of myopia but the effect was due to corneal flattening and not to a change in axial length of the eye. In the later study, over a longer period, myopia was significantly reduced in contact lens wearers. While some of the effect was due to corneal flattening, not all of it was, suggesting that the lenses do affect the axial elongation of the eye.
There has been a resurgence of interest in orthokeratology to prevent myopia as a result, due to the high prevalence of myopia, especially among Asian populations. An early scientific study of the efficacy of orthokeratology by Kerns found that the approach reduced myopia in some cases but the permanence of the changes once the lenses are removed, was not clear. A large study by Polse and colleagues published in 1983 also showed that while it is possible to reduce myopia in this way, the effect is temporary and continued wearing of lenses, at least intermittently, is required to maintain an effect. Moreover, during periods when lenses are not worn, visual acuity can be unstable.
Another relatively large study involving 100 children followed for three years involved the use of regular (non-orthokeratology designed) gas permeable contact lenses. The results also indicate that there is a reduction in myopia and not all of this change is a result of changes in the cornea, suggesting a possible effect on the growth of the eye. A retrospective study in young adults more than 20 years of age indicates that while myopic progression does take place in young adults, there is no association with type of refractive correction (spectacles or contact lenses).
In 2008, a retrospective study by Chan, Cho and Cheung on Hong Kong children seen in a university clinic reported that orthokeratology lenses for overnight wear were effective in reducing low to moderate levels of myopia. That orthokeratology retards the axial elongation of the eye is supported by a case report, which indicates that suspension of orthokeratology results in faster axial growth of the eye. Finally, a Japanese study shows that while orthokeratology cannot stop axial elongation in progressive myopia, it can slow it down.
The increase in the wearing of soft contact lenses made of varying materials and varying oxygen permeability has also led to retrospective[96, 100] and prospective studies of the effects of various contact lenses on myopic progression. The results indicate that lens material can affect myopic progression, although whether the differences are due purely to varying mechanical effects on the cornea is not clear.
There is a substantial history of efforts to manage the development of myopia through the use of pharmaceuticals. The drugs used have been primarily cycloplegics meant to paralyse the ciliary muscle to cause a loss of accommodation, but mydriatics and anti-glaucoma drugs have been used as well. Thorough reviews of the earlier efforts are provided by Curtin and by Jensen and Goldschmidt, who refer to studies that were carried out without appropriate controls, with high dropout rates and also poor subject compliance. Both reviews express reservations about the use of drugs in dealing with myopia, in terms of potential dangers from side effects and both refer to the possible danger to the retina from larger than normal amounts of solar radiation entering the eye through a dilated pupil. In addition, Jensen and Goldschmidt note that: ‘In our opinion, it is of very limited value to reduce the final amount of myopia from 4.00 D to 3.00 D, especially if the reading capability or binocularity of the child is negatively influenced by the drug.’ One can add the additional concern about the use of cycloplegic drugs to control myopia when the original rationale was based on an unproven link between accommodation and myopia.
There have been animal studies that have addressed the issue of accommodation, drugs and myopia, either directly or indirectly. As mentioned earlier, myopia can be induced by depriving the eye of form vision in the grey squirrel (Sciurus carolinensis), one of the few non-primate diurnal mammals, even though the eye has no accommodative ability. Experiments showing that form deprivation myopia can still be produced in chicks, in which the optic nerve is cut, also suggest that accommodation is not involved. This point is also emphasised by the fact that myopia and hyperopia can be induced in fish,[64, 65] where accommodation takes place by movement of the lens, not through a change in lens shape.
McBrien, Moghaddam and Reeder also found that while atropine prevents the experimental induction of myopia and the axial elongation of the eye in chicks, the effect is not related to the accommodative mechanism. Follow-up studies[104, 105] showed that daily injections of the M1 antagonist pirenzepine can also prevent form deprivation myopia in chicks and tree shrews by selectively blocking M1 receptors in the retina or choroid. A similar result was found in form deprivation experiments with guinea pigs, in which the topical application of pirenzepine prevents the development of myopia.
In 2003, Bartlett and colleagues reported positive results on the tolerability of the application of 0.5 to 2.0 per cent pirenzepine to the eyes of children to evaluate its potential use in the prevention of myopia without producing the side effects on pupil size and accommodation associated with atropine. In 2008, the results of a large multi-centre study involving 174 children (eight to 12 years of age), in which the treated group (n = 117) received daily applications of pirenzepine (2 per cent) were published. The results show that after two years there was a mean 0.58 D increase in myopia for the pirenzepine group, compared to a mean increase in myopia of 0.99 D for the placebo group; however, 14 of the treated group dropped out of the study because of adverse effects such as medication residue and accommodative dysfunction. Also, the difference in axial length of the eyes of the two groups was not significant. Chen and colleagues reported the effects of racanisodamine (0.5 per cent), a racemic isomer of anisodamine on the pupil and accommodation of the eyes of 20 children, nine to 12 years of age. There was a moderate effect on pupil size and no effect on accommodation, suggesting that this drug, a non-selective muscarinic antagonist, may be a good candidate for study as a means of retarding the progression of myopia.
In summing the situation vis à vis the use of pharmaceuticals as a treatment of myopia, it is instructive to repeat the point made by Jensen and Goldschmid, which questions the value of modest reductions of myopia if there are associated risks, particularly in children. Of course, the same point can be made with regard to other treatments, such as bifocals or progressive lenses. Why promote treatments that may have, at best, moderate results, if it means that there is the possibility of negative side effects?