Methodological quality of the review: Low confidence

Author: Flynn JT, Schiffman J, Feuer W, Corona A.

Region: Not specified

Sector: Therapy

Sub-sector: Therapy, amblyopia therapy, vision

Equity focus: None specified

Review Type: Effectiveness Review

Quantitative synthesis method: Narrative analysis

Qualitative synthesis methods: Not applicable

Background

Recent studies on prevalence indicate that amblyopia remains a very common cause of unilateral visual impairment and ranked first to third among adults up to the age of 65 years. However, therapeutic interventions for common eye disorders have received relatively little attention. The few efforts seem to suggest that therapy has become completely satisfactory and therefore required no change or critical analysis. Despite this, familiarity with its use in the clinic seems to suggest otherwise.

Research objectives

To determine the outcome of occlusion therapy in patients with anisometropic, strabismic, and strabismic-anisometropic amblyopia.

Main findings

The authors identified 961 amblyopic patients, participants in 23 studies, undergoing patching therapy for amblyopia from 1965 to 1994 with anisometropia, strabismus, or anisometropia-strabismus.

The study found that the mean spherical equivalent difference between the two eyes was striking in the anisometropic and anisometropic-strabismic patients but was minimal in the strabismic patients. The study again found that across different patient populations, amblyopia is successfully treated by occlusion therapy in the short term. Overall, the occlusion therapy for the three types of amblyopia was successful in 66.7% anisometropic, 77.6% strabismic and 58.7% of strabismic-anisometropic. Success was not related to the duration of occlusion therapy, type of occlusion used, accompanying refractive error, patients’ sex or eye. Initial acuity and age had a highly significant relationship with the rate of success (p=0.001). After adjusting for other co-variates, the type of amblyopia was statistically significant in influencing the risk of treatment failure. For instance, strabismic and anisometropic amblyopia shared the same risk, however, the odds of failure was two times higher for strabismic-anisometropic amblyopia.

The study proposed a further research question relating to the level at which difference in refractive error began to contribute to the decrement in visual acuity in the strabismic patient.

Methodology

The authors included studies that had patients with one of the three most common types of amblyopia – strabismic, anisometropic and strabismic-anisometropic; and occlusion in its various forms (part-time or full-time, total adhesive, shield, and pirate patch).

The authors conducted a search in databases including Medline for the years 1965 to 1997. They first reviewed the abstract and if the abstract contained data, these articles were then reviewed individually. Authors of studies that included substantial numbers of patients (more than 10) were also contacted for additional information; including individual patient data.

The authors constructed a three-dimensional risk diagram which employed the significant independent risk predictors of failure of occlusion therapy to portray how these factors contributed to that outcome in the population.

Applicability/external validity

The authors did not discuss the applicability/external validity of findings.

Geographic focus

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

Quality assessment

Overall, low confidence was attributed in the conclusions about the effects of this study as major limitations were identified. Although the search periods for included studies were comprehensive enough that relevant studies were unlikely to be omitted, the study did not explicitly mention the language used in searching for included studies, hence, we cannot be sure if language bias was avoided or not. The authors were silent on whether or not the reference lists of included articles were checked. Nor did they not specify whether or not the screening and extraction of text was independently done by at least two reviewers. No list of included and excluded studies was provided.