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.

Intermittent exotropia

Intermittent exotropia (IXT) is the most prevalent subtype of exotropia in children [S8]. Most untreated children aged 3 to 10 years remain stable or improve slightly over 3 years [S3], so observation, part-time patching, overminus lenses and surgery are all options, and evidence on which is best remains limited [S7].

Key points - Deterioration to constant exotropia or loss of stereoacuity over 3 years without treatment was uncommon in children aged 3 to 10 years (PEDIG cohort) [S3]. - Part-time patching improved alignment compared with active observation at six months (Cochrane, high-certainty evidence) [S7], but both patching and observation are reasonable [S1]. - Overminus lenses improve control while worn, but the benefit was not maintained after stopping and they increase myopic shift, which persisted at 3 years [S6] [S11]. - After surgery, unilateral recession-resection may give better 8-year outcomes than bilateral lateral rectus recession, though the primary outcome did not differ significantly [S9].

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

Red flags

• Progression to a constant exotropia (10 prism dioptres or more at distance and near), the motor deterioration criterion used in trials [S3] [S5].
• Loss of near stereoacuity of 2 octaves or more [S3].
• Double vision: diplopia was one reason treatment was started in the observation cohort [S3].
• Myopic shift on overminus lenses: 17% of treated children had a shift of more than 1.00 D at 12 months vs 1% without overminus [S6].

Recognise it

• Trial definitions: IXT at distance, or constant exotropia at distance with IXT or exophoria at near, with an exodeviation of at least 15 PD at distance or near and at least 10 PD at distance by prism and alternate cover test (PACT) [S1] [S2].
• Parents may seek treatment for social concern: 7 of 25 deteriorations in one cohort were treatment started for social concern [S3].

Confirm it

• Measure the angle with PACT at distance and near, grade distance and near control, and measure near stereoacuity: these were the outcomes used across trials [S1] [S6] [S7].
• A masked examiner and confirmation by retest were used to define deterioration in trials [S1].

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)

• Refer children with an outward-drifting eye to an eye clinic. Deterioration over 6 months was uncommon with or without patching [S1], so this is not an emergency unless other signs are present.

Tier B (general eye clinic)

1. Observation. In children aged 3 to 10 years, the cumulative probability of deterioration by 3 years was 15%, likely an overestimate; control, stereoacuity and angle stayed stable or improved slightly [S3]. In children aged 12 to 35 months, motor deterioration by 3 years was 10%, and watchful waiting may be reasonable [S5]. Trials reviewed children at 3 months, 6 months, then every 6 months [S3] [S5].
2. Part-time patching. In children aged 3 to 10 years, patching 3 hours daily for 5 months reduced deterioration at 6 months (0.6% vs 6.1%), but both approaches were judged reasonable [S1]. In children aged 12 to 35 months, patching did not significantly reduce deterioration (2.2% vs 4.6%), and there was insufficient evidence to recommend it [S2]. A meta-analysis of 4 RCTs found better control, smaller distance deviation and better near stereoacuity with patching than observation [S8].
3. Overminus lenses. In a PEDIG RCT of 386 children aged 3 to 10 years, overminus spectacles (-2.50 D for 12 months, then -1.25 D for 3 months, as reported in [S6]) improved distance control at 12 months (1.8 vs 2.8 points), but not at 18 months after tapering [S6]. The greater myopic shift remained at 3 years (-0.74 D vs -0.44 D), and the risk should be discussed with parents [S11]. A meta-analysis of 4 RCTs found better control and smaller angles with overminus lenses, without better near stereopsis [S10].

Tier C (specialist centre)

• Choice of procedure. The Cochrane review found that bilateral lateral rectus recession likely results in little difference in alignment compared with unilateral recession-resection (moderate-certainty evidence) [S7]. In the 8-year follow-up (bilateral recession vs recession-resection), suboptimal outcome was 68% vs 53% (not significant), complete or near-complete resolution 15% vs 37%, and reoperation 30% vs 11%, the last two favouring recession-resection [S9].
• Age at surgery. In a secondary analysis, suboptimal outcome by 3 years was 28% for children aged 3 to under 5 years vs 50% for those aged 5 to under 11 years [S4]; a randomised trial of immediate vs delayed surgery is needed to confirm this [S4].
• Comparing non-surgical options. A network meta-analysis of 11 RCTs (1411 patients) ranked overminus lenses first; part-time occlusion, binocular vision training and base-in prisms showed no significant effect on control or angle, and none improved near stereoacuity [S12].

Refer when

Pitfalls

• Operating on every child at diagnosis: most remain stable over 3 years [S3].
• Prescribing overminus lenses without discussing myopic shift [S11].
• Expecting overminus benefit to persist after stopping [S6].
• Patching very young children (12 to 35 months) with high expectations: benefit was not shown [S2].
• Promising better stereopsis from non-surgical treatment [S10] [S12].

Evidence notes

• Conflict: the patching meta-analysis found better near stereoacuity with patching [S8], while the Cochrane review found little or no difference (low-certainty evidence) [S7] and the network meta-analysis found no significant effect of patching on control [S12].
• Most RCTs were from the United States; the Cochrane review included two trials from Turkey and India [S7]. Masking of participants was not possible, giving high risk of performance bias [S7].
• The 8-year surgical follow-up continued in 123 of 197 randomised children [S9]; long-term follow-up of the overminus trial lost 42% [S11].
• Not covered by these sources: convergence insufficiency-type IXT, surgical doses, orthoptic exercises in detail, and evidence from low-income settings.

Related articles

• Childhood esotropia
• Amblyopia: patching, atropine and binocular treatment

References

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