Controlling Myopia Progression - A Confusing Story
From Wildsoet Lab, UC Berkeley
The dilemma of managing young progressing myopes
Two recently published clinical trials involving under-correction and PALs as alternative myopia control strategies add more rather than less confusion
“Eye correction is seriously short-sighted” was the headline to an article by Andy Coghlan and Michel le Page in the New Scientist late last year (11-26-02) covering a randomized clinical trial of undercorrection as a treatment for myopia progression in children.
The New Scientist article was referring to a paper published in the journal, Vision Research (2002, 42: 2555-9), describing a 2-year Malaysian-based study comparing the effects of undercorrecting myopia with full correction on myopia progression in children. The message from the principal investigator on this study, Dr O’Leary, to doctors, patients and parents, as reported in the New Scientist article is “No glasses is the worst option of all, But don’t undercorrect. GO for full correction.”
The New Scientist article is likely to have cause both alarm in patients and guilt in parents and clinicians alike with its headline statement: “Millions of people worldwide may have worse vision and even be more likely to go blind because of a long-held but misguided idea about how to correct short-sightedness.” The article goes on to claim that: “A study intended to confirm the theory has instead been stopped because the children’s eyesight was getting worse.” A closer examination of the Vision Research paper covering the Chung et al study suggests that there are reasons to question undercorrection as a management study for all progressing myopes but also a case of serious exaggeration by the New Scientist reporters.
The Chung et al study is a small (n=94), 2 year randomized and masked prospective study comparing the effects of full-time undercorrection (UC, by approx 0.75 D) with full-time fully correction (FC) in young myopes (mean: -2.86 D). The study group comprised approximately 1.4 time the numbers of girls as boys with Chinese and Malay ethnic groups being approximately equally represented. Over the 2 years of the study, the FC group showed a progression of –0.77 D compared to the UC group that exhibited a progression of –1.00 D. Rates of eye growth also differed between the two groups, as expected, being slower for the FC group.
What message can we take home from the Chung et al study? Unfortunately this study is sparse on detail – and thus difficult to interpret. We are told nothing about the refractive distribution of the subjects although we are told that analyses were based on 4 refractive error categories, including one covering >3 D myopia. We can not assume that all refractive groups would have behaved similarly – for example, the more recently published COMET study observed significant refractive error differences (see below). We are not provided with any information about the binocular vision or accommodative status of the subjects yet these parameters are predicted to also influence outcomes based on other studies (see COMET study below). Finally, while full compliance was defined as wearing the glasses for at least 8 hr per day, we are not told when the children typically went to bed. Thus it seems impossible to rule out the possibility that compliant children from either or both treatment groups may have undertaken near activities without their glasses at night.
It is interesting to compare progression rates for the two groups in the Chung et al study with values from the more recently published COMET study, converted in both cases to a D per year rate. The progression rate for the UC group (-0.5 D per year) corresponds closely to the mean rate reported for participants allocated single vision lenses in the COMET study (-0.49 D per year). A conclusion based on this comparison alone would be that undercorrection neither exacerbates or slows the progression of myopia, when applied unselectively. This outcome is predicted if we assume that the benefit of undercorrection is limited to those with poor accommodation. Animal model studies predict increased (myopic) eye growth with sustained poor accommodation in fully corrected eyes (see Wildsoet, 1997, for a more extensive discussion of animal-based emmetropization studies and their clinical implications). However undercorrection should improve the state of focus at near as less accommodation is required. A potential parallel with animal studies involves the imposition of binocular low powered positive lenses on young monkeys; their eye growth slows, presumably because their eyes now have almost perfect focus at near distances, the limit of the visual world of these young animals (Smith & Hung, 1999). It might also be argued based on animal studies, that because the undercorrection strategy imposes myopic defocus at distance, that eye growth should be slowed in all children. However as school children spend little time outdoors focussed on more distant tasks, they are likely to gain little benefit from this correction strategy in terms of reduced eye growth. Nonetheless, this interpretation leaves us with the dilemma of explaining why full correction slows myopia progression (rate: -0.39 D per year) in a population that is typically more susceptible to myopia compared to Western populations (and the COMET study subject base). However children from the FC group with poor accommodation and/or over-convergence problems had potentially the most to benefit from not wearing their glasses at night. Only another study will provide us with the answer as to whether such behavior is the explanation for the apparently slowed progression of FC myopes.
The message that the Chung et al study should send to other would-be clinical myopia researchers is that randomization and masking are good but too little information about the participants can still render even the most well-controlled study useless in terms of clinical applicability.
As a final aside, the New Scientist article quotes the principal investigator as saying: “The study was meant to run for three years but after two years, when we found out we were making the children’s eyes worse, we had to stop it prematurely”. This statement seems at odds with the statement in the methods section of the Vision Research paper that: “results were only analyzed after the last reading of the last patient was collected.” What really did happen?
Chung K, Mohidin N, O'Leary DJ. Undercorrection of myopia enhances rather than inhibits myopia progression. Vision Res. 2002, 42: 2555-9.
Smith EL 3rd, Hung LF. The role of optical defocus in regulating refractive development in infant monkeys. Vision Res. 1999, 39: 1415-35.
Progressive addition lenses (PALs) should not be prescribed as a myopia-control treatment
Evidence from The COMET (Correction of Myopia Evaluation Trial) study – A randomized clinical trial of progressive addition lenses (PALs) for control of the myopia in children
The long awaited data from this multi-center US-based study, funded by the National Eye Institute (NEI), was published earlier this year in the prestigious journal, Investigative Ophthalmology and Visual Science (2003, 44: 1492-1500). The results of this study were disappointing, adding to an accumulating number of studies reporting only limited benefit from variable focus lenses. Indeed, around the same time as the paper appeared, NEI released a statement covering the study, indicating that progressive addition lenses should not be prescribed as a myopia-control treatment.
What is the principle behind prescribing variable focus spectacle lenses for myopia control and why might the results of clinical trials not bear out the expected benefit of such lenses? Three key factors have driven the use of these lenses for myopia control: 1. accumulating data linking myopia with excessive near work; 2. observed association between high accommodative lags and progressing myopia; and 3. increased eye growth, as characteristic of myopia, in animals exposed to negative defocusing lenses. Both accommodative lags and negative lenses imposed on otherwise normal eyes, create conditions of hyperopic defocus where the true image plane lies behind the retina (the seeing layer of the eye). The ocular growth response to defocusing lenses reported in animal studies appears to be compensatory, reducing the amount of defocus experienced by the eye with the defocusing lens in place. It is conceivable that a similar growth response could be triggered in humans by accommodative errors during excessive near work. The resulting myopia, if left uncorrected, should lessen the demand on accommodation and thus accommodative lag. However, such changes are typically corrected with an updated prescription, thereby reintroducing the problem triggering the growth response.
The main outcome of the COMET study was a small but significant slowing of myopia progression with PAL lenses that was limited to the first of the 3 years of the clinical trial. The 3-year difference in progression between participants wearing PALs compared to those wearing regular (single vision; SV) lenses was 0.20 +/-0.08 D. The study cannot be faulted in terms of its design – it is a large (469 participants) randomized double-masked, multi-center (4 sites), ethnically diverse study with an excellent record of retention and masking throughout. None-the-less there are weaknesses. All children in the PAL group were prescribed the same +2 D lens addition based on an earlier report (Leung & Brown, 1999) that +2 D additions were more effective than +1.5 D additions in controlling progression of myopia in adolescents. It could be argued that lenses intended to correct near focusing errors should be designed on an individual basis. A second weakness of the study is the failure to verify that the children used the near addition during close work although the lenses were fitted in the spectacle frames in a way to encourage reading through the near addition. In general children have little incentive to use the near addition of their PALs; indeed, the distortions occurring in the lower periphery of theses lenses might act as a disincentive. In contrast, adults who are typically prescribed them to compensate for failing near focusing ability have the incentive of clear near vision to use the lower portion of their lenses. It is of potential significance that the study reporting the most promising outcome using PALs also involved adolescents who may have been better able to follow instructions in the use of their lenses.
Although the overall trend of less than 0.50 D difference in progression in myopia between the PAL and SV groups after 3 years can not justify the wide-spread prescribing of PAL lenses for myopia control, a closer inspection of the COMET data provides some support for their more limited prescribing. Specifically, those with low myopia (<2.25 D initially) and low accommodation (<2.57 D in respond to a 3 D demand), showed greater benefit from PALs (0.55 D treatment effect for those showing both low myopia and low accommodation). It is of interest to know whether or not low accommodation was associated with a tendency to overconverge at near. There is suggestion from the COMET data that those exhibiting near esophoria and orthophoria showed more benefit from PALs compared to those with near exophoria although the differences were not statistically significant. Nonetheless this trend is consistent with other studies showing greater treatment effects in near esophoric participants (e.g. Grosvenor et al, 1987; Goss & Grosvenor, 1990; Fulk et al, 2000). That such participants would show greater accommodative lags through their normal SV corrections is also predictable.
Where to now? The COMET study provides convincing evidence that PALs are not for everyone. It might also be argued that spectacles are not the optimum form of correction for proving near additions to young children because of their more limited understanding of the issues involved. A recent presentation at ARVO (a international vision conference held annually in Florida), tended to confirm this by showing that during reading, children rarely use the optimum part of the lenses for near work (Nakatsuka et al., 2004). Dr Tom Aller OD (San Bruno, California) has reported promising retrospective data involving variable focus soft contact lenses as a myopia control treatment. Dr Aller recently received a research grant to fund a year long randomized prospective clinical trial of variable focus soft contact lenses. The project is being sponsored by a major contact lens company. Forty-five children already have been recruited into the study. The target number is around 100 children. If you have a child who is showing rapid increases in their myopia, and are interested to know more about this study, please click here. This study was also the subject of a feature article in Fremont Argus, a local Northern California newspaper. Look out for follow-up data from this clinical trial!
As a final aside, the report on the COMET study makes the curious observation that the treatment effect was limited to the first year of the 3 years of the study. Overall, there were no differences in progression between the PAL and SV groups across the 2nd and 3rd years of the study. The authors of the report put forward a number of possible explanations for the short-term nature of the treatment effect, including the possibility that a single environmentally-based treatment intervention may have only a limited capacity to restrain progression that is influenced by additional visual and genetic factors as well. As studies involving drug intervention (Shih et al, 2001, Chua et al, 2002), and contact lenses (Khoo et al, 1999) for myopia control have reported similar time constraints related to treatment efficacy, this issue suggests a major reevaluation of the hypotheses underlying various treatment strategies.
Aller T, Grisham, D. Myopia control with bifocal contact lenses, Optom Vis Sci (Suppl), 2000, 77: 187.
Aller T. Myopia progression with bifocal soft contact lenses – A twin study Optom Vis Sci (Suppl), 2002, 79: 179.
Chua W-H, Balakrishnan V, Tan D, Chan Y-H, ATOM Study Group. Efficacy results in the treatment of myopia (ATOM) study. ARVO 2003, Abstract#3119.
Fulk GW, Cyert LA, Parker DE. A randomized trial of the effect of single-vision vs. bifocal lenses on myopia progression in children with esophoria. Optom Vis Sci. 2000, 77: 395-401.
Goss DA, Grosvenor T. Rates of childhood myopia progression with bifocals as a function of nearpoint phoria: consistency of three studies. Optom Vis Sci. 1990. 67: 637-40.
Grosvenor T, Perrigin DM, Perrigin J, Maslovitz B. Houston Myopia Control Study: a randomized clinical trial. Part II. Final report by the patient care team. Am J Optom Physiol Opt. 1987, 64: 482-98.
Gwiazda J, Hyman L, Hussein M, Everett D, Norton TT, Kurtz D, Leske MC, Manny R, Marsh-Tootle W, Scheiman M. A randomized clinical trial of progressive addition lenses versus single vision lenses on the progression of myopia in children. Invest Ophthalmol Vis Sci. 2003, 44: 1492-500.
Khoo CY, Chong J, Rajan U. A 3-year study on the effect of RGP contact lenses on myopic children. Singapore Med J. 1999, 40: 230-7.
Leung JT, Brown B. Progression of myopia in Hong Kong Chinese schoolchildren is slowed by wearing progressive lenses. Optom Vis Sci. 1999, 76: 346-54.
Nakatsuka C, Hasebe S, Nonaka F, Ohtsuki H. Assessment of downward deviation of progressive addition lenses in a myopia control study. Invest Ophthalmol Vis Sci 2004, (ARVO abstract #2732).
Shih YF, Hsiao CK, Chen CJ, Chang CW, Hung PT, Lin LL. An intervention trial on efficacy of atropine and multi-focal glasses in controlling myopic progression. Acta Ophthalmol Scand. 2001, 79: 233-6.
Wildsoet CF. Active emmetropization – evidence for its existence and ramifications for clinical practice. Ophthalmic Physiol Opt 1997, 17: 279-290.