Amblyopia, or “lazy eye,” is the most common cause of monocular visual impairment in both children and young adults. Amblyopia occurs when a disruption in the processing of visual information during a critical period of visual development leads to a unilateral or, less commonly, bilateral decline in best-corrected visual acuity (BCVA). The reduced vision in an amblyopic eye is unable to be improved with glasses, surgery or other interventions. Importantly, the diminished visual acuity (VA) cannot be attributed to any structural abnormality of the eye or the posterior visual pathway.(1) Rather, the functional connectivity of visual circuits between the eyes and the brain fails to develop normally as a consequence of altered visual input from various causes.

Causes: Three Mechanisms

Three general mechanisms exist whereby a disruption in visual processing during the critical period can cause amblyopia: form deprivation, uncorrected refractive errors and strabismus.

Deprivation amblyopia results as a consequence of obstruction of the image in the visual axis, blocking a clear image from reaching the fovea in the eye. The blockade may be anywhere along the visual axis, from the lid (ptosis), cornea (Peter’s anomaly), lens (cataract), posterior segment (vitreous hemorrhage), etc. The lack of a formed visual stimulus results in maldevelopment of cortical visual circuits and subsequent amblyopia. It can be unilateral or bilateral.(2)

Similarly, refractive amblyopia results as a consequence of abnormal visual experience due to uncorrected refractive errors. Isometropic amblyopia (from Greek, iso+metros of equal measure; ops, eye) is a bilateral loss of BCVA that occurs as a consequence of uncorrected symmetric refractive error, most typically hyperopia or astigmatism, at young ages. Images from both eyes are degraded resulting in abnormal visual processing. Appropriate refractive correction improves VA significantly but fails to achieve normal BCVA for age. Anisometropia (anisos + metros, unequal measure, ops, eye) is the most common cause of amblyopia. It occurs as a consequence of a difference in refractive error between the two eyes, which leads to a projection of unequal images on the fovea and resultant unilateral blur. If untreated, functional circuits between the eye with the blurred image and the brain fail to form and those from the sound eye expand. Indeed, fMRI studies show that sound eyes occupy more cortical territory in the visual cortex than the amblyopic eye if the defect develops in infancy.(3)

The third type, strabismic amblyopia, is fundamentally different. Here, ocular misalignment results in a different image being cast onto the fovea of each eye. Despite this misalignment, strabismic children are rarely diplopic because the image from the non-dominant eye is actively suppressed in the brain during binocular viewing. Chronic suppression can result in amblyopia. The age of presentation for strabismic amblyopia is typically younger than that of anisometropic amblyopia.(4)

Age and the Critical Period

The role of age in the development of amblyopia is important. Deprivation or interocular competition does not result in amblyopia once visual development has concluded: it is only during a critical period that these deficits are manifest. A critical period describes the period of time during which disruption of normal sensory stimulation results in a permanent functional deficit. Different visual functions mature over many distinct visual critical periods that often overlap, and these have been well delineated in animal models.(5)

Ascertainment of absolute critical periods in human children is less certain, but can be deduced from residual deficits in children with visual correction at defined ages. For example, it is known that binocular function is disrupted by deficits that occur prior to the age of 24 months,(6)(7) while amblyopia can develop up to eight years of age.(8) Thus early intervention is critical for correcting visual deficits to minimize permanent visual dysfunction as a consequence of abnormal visual development.

Visual Pathophysiology

The Nobel Prize winning work by Hubel and Wiesel demonstrated remodeling of afferent projections in ocular dominance columns in primary visual cortex (V1) following visual deprivation in kittens.(9) While their research, which commenced half a century ago, laid the foundation for investigations of the neuroanatomic basis of visual circuit maturation, the neural mechanisms underlying the visual pathophysiology of amblyopia remain an area of active investigation.

Both anatomic and physiologic deficits have been demonstrated following visual deprivation in all mammalian species studied, which clearly shows that amblyopia is a disorder of the brain, primarily of V1. The model that emerges from these collective studies suggests that any deficit (e.g., strabismus, surgical eyelid closure) that causes disruption of coordinated activity of inputs from the two eyes onto target cells in V1 results in binocular suppression with loss of responses from the amblyopic eye.(10) The suppression appears predominantly as a result of the loss of excitatory inputs onto cortical neurons as opposed to through inhibitory circuits.(11) This likely reflects the fact that during normal development cortical neurons with similar response properties make reciprocal connections (i.e., cells that fire together wire together) that are disrupted when interocular activity is de-correlated during abnormal visual experience.(12)

Other studies in non-human primates and in human subjects using psychophysical testing and functional imaging studies have also noted deficits in higher visual cortical areas but the physiology of these abnormalities is not known.(4) Studies in amblyopic adults have shown that a panel of psychophysical tests exploring 11 elements of visual function can discriminate amblyopic etiology based upon performance. For example, deprivation and anisometropic amblyopia share similar response properties whereas strabismic amblyopes are quantitatively different, as might be expected mechanistically. This finding tells us that all amblyopia is not the same and that neural mechanisms yet to be elucidated in each type may differ.(4)

Who gets it? How do we find it?

The prevalence of amblyopia is approximately 2.5% in the general population(13) and does not vary with age. Boys and girls are affected equally. Occurrence of amblyopia varies with ethnicity, with a prevalence of 1.5% among African-American children and 2.6% among Hispanic children.(14) Seventy-five percent of amblyopia cases are attributable to refractive error, with anisometropic amblyopia being the most common subtype. Strabismus is the next most common cause of amblyopia and form-deprivation amblyopia is least common but the most visually devastating.(13)

Consequently, the screening and identification of those at risk for amblyopia is of considerable interest. Screening can be performed by many members of the care taking team, including schools, pediatricians, eye care professionals and automated photo-screening programs. Each of these routes has its pros and cons. Access, efficacy and cost vary with each approach. While mandatory eye exams have been advocated in Missouri, Illinois and Kentucky, this is a contentious requirement, particularly because of the excessive cost and poor compliance with obtaining the exams. (According to the Missouri Children’s Vision Commission, <40% of children required to obtain mandatory eye exams in Missouri prior to kindergarten complied with this mandate during the years 2008-2010.(15))

For any screening program, the goal is to reach the largest number of children possible and, in the case of amblyopia, at the youngest reasonable age. Access to care is improved with school and pediatrician-based screening; however, compliance with the American Academy of Pediatrics vision screening recommendations is variable and most children do not receive a first screening exam until kindergarten. This leaves little time for intervention and treatment.(16) To address these concerns recent studies have evaluated the efficacy of lay screeners to identify index cases for referral and have shown promise for reducing the cost and expanding access to increase the number of children screened at younger ages.

The Vision in Preschoolers study (VIP) was designed to evaluate the effectiveness of vision screening tests commonly used for preschool-aged children in the hands of eye care professionals and lay testers. This was a three-phase study. Phase I had licensed eye care professionals testing different screening modalities -- non-cycloplegic retinoscopy, autorefraction, photoscreeners, Lea symbols test, HOTV vision test, stereo smile test, cover/uncover test and random dot E. Results indicated that non-cycloplegic refraction, Retinomax autorefraction, SureSight vision screener and Lea symbols are the most sensitive methods.(17) These methods were then used in Phase II in a controlled environment by both trained eye-nurses and lay testers to determine which screening tool was most effective in the hands of the various caregivers.(18) Phase III is currently ongoing and will investigate the effectiveness of the best screening tools in a more general population and setting.

Federal initiatives to expand vision screening for children will increase the number of children identified with amblyopia, strabismus and/or significant refractive error. While the benefits of identifying and treating these conditions are undeniable for an individual, the cost to society for the screening and identification of that one child is a relevant concern. The VIP studies were able to show that lay testers, armed with the right screening tools, were effective. With cheaper and more reliable screening tools, the utility balance is easily tipped in favor of screening – both more often and younger. New directions in screening include home-based screenings. In one study, home-based amblyopia screening packages were distributed to parents through the kindergartens. Children were later examined by an eye care professional. Sensitivity of the screening package was found to be 80%. Specificity of the screening package is 94.1%.(19) The producing cost for each package was $1.17, and it took just $266.00 to screen out an amblyopic child.

Treatment: optical correction alone, patching, atropine

Occlusion therapy with patching of the sound eye has long been the mainstay of amblyopia treatment, yet there was no decisive consensus for duration of occlusion or length of treatment. Furthermore, some physicians advocated atropine penalization and other alternative treatments despite limited data to assess efficacy. The Pediatric Eye Disease Investigator Group (PEDIG), a collaborative network dedicated to facilitating multicenter clinical research in strabismus, amblyopia and other eye disorders that affect children, initiated the first of several Amblyopia Treatment Studies (ATS) to examine the many questions surrounding the treatment of amblyopia.

Amblyopia can be classified both by type and by severity. Classification by severity is utilized for purposes of stratifying patients to assess treatment efficacy. Moderate amblyopia is defined in the ATS as BCVA between 20/40 – 20/80, while BCVA between 20/100 - >20/400 is classified as severe amblyopia. BCVA better than 20/40 is considered mild amblyopia.

The initial ATS study (ATS 01) examined patching versus atropine penalization for moderate amblyopia.(20) Children ages 3-7 with moderate amblyopia were randomly assigned to patching or atropine. Improvement in BCVA was initially faster in the patching group, but equalized after six months. Atropine was found to be both cheaper and easier, although it was associated with a transient decrease in vision in the good eye.

Severe amblyopia was treated with either full-time or part-time patching in ATS 02A.(21) Full-time patching is defined as patched for all waking hours, part-time patching occurs for six hours per day. After four months, treatments were equal with an average improvement of >4.5 lines. A greater treatment effect was seen with worse baseline VA.

The use of two versus six hours of patching for moderate amblyopia was examined in ATS 02B.(22) After four months, BCVA in the amblyopic eye improved a similar amount in both groups. ATS 04 looked at the frequency (weekend versus daily) of atropine administration in moderate amblyopia.(23) Weekend atropine provides an improvement in VA similar to that provided by daily atropine with both groups improving an average of 2.3 lines from baseline to 17 weeks, but compliance was better in the daily atropine group.

Another significant question raised was that of the recurrence rate of amblyopia once treatment was discontinued. ATS 02C demonstrated that regression was seen in 25% of patients off atropine or two-hour patching, while 42% of patients using six hours of patching regressed. The study authors concluded that prior to cessation patching should be weaned from six to two hours in patients with initially longer patching regimens.

The Monitored Occlusion Treatment of Amblyopia Study examined spectacle correction alone versus two hours of patching for strabismic or anisometropic amblyopia in children ages 3-7. This study demonstrated that improvement in VA can be achieved with eyeglasses alone in both strabismic and anisometropic amblyopia.(24)(25) BCVA of 20/25 or better was seen in 74% of children treated with patching and spectacle correction compared to 43% of children treated with spectacles alone. Similar results were found for both moderate and severe amblyopia.

Who gets better?

Numerous factors contribute to vision outcomes in the treatment of amblyopia such as cause of amblyopia, initial visual acuity or severity of amblyopia, age of the child and treatment strategy.

While all forms of amblyopia can cause reduced vision in the affected eye, the prognosis and challenges of treatment differ. Deprivation amblyopia requires prompt treatment of whatever is impeding visual input, followed by intensive amblyopia treatment. Otherwise, these eyes have very poor visual potential. Refractive amblyopia is typically discovered at older ages because the child’s eyes are not obviously misaligned and the non-amblyopic eye has normal vision limiting the time for intervention and correction.

The initial vision in the amblyopic eye has been suggested to impact the amount of visual improvement seen with the initial treatment. Repka and Ray reported on 79 patients treated with atropine; more improvement was seen in eyes with worse initial vision. They stratified patients on the basis of initial amblyopic eye acuity (20/100 or worse, 20/80 - 20/50, and 20/40 or better), the 20/100 or worse group improved the most.(26) This suggests that even those eyes with severe baseline loss of vision from amblyopia have the greatest potential for improvement with treatment.

The age of the child at the initiation of amblyopia treatment is quite significant. A meta-analysis of outcome data from four randomized clinical trials confirmed that a greater gain in visual acuity is seen with the treatment of younger children. ATS 03 and 09 showed children over the age of seven were less responsive than younger children to amblyopia therapy. Among the older children, treatment was least effective when there was a history of prior amblyopia treatment or severe amblyopia. In eyes with severe amblyopia, children 3-<5 were somewhat more responsive to either six hours/day of patching or full-time patching than were even slightly older children (5-<7).(25)(27)

ATS 03 generated the significant and surprising result that up to one-quarter of children over the age of seven will respond to treatment. Among children 7-17, 25% had improved visual acuity with spectacle correction alone. This indicates amblyopia responds to treatment even after age seven, challenging the belief held by many eye care professionals that intervention beyond a certain age is ineffective. Fifty percent of children 7-12 saw an improvement in visual acuity with the addition of 2-6 hours of patching +/- atropine.(28) This was true even if the amblyopia had been previously treated. Even among children 13-17, 25% of the treatment group and 23% of the spectacles-only controls had an amblyopic eye that improved ≥2 lines of vision. However, 50% of older children without a prior history of amblyopia treatment showed significant gains in visual acuity with optimal optical correction and two to six hours of patching.(25) ATS 03 removed the previously imposed “statute of limitations” on the treatment of amblyopia, clearing the way for further research confirming atropine and patching achieve similar results among 7-12 year olds with unilateral amblyopia.(29)

Can adults with amblyopia improve their vision?

An area of ongoing interest and investigation is whether vision can improve in adult patients with amblyopia. The fact that older children without a prior treatment history may continue to show improvement suggests there may be hope outside of the critical period. Patients with strabismic amblyopia may show a “slippage” of visual acuity in the amblyopic eye long after early childhood. This acuity loss can be reversed with occlusion of the non-amblyopic eye even in adults, although it may subsequently “slip” again with cessation of occlusion therapy.(30)

Recently, investigation into residual plasticity of the adult brain supports the idea that intervention well outside of the critical period may improve vision in an amblyopic eye. One concept, perceptual learning, the act of repeatedly practicing a challenging visual task, appears to generate substantial and lasting improvement in visual performance. In the normal visual system, the improvements remain tightly linked to the specific training stimulus. However, in the amblyopic visual system, learned improvements have been shown to generalize to novel tasks. This may translate to improvement in visual acuity.(31)

Newer novel treatments: opaque CTL, refractive surgery (PRK, phakic IOLs)

Occlusion treatment for strabismus was first described in 1722.(32) Given the challenges, social stigma and compliance issues associated with both patching and atropinization, alternative therapies are of considerable interest.

Opaque contact lenses (CTL) are hydrogel CTLs designed for extended wear and significant reduction in visual stimuli. The efficacy of opaque CTLs was examined in a population of children who were patch-intolerant and had failed conventional treatment. In this prospective study, 92% of patients improved at least one line of visual acuity. However, 11/25 patients had amblyopia rapidly return to a reduced level when the CTL was discontinued.(33) A retrospective chart review of the use of opaque CTLs in children >5 showed they are an effective alternative to patching, although management issues related to lens hygiene and handling exist.(34) Nonetheless, they remain a treatment option for select patients and allow for the avoidance of patching, a treatment which does place a child at a greater risk for teasing and the social stigma associated with strabismus.(27)

Another novel treatment is the sutured occluder, a translucent sew-on occluder that can be fixed to the orbital rim for a month of amblyopia therapy.(35) The occluders are manufactured from heat-moldable plastic and reduce the VA of the sound eye from 20/20 to 20/400 but are not so opaque as to prevent examination of the eye. This technique was examined in patients with severe amblyopia and a combination of anisometropia and strabismus. The primary outcome was shield tolerance by patients and their families, which was achieved by 100% of families. The secondary outcome was impact of this therapy on VA. The amblyopic eyes demonstrated significant improvement from initial mean visual acuity, 20/119, to a mean visual acuity, 20/57.

Another strategy to surgical occlusion is that of silicone-eyelid closure.(36) In this method, the good eye is closed by passing one limb of a double-armed suture from one eyelid margin through a silicone sleeve and through the corresponding eyelid margin. The other limb is passed behind the silicone sleeve and then through the opposite eyelid margin to form a barrier between the sleeve and the cornea. Two to four weeks later, the suture is cut and the silicone sleeve removed. This method had promising results in severe amblyopes: pre-occlusion, average VA was counting fingers at 3 m; at the end of occlusion, 12/15 patients improved to an average VA of 20/67. VA in the non-amblyopic eye remained 20/20, which is an important finding with full-time closure.

An entirely distinct strategy for correction of amblyopia is that of pediatric refractive surgery. Two types of pediatric refractive surgery are intraocular refractive surgery with implantation of intraocular lenses (IOL) and corneal refractive surgery for patients who have been noncompliant with spectacle correction or contact lens therapy. Spectacle correction is not an ideal therapy in these children because lenses in the spectacle plane act to magnify (hyperopic or plus lenses) or minify (myopic or minus lenses) images in addition to their role in focusing on the retina, Differences between the prescription of two eyes of greater than 3.0 prism diopters causes a discrepancy in the size of an image perceived by each eye (aniseikonia) that causes a secondary loss of sensory binocular fusion.

Pediatric laser refractive surgery was first reported in 1995.(37) In most cases of refractive amblyopia, corrective spectacles or lenses, patching and pharmacologic penalization of the sound eye are effective. However, these strategies have been proven ineffective in a subset of children, in particular those with comorbid neurobehavioral issues or who are noncompliant with therapy.(38)(39) Photorefractive keratectomy (PRK), laser epithelial keratomileusis (LASEK) and laser-assisted in situ keratomileusis (LASIK) have all been used in the pediatric population.(40)(41)(42)(43)

Surface ablation techniques include LASEK and PRK. Studies have shown improvement in both VA and stereopsis (i.e., depth perception—which requires visual input from both eyes) with LASEK.(35) The procedure was well tolerated with minimal complications. Another study of both PRK and LASEK to correct ametropia in a cohort of anisometropic children showed significant gains in VA and binocularity. Over 29 months of follow-up, VA improved in 97% of the children, correction within 2.0 D of the target refraction was achieved in all eyes. An impressive 69% gained binocular vision. Complications frequently encountered with surface ablation procedures include myopic regression and corneal haze, which are adequately managed with frequent use of steroid drops.(36)

LASIK employs the creation of a corneal flap and a laser for remodeling of the underlying corneal stroma. Advantages over surface ablation include faster recovery and less post-operative discomfort. Good results are seen in children with myopic anisometropic amblyopia not responding to traditional treatment, with a reported 93.5% reduction in myopic anisometropia.(37) LASIK does have more associated complications, corneal haze still being most common; others include persistent haze from a free flap, flap striae, deposits on the interface, epithelial in-growth and wrinkles in Bowman’s membrane.(33) Results have suggested pre- and post-operative visual outcomes were similar in LASIK and surface ablation procedures. The best visual outcomes were observed in myopic eyes.(38)

Implantation of IOLs, either following lensectomy or as a phakic IOL, is another form of refractive surgery with impressive results in the treatment of amblyopia.(39)(40) Eighty-five percent of patients were observed to have improved functional vision with significantly improved uncorrected VA.(41)

Secondary IOLs have been utilized in highly anisometropic myopic (>−8.0 D) and high myopic ametropia (>-10.00 D) patients who are spectacle or contact lens intolerant.(44) Because of the object minification induced by minus lenses, highly myopic spectacle correction results in loss of best corrected visual acuity. Among lens options, iris-fixated anterior chamber IOLs seem to be the preferred choice.(45) Following secondary IOL implantation, significant improvement (>6 lines) in VA and improved stereoacuity were seen. Parents reported improved physical coordination and reading and writing skills. An unknown outcome in secondary IOLs is the potential for endothelial cell loss and is an area of ongoing investigation.

Consequences of monocular vision loss: health of the “good” eye, psychosocial impact, reading

Amblyopia is not just a childhood condition; the impact of having decreased vision secondary to amblyopia extends into adulthood. Amblyopia is a significant public health problem. It is the leading cause of monocular vision loss in young and middle-aged Americans.(46) As discussed above, early detection and initiation of treatment remains the most effective way to prevent decreased VA or impaired binocularity. People with impaired vision, even if the impairment is unilateral, are more likely to have an unskilled manual labor job and to be unable to work because of permanent illnesses. And worsening distance acuity further negatively impacts social and economic outcomes.(47) These people have poorer fine motor skills compared with control subjects secondary to their impaired binocularity. Amblyopia may even be associated with a borderline significant decreased rate of completion of university degree.(48) However, this finding has not been validated by other studies.(49)

While the majority of amblyopes have excellent vision in their better seeing eye, they are at greater risk of lifetime visual impairment than the general population. People with an amblyopic eye have nearly three times the risk of visual impairment in the better-seeing eye compared with those without amblyopia.(48) Young adults have an increased risk of a traumatic eye injury in their better seeing eye and older adults have greater rates of visually disabling age-related macular degeneration. Most state agencies define visual impairment as best-corrected vision of 20/60 or worse in the better-seeing eye, which allows an individual access to special services and accommodations. Detection and treatment of amblyopia, therefore, can prevent blindness and its additional cost to society. This benchmark is fairly low: identifying and treating just 12.5 individuals with amblyopia allows for the successful prevention of one case of bilateral visual impairment.(50)

The increased risk of bilateral visual impairment among individuals with amblyopia supports the efforts to screen for and treat amblyopia. Screening is vital to preventing visual loss from amblyopia, especially individuals in the anisometropic subtype, who are often asymptomatic early in the course of the disease. While no randomized trials have demonstrated that screening directly reduces rates of blindness, we know that the maturing visual system does not tolerate delayed treatment of amblyogenic factors and that, therefore, the most successful programs would screen the very young and funnel affected individuals to an eye care professional for diagnosis and management.

This is not to say that formal eye examinations are necessary. In fact, in states where such exams have been mandated, it appears that they result in unnecessary prescription of spectacles, potentially increasing the costs associated with amblyopia screening.(51) Prescription of hyperopic correction for refractive errors not meeting consensus criteria for spectacle wear occurred for 1.8% of children examined by a pediatric ophthalmologist, 11.7% of children examined by comprehensive ophthalmologists and 35.1% of children seen by optometrists. Both optometrists and comprehensive ophthalmologists prescribe spectacles to approximately half of those children who have what most pediatric ophthalmologists believe to be insignificant (<2.00 diopters) hyperopic refractive error. Spectacles were never prescribed by an “experienced” pediatric ophthalmologist to a child having hyperopia of <2.00 diopters.(52) If these results were extrapolated to the entire U.S. population, over $200 million would be spent on unnecessary glasses yearly.

Mandatory comprehensive eye exams have enormous direct and indirect costs and still do not achieve the goal of detecting amblyogenic factors in a majority of children.(46) During the last decade, detection and treatment of amblyopia have received increased emphasis. The VIP studies and PEDIG investigators encouraged the identification and treatment of amblyopia, with particular attention to cost effectiveness and younger age at screening. The cost to society, however, due to unilateral vision loss is minimal.(53) Traditional cost-utility analysis of amblyopia screening in these studies relied on offsetting the socioeconomic cost of bilateral blindness. However, these conclusions fail to take into account the psychosocial impact of amblyopia and the economic hindrances of monocularity.(45)(46)(49)(51) To accurately reflect the true cost benefit of screening, these factors must be acknowledged and greater weight given to the avoidance of amblyopia. Ideally, making screening as inexpensive as possible – through the use of photoscreeners and layperson evaluations, as supported by the VIP study, will both improve our identification and treatment of individual affected children and the cost-utility of screening for the general population.