The late 1980s was a most exciting period for infant vision research. For the first time it became possible to measure—simply and accurately—the vision of babies and young children. Dobson was at the forefront of this activity and the vision scientist most responsible for its incorporation into clinical practice and ophthalmic research. Much of the work we review here draws on the rigorous approach pioneered by Dobson whose legacy is that the quantitative assessment of vision, rather than observation of the eye alone, now lies at the heart of our approach to the management of children with eye disease.
Even up to and beyond the time period referred to above, it was firmly held that the correction of refractive error played only a minor role in the clinical management of amblyopia. Indeed, reference to the failure of refractive correction to ameliorate amblyopia often features in textbook definitions of the condition,1 which generally fails to distinguish between immediate and gradual effects of spectacle or contact lens wear.
Sitting uneasily with this viewpoint was the observation that amblyopic children would often not obtain their best “pre” treatment visual acuity until a refractive correction had been in place for some time. Although this intervention—commonly referred to as spectacle adaptation—had formed a component of the treatment for anisometropic ambyopia since at least the middle of the last century,2 it was seen only as an adjunct to treatments such as penalization or occlusion, rather than actively therapeutic in its own right.
It was not until 2002 that any attempt was made to quantify the effect that refractive correction might have on amblyopia. Up until this point there was no evidence as to what proportion or what type of patient might benefit from this intervention or to what extent and what timescale. In a short report published in Ophthalmic and Physiological Optics,3 we analyzed the improvement in visual acuity subsequent to the correction of refractive error in a small group (n = 13) of typically presenting amblyopic children and were greatly surprised by the magnitude of improvement [0.1 to 0.5 logMAR (minimum angle of resolution) units], which occurred among these mostly refractive amblyopes. Further, refractive correction was seen to benefit all the children. Time to best acuity ranged from 4 to 24 weeks (Fig. 1). Control subjects who did not undergo the repeated (weekly) testing of the experimental group but who otherwise were treated identically also showed comparable acuity gains, essentially ruling out practice or training effects as a more parsimonious explanation of the improvements seen. Further, the time period over which improvements occurred was not so prolonged that visual maturation could be considered a significantly contributing factor. Perplexing to us and others was the finding that one of our amblyopic participants who had both anisometropia and esotropia with their amblyopia demonstrated a gain in acuity (0.28 logMAR) that was on a par with the straight-eyed study participants. Subsequently, when similar observations were again reported by ourselves4 and by others,5 this became less controversial; we shall return to this issue later.
Although “refractive adaptation”—as the improvement in acuity attributable to the correction of refractive error became known—did not immediately impact on clinical practice, it soon gained significance in the context of clinical trial design. A much-publicized systematic review of amblyopia6 treatment had, by this time, provided an impetus for a more evidence-based approach to amblyopia therapy, resulting in the design of dose response and controlled clinical trials. Crucially, it was now considered necessary when attempting to evaluate the effectiveness of, say, a regimen of occlusion, that the study account for the effects of prior refractive adaptation. The first such study we conducted—the Monitored Occlusion Treatment for Amblyopia Study (MOTAS7)—put this important aspect of study design into practice. It comprised three phases. In the first (“baseline”) phase, participants were clinically assessed and stable baseline visual performance established. In the second, participants underwent a period of refractive adaptation until we were reasonably certain that all improvement attributable to this process would have occurred (the duration of this phase—18 weeks—was based on our previous study,3 where no gains in acuity exceeding 0.1 logMAR occurred beyond this period). Only at this point did phase 3 begin, within which occlusion was prescribed.
MOTAS, it could be said, put the concept of refractive adaptation firmly on the map, such that we felt that the findings of the refractive adaptation phase merited publication in their own right.4 Sixty-five children [mean (standard deviation) age = 51(1.4 years)] were enrolled, of which just under half were anisometropic and strabismic, and the remainder either anisometropic or strabismic in roughly equal proportions. LogMAR visual acuity improved on an average by 0.24 log unit (range: 0.00 to 0.60) over the 18-week adaptation phase (Fig. 2). The improvement was not seen to differ as a function of age or type of amblyopia. Fourteen children (22%) improved to such an extent during refractive adaptation that they became ineligible to proceed to the final phase of the study in which they would have been prescribed occlusion. This outcome led us to ponder the fact that had these children undergone routine clinical management as practiced at that time, they would likely as not, have undergone a quite unwarranted period of occlusion therapy. Where fellow eyes had significant refractive errors (≥1.75 D), their acuity was also observed to improve, on an average, by 0.1 log unit. However, it was unclear to what extent such gains can be interpreted as arising from the remediation of bilateral amblyopia or from the simple optical benefits of refractive correction.
In 2006, our MOTAS findings were corroborated and extended by the Pediatric Eye Disease Investigator Group (PEDIG).8 In their study, subjects with anisometropic amblyopia (n = 84) initially underwent between 5 and 30 weeks of refractive correction during which their mean minimum angle of resolution reduced by almost half (0.29 logMAR gain). This corresponded to an improvement of ≥0.2 logMAR and ≥0.3 logMAR in, respectively, 77 and 60% of participants. Resolution of amblyopia occurred in 23 (27%) children. There was no apparent effect of age on the improvement of acuity seen but poorer initial acuity and greater degrees of anisometropia decreased the likelihood of resolution of amblyopia. Of note is that this study used a more stringent inclusion criterion for the definition of amblyopia (≥0.2 log unit intra-ocular difference) compared with that of MOTAS (≥0.1 log unit intra-ocular difference). Given the outcome of the PEDIG study, it now seems unlikely that the findings of MOTAS could have been accounted for by the inclusion of some non-amblyopic ametropes, whose acuity gains arose solely from the optical benefits of refractive correction.
Our most recently conducted trial of occlusion therapy (ROTAS9) provided further confirmation that refractive adaptation in itself constitutes a robust treatment for amblyopia.a Forty-four children undergoing 18 weeks of spectacle wear before scheduled occlusion gained, on an average, 0.22 logMAR unit of acuity. Statistical power constraints did not permit a secondary analysis by age and amblyopia type.
REFRACTIVE CORRECTION IN BILATERAL REFRACTIVE AMBLYOPIA
Where the refractive error is symmetrical and bilateral, refractive correction is an established and uncontroversial treatment. The improvements in acuity seen are of the greatest magnitude reported for all types of amblyopia. For example, in this category of patients, PEDIG5 observed a mean (95% confidence interval) improvement in the binocular logMAR acuity of 113, 3 to 10-year-old children of 0.39 (0.35 to 0.41) log unit.
REFRACTIVE CORRECTION IN NON–STRAIGHT-EYED AMBLYOPIA
That refractive correction has now been convincingly established to be beneficial,10 where the primary clinical association of the amblyopia is refractive error had, as already mentioned in the Introduction section, is highlighted in the clinical literature. However, our findings in 2004 (and to a limited extent in 2002) that even in the absence of significant anisometropia and in the presence of a constant strabismus, refractive correction still appeared to exert an ameliorating effect on the amblyopia3,4 was not a readily predictable finding, or as one comment in the literature put it, “… somewhat surprising ….” 5 However, an analysis of the changes in acuity occurring during a refractive correction “run-in” phase to a randomized trial of occlusion therapy5 again revealed improvements entirely comparable (i.e., ≥0.2 logMAR) with those seen among straight-eyed amblyopes. In the case of MOTAS, although a small amount of this improvement could be accounted for by some subjects (n = 7, 20%) having small angle strabismus and rudimentary binocular vision, or strabismus of a refractive nature, the majority of those categorized as strabismic amblyopes had large angle strabismus without demonstrable binocular vision at the start or end of the treatment.
NEUROPHYSIOLOGICAL BASIS OF TREATMENT EFFECT
To date, most research that has examined the effect of refractive correction on amblyopia has predominantly concerned itself with establishing the magnitude and time course of the treatment gains, and to a lesser extent with the categories of amblyopic patients who might benefit. It does not appear to be an artifact of repeated testing,3 and clearly, the time course of the improvement in vision is incongruent with any explanation based on simple optics. Unfortunately, we know of no experimental models that mimic or manipulate the effects of refractive adaptation. However, it might prove insightful to search for clues among the considerable body of evidence arising from experimentally induced ocular deprivation? After all, refractive correction can, at one level, simply be viewed as a more subtle alteration of spatial visual input in comparison with the typically gross manipulation (i.e., total provision or total elimination) used in ocular deprivation studies. Consider, for example, recent findings that have highlighted the role of (non-competitive) binocular experience in reversing the effects of monocular deprivation. The initial experimental manipulation undertaken by Mitchell and Gingrass11 involved classical monocular deprivation of a cat eye by eyelid suture. After 6 days, the eye was opened with no attempt to eliminate the competitive advantage of the previously non-deprived eye (no reverse occlusion). Subsequent recording of grating acuity (jumping stand paradigm) for up to 6 weeks showed an orderly recovery in acuity to around 5 c/deg in the formally deprived eye with the acuity of the non-deprived eye reaching 7.05 c/deg (within normal range). Indeed, analogous findings have been reported in human infants, where the restoration of binocular visual input on removal of a congenital cataract facilitates a rapid improvement in visual acuity.12 However, at this juncture it is important to compare the experimental model and human research. In the former, it seems that restoration of acuity occurs only when the binocular visual input is correlated.13 This is very different from the human situation in which refractive correction improves acuity in amblyopic children without and with strabismus: the latter being a de facto clinical example of uncorrelated visual input. However, recent studies14,15 have shown that, contrary to our previous understanding, subjects with strabismic amblyopia have mechanisms that use and combine information from the affected and fellow eye, and they can both drive binocular cells in visual cortical area V1. Although this was observed under exacting experimental conditions with stimulus presentation at corresponding retinal locations, it at least hints at the possibility that fellow eye suppression may not result in complete inhibition of visual input modified by refractive correction.
A TREATMENT PROTOCOL
The evidence base suggests that the majority of children with refractive error and amblyopia will benefit from a period of refractive adaptation, negating the need for occlusion in around one quarter to one third of patients.4,8 However, what should the guidelines for refractive adaptation be and what is the present uptake of clinicians worldwide to this evidence base? Possible guidelines for refractive adaptation would include a minimum period of full-time spectacle wear of 12 weeks for all children with amblyopia and significant refractive error. Follow-up should be 6 to 8 weeks until substantive gains in visual acuity cease or visual acuity becomes good and equal. In the event that little or no gains in visual acuity are seen after 12 weeks, refractive error should be reassessed and refractive adaptation restarted if significant differences are seen. Subjects showing no improvement in visual acuity or difference in refractions should commence occlusion.4,8
Informal observation and feedback suggest that many orthoptists in the United Kingdom and Europe have implemented protocols for prescribing optical treatment to children with amblyopia,16 and it is our understanding that likewise in the United States, clinical practice is now moving in a similar direction.
In this review, we have examined the emerging evidence that refractive correction can, over a period of time, significantly reduce the acuity deficit in the most commonly presenting types of amblyopia. Such findings directly impinge on the way in which we should manage our patients and the expectations of outcome we can offer to carers. Although the neural mechanisms underpinning optical treatment remain obscure (particularly so in the case of strabismic amblyopia), we have highlighted some areas of research that we consider insightful. Hopefully, the increasing implementation of optical treatment as a component of clinical management should provide an impetus to elucidating the basis of this important, but hitherto neglected, treatment modality.
This work was supported by the Guide Dogs for the Blind Association and Fight for Sight, UK.
Department of Optometry and Visual Science
London EC1V 0HB, United Kingdom
1. Rutstein RP, Daum KM. Anomalies of Binocular Vision: Diagnosis and Management. St. Louis: Mosby; 1998.
2. Bishop JW. Treatment of amblyopia secondary to anisometropia. Br Orthopt J 1957;14:68–74.
3. Moseley MJ, Neufeld M, McCarry B, Charnock A, McNamara R, Rice T, Fielder A. Remediation of refractive amblyopia by optical correction alone. Ophthal Physiol Opt 2002;22:296–9.
4. Stewart CE, Moseley MJ, Fielder AR, Stephens DA. Refractive adaptation in amblyopia: quantification of effect and implications for practice. Br J Ophthalmol 2004;88:1552–6.
5. Cotter SA, Edwards AR, Arnold RW, Astle WF, Barnhardt CN, Beck RW, Birch EE, Donahue SP, Everett DF, Felius J, Holmes JM, Kraker RT, Melia BM, Repka MX, Wallace DK, Weise KK. Treatment of strabismic amblyopia with refractive correction. Am J Ophthalmol 2007;143:1060–3.
6. Snowdon SK, Stewart-Brown SL. Preschool vision screening. Health Technol Assess 1997;1:1–83.
7. Stewart CE, Fielder AR, Stephens DA, Moseley MJ. Design of the monitored occlusion treatment of amblyopia study (MOTAS). Br J Ophthalmol 2002;86:915–9.
8. Cotter SA, Edwards AR, Wallace DK, Beck RW, Arnold RW, Astle WF, Barnhardt CN, Birch EE, Donahue SP, Everett DF, Felius J, Holmes JM, Kraker RT, Melia M, Repka MX, Sala NA, Silbert DI, Weise KK. Treatment of anisometropic amblyopia in children with refractive correction. Ophthalmology 2006;113:895–903.
9. Stewart CE, Stephens DA, Fielder AR, Moseley MJ; ROTAS Cooperative. Objectively monitored patching regimens for treatment of amblyopia: randomised trial. BMJ 2007;335:707.
10. Shotton K, Powell C, Voros G, Hatt SR. Interventions for unilateral refractive amblyopia. Cochrane Database Syst Rev 2008;8:CD005137.
11. Mitchell DE, Gingras G. Visual recovery after monocular deprivation is driven by absolute, rather than relative, visually evoked activity levels. Curr Biol 1998;8:1179–82.
12. Maurer D, Lewis TL, Brent HP, Levin AV. Rapid improvement in the acuity of infants after visual input. Science 1999;286:108–10.
13. Kind PC, Mitchell DE, Ahmed B, Blakemore C, Bonhoeffer T, Sengpiel F. Correlated binocular activity guides recovery from monocular deprivation. Nature 2002;416:430–3.
14. Baker DH, Meese TS, Mansouri B, Hess RF. Binocular summation of contrast remains intact in strabismic amblyopia. Invest Ophthalmol Vis Sci 2007;48:5332–8.
15. Baker DH, Meese TS, Hess RF. Contrast masking in strabismic amblyopia: attenuation, noise, interocular suppression and binocular summation. Vision Res 2008;48:1625–40.
16. Newsham D. The effect of recent amblyopia research on current practice in the UK. Invest Ophthlamol Vis Sci 2008;49:E-Abstract 2585.
aIndeed PEDIG had already adopted the term “optical treatment of amblyopia.”8