Amblyopia

Methodological quality of the review: Medium confidence

Author: West S, Williams C.

Region: Details not provided

Sector: Amblyopia

Sub-sector: Screening, treatment

Equity focus: Children aged up to 15 years

Review type: Effectiveness review

Quantitative synthesis method: Narrative analysis

Qualitative synthesis methods: Not applicable

Background

Amblyopia is commonly associated with squint (strabismus) or refractive errors resulting in different visual inputs to each eye during the sensitive period of visual development (under seven to eight years of age). The cumulative incidence is estimated at 2% to 4% in children aged up to 15 years. The population prevalence is affected by whether there have been any interventions to prevent or treat the condition.

Research objectives

To determine the effects of interventions to detect amblyopia early and the effects of medical treatment for amblyopia.

Main findings

Authors included 33 systematic reviews, randomized controlled trials (RCTs) or observational studies. Studies included reported effects of screening or treatment of amblyopia in children. Authors reported the following findings:

• Vision screening before school entry may increase detection rates of amblyopia. However, pre-school screening may not improve treatment outcomes at seven years compared with school-entry screening. It is not clear whether children with higher risk of eye problems should be targeted for vision screening.
• Most evidence was available for children aged under seven years, in whom wearing glasses for up to 30 weeks can improve amblyopia and may cure it. Children with suspected amblyopia who had clinically important refractive error were prescribed glasses; therefore, most data available on other interventions assessed their effectiveness in combination with glasses.
• Occlusion (covering the fellow eye using a patch) may have be more effective than glasses alone in children up to 13 years of age not fully treated with glasses. Further data assessing occlusion in combination with near-vision tasks, such as encouraging the child to do close work while wearing their patch, confirmed that combined interventions were more effective than glasses alone in younger children. Some older children might improve with treatment, although there were few data available to support this. It was not clear whether prescribing occlusion of the fellow eye for longer periods every day was more effective than prescribing for shorter periods of daily occlusion, but success rates do increase in proportion to objectively measured compliance.
• Penalisation with atropine may have been as effective as occlusion when given in combination with other interventions for improving amblyopia in children aged under seven years who were not fully treated with glasses. It was not clear whether near-vision tasks are effective alone as adjuvant treatment to glasses for amblyopia. Near-vision tasks may have further enhanced visual acuity when added to occlusion or penalisation, but the contribution of near-vision tasks to the effects of these combination interventions remained unclear.

Authors noted that future research was needed to determine whether more practicable screening protocols could be cost-effective.

Methodology

The following databases were used to identify studies for this systematic review: Medline 1966 to June 2010, Embase 1980 to June 2010, and The Cochrane Database of Systematic Reviews, 2010, Issue 2 (1966 to date of issue). An additional search within The Cochrane Library was carried out for the Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment (HTA). The authors also searched for retractions of studies included in the review. Abstracts of the studies retrieved from the initial search were assessed by an information specialist. Selected studies were then sent to the contributor for additional assessment, using predetermined criteria to identify relevant studies.

Study design criteria for inclusion in this review were: published systematic reviews (including systematic reviews of observational studies), RCTs, and prospective and retrospective cohort studies in any language, including open studies, and containing more than 20 individuals, of whom more than 50% were followed up. There was no minimum length of follow-up required to include studies. Systematic reviews of RCTs and RCTs were included where harms of an included intervention were studied applying the same study design criteria for inclusion as we did for benefits. In addition, the authors continue use a regular surveillance protocol to capture harms alerts from organizations such as the FDA and the MHRA, which are added to the reviews as required.

To aid readability of the numerical data in the reviews, the authors round many percentages to the nearest whole number. Readers should be aware of this when relating percentages to summary statistics such as relative risks (RRs) and odds ratios (ORs). A GRADE evaluation of the quality of evidence for interventions was included in this review. The categorization of the quality of the evidence (high, moderate, low, or very low) reflected the quality of evidence available for the authors’ chosen outcomes in the defined populations of interest. These categorizations were not necessarily a reflection of the overall methodological quality of any individual study, because the Clinical Evidence population and outcome of choice may have represented only a small subset of the total outcomes reported, and population included, in any individual trial.

Applicability/external validity

Authors did not discuss the applicability/external validity of findings.

Geographic focus

Authors did not provide the geographic focus of studies included in the review.

Quality assessment

There is medium confidence in the conclusions about the effects of this study. Although the authors conducted a comprehensive search of relevant databases avoiding language bias, they did not contact authors or experts for additional potentially relevant studies as part of the search strategy. They used appropriate criteria for assessing included studies, however the risk of bias was not avoided in regards to the methods used to include studies and extract data of included studies. The authors used descriptive syntheses to analyse the findings of included studies and were clear in terms of which studies were low or high risk of bias.