Author: Claude (AI), from PubMed abstracts · Reviewed by: Dr Mina Abdelmseih · Reviewed: 2026-11-15 · Review due: 2027-11-15

Educational content only. Not personalised medical advice. Clinical decisions remain the responsibility of the treating clinician.

Amblyopia: patching, atropine and binocular treatment

Amblyopia is impaired visual acuity in one or both eyes without a demonstrable abnormality of the visual pathway, and it is not immediately resolved by wearing glasses [S5]. It is the most common visual impairment of childhood [S8] and the leading cause of monocular reduced best-corrected vision in children [S12]. Glasses first, then patching or atropine, remain the evidence-based core of treatment, and treatment works best in young children [S9].

Key points - Correct refractive error first: glasses alone improve acuity with a large effect size in refractive and strabismic amblyopia [S4]. - In children aged 3 to 8 years, starting patching after 3 weeks of glasses was more successful than waiting 18 weeks (67% vs 54% success) in the EuPatch RCT [S8]. - Atropine penalisation appears as effective as patching, costs less and has better adherence, but causes light sensitivity and mild sound-eye blur more often (Cochrane review) [S5]. - Binocular (dichoptic) treatment gave gains similar to patching in the only RCT in a Cochrane review; long-term data are lacking [S7]. - Treatment is mostly effective under 7 years of age and patching should be tapered over at least 5 weeks, not stopped abruptly (Brazilian guideline) [S9].

Evidence last checked: 2026-10-17 (PubMed abstracts).

Red flags

• Reduced vision with a visible ocular abnormality is not amblyopia. By definition amblyopia has no demonstrable abnormality of the visual pathway [S5]; trials required ophthalmic examination including the fundus before diagnosis [S7]. Do not start patching before the eye has been examined.
• Vision loss in the good eye during treatment. Atropine is associated with mild reduction in sound-eye acuity [S5]; reverse or worsening amblyopia was a reported non-serious harm of patching and digital treatment [S10]. Check both eyes at every visit.
• Older age at detection. Treatment is mostly effective under 7 years [S9], and the benefit of glasses falls with age [S4]. Do not delay treatment in a young child.
• Double vision. Double vision was reported as a non-serious harm of treatment [S10].

Recognise it

• Causes (amblyogenic factors): anisometropia, strabismus or both [S7] [S8]. Strabismus develops in approximately 5% to 8% of the general population [S3].
• Deficits beyond acuity: reduced stereoacuity and contrast sensitivity, and perceptual changes even in the sound eye [S9].
• Common working definition in trials: an interocular difference of 0.30 logMAR or more in best-corrected acuity [S8], or amblyopic eye acuity worse than 0.200 logMAR with the fellow eye 0.200 logMAR or better [S7].

Confirm it

• Use an age-appropriate acuity test and measure each eye separately [S7].
• Trials required cycloplegic refraction and an ophthalmic examination including the fundus [S7].
• A consensus decision-support tool from the Pediatric Eye Disease Investigator Group (PEDIG) gives testing methods, pass/refer criteria and monitoring schedules for children aged 3 to 17 years; where direct evidence was lacking its recommendations rest on committee consensus [S12].

Manage it

Before acting on this section, check the national and local guidelines, product information and regulatory approvals that apply where you practise. See the medical disclaimer.

Tier A (primary care / no specialist)

• Identify children with reduced acuity in one eye and refer for refraction and examination: in trials, treatment started only after cycloplegic refraction and an ophthalmic examination including the fundus [S7].
• Support families: adherence limits the effectiveness of treatment [S10].

Tier B (general eye clinic)

1. Glasses first. Prescribe full refractive correction when indicated [S9]. Optical treatment alone gave a large positive effect size of 1.07 on acuity in a meta-analysis; effects were larger with longer wear and better starting acuity [S4]. In one trial, glasses alone improved acuity compared with no treatment (borderline significance) [S1].
2. Do not wait too long before patching. In EuPatch (children 3 to 8 years, intensive patching of 10 hours a day, 6 days a week), early patching after 3 weeks of glasses succeeded in 107 of 159 (67%) vs 86 of 158 (54%) after 18 weeks of glasses [S8]. This conflicts with guidelines that recommend an extended glasses-only phase [S8] and with the meta-analysis favouring optical treatment first [S4].
3. Patching or atropine. Adding occlusion improves vision when amblyopia persists in glasses, in refractive [S1] and strabismic [S3] amblyopia. Atropine penalisation is as effective as occlusion for acuity, with no difference in alignment or stereoacuity; it costs less and gives better adherence and quality of life, but more light sensitivity and mild sound-eye blur, while patching causes more skin, lid or conjunctival irritation [S5]. The Brazilian guideline restricts atropine to hyperopic children whose amblyopic eye sees better under cycloplegia than the fellow eye [S9].
4. Dose. Patching regimens of 2 vs 6 hours (moderate amblyopia) and 6 hours vs full time (severe amblyopia) did not differ significantly in single trials [S1]. Atropine dosing is not covered by these abstracts: follow local formularies.
5. When improvement stops. In children with stable residual amblyopia after at least 12 weeks of 2 hours daily patching, increasing to 6 hours gave 0.6 line more improvement at 10 weeks; 40% vs 18% gained 2 or more lines (PEDIG RCT) [S2].
6. Stopping. Taper patching over at least 5 weeks [S9].

Tier C (specialist centre)

• Combinations. A 2026 network meta-analysis of 22 RCTs found that only full-time patching combined with video games (MD 0.27) and atropine combined with patching (MD 0.25) were significantly better than spectacles; other treatments did not differ significantly [S11].
• Binocular (dichoptic) treatment. A Cochrane review found one RCT (68 children): acuity gain at 16 weeks was similar to patching (0.21 vs 0.24 logMAR), adherence was lower (47% vs 90% achieved over 75% compliance), and no permanent diplopia occurred [S7]. A later phase 3 RCT of a dichoptic digital therapeutic plus glasses vs glasses alone in children aged 4 to 7 years found 1.8 vs 0.8 lines of gain at 12 weeks and was stopped early for success [S6]. These require devices and may not suit low-resource settings.
• Adherence interventions. A 2025 Cochrane review (18 studies) found very low-certainty evidence and could not conclude that behavioural, digital or alternative patching approaches improve objectively measured adherence [S10].

Refer when

Pitfalls

• Patching before excluding structural disease [S5] [S7].
• Skipping glasses: optical treatment alone has a large effect [S4].
• Prolonged glasses-only periods in children who need patching [S8].
• Missing the sound eye's acuity on atropine [S5].
• Stopping patching abruptly [S9].
• Assuming games or rewards will fix poor adherence: the evidence is very low certainty [S10].

Evidence notes

• The optical-treatment meta-analysis had very high heterogeneity (I² = 96.65%) [S4].
• EuPatch was not masked and dropout was higher in the extended-glasses group (14% vs 6%) [S8]; its finding conflicts with the "glasses first for a long period" approach supported by [S4].
• No Cochrane review found trials in bilateral refractive amblyopia [S1]; no RCTs assessed partial occlusion or optical penalisation for strabismic amblyopia [S3].
• The dichoptic therapeutic trial compared against glasses alone, not patching [S6].
• Participants in the adherence review were mostly White, limiting generalisability [S10].
• Funding: EuPatch was funded by Action Medical Research, NIHR Clinical Research Network and Ulverscroft Foundation [S8]; the 2025 Cochrane review had no dedicated funding [S10].
• Not covered by these sources: atropine concentration and frequency, deprivation amblyopia (cataract, ptosis), treatment of adults, recurrence rates, and evidence from low-income settings.

Related articles

• Vision screening in children
• Childhood esotropia
• Refractive error in children and spectacle provision

References

See the numbered source list below. Each marker [S#] links to its PubMed and DOI record.