Non-surgical therapy for intermittent exotropia: a systematic review and network analysis

Abstract

Objective: This study aimed to conduct a network meta-analysis to evaluate the efficacy of various non-surgical treatments for intermittent exotropia(IXT).

Methods: A comprehensive search of the PubMed, EMbase, and Cochrane Library databases was performed to identify relevant randomized controlled trials (RCTs) up to June 2024. Following independent screening, data extraction, and bias assessment by two researchers, network meta-analysis was conducted using R 4.2.2 software.

Results: A total of 11 RCTs involving 1411 patients were included. Treatment options included overminus lenses (OML), conventional prisms(base-in prism) (CP), part time occlusion (PTO), and binocular vision training (BVT). OML demonstrated superior efficacy in improving distance and near control(1.1, 95% confidence interval (CI), 0.22 ∼ 1.8); 0.67, 95% CI :0.027 ∼ 1.2), as well as reducing near exodeviation compared to observation(4.5, 95% CI, 1.9 ∼ 6.9), but failed to reduce distance angle of deviation(3.2, 95%CI, -1.1 ∼ 6.4). No significant effect in improving control and reducing exodeviation angle at both distance and near was observed in PTO, BVT, and CP. Probability ranking indicated that the top-ranking three non-surgical interventions were OML, BVT and PTO for improving distance control and reducing near exodeviation; the top three non-surgical interventions for reducing distance exodevition were OML, PTO and BVT; the best non-surgical intervention for improving near control was OML, BVT、 PTO and CP have similar effects. The four non-surgical treatments had no significant impact on near stereoacuity.

Conclusion: Overall, OML ranks first among the four conservative treatment methods. These four commonly used non-surgical interventions did not significantly impact near stereoacuity. Clinicians should tailor personalized treatment strategies for patients with intermittent exotropia based on disease severity, characteristics, efficacy, and cost considerations.

Keywords: Intermittent exotropia; Meta-analysis; Non-surgical therapy.

Fig. 1

Flow diagram showing selection process for inclusion of studies in network meta-analysis

Fig. 2

Network plots for efficacy: mean distance control score changes. Each node represents 1 non-surgical intervention. The node size corresponds to the number of participants assigned to each treatment. Treatments with direct comparisons are linked with a line. The line thickness corresponds to the number of trials evaluating the comparison

Fig. 3

Results of network meta-analysis using observation as referent intervention in terms of distance and near control score changes: (A) distance control score changes, (B) near control score changes. Each non-surgical interventions was compared with the control treatment, which was the reference group. CrI = confidence interval

Fig. 4

Net league table of head-to-head comparisons for different non-surgical interventions in improving IXT control ability: (Lower left) mean difference in distance control score changes and (Upper Right) mean difference in near control score changes. The treatment comparisons should be read from left to right. The estimate is shown in the shared cell between the treatment column and row. Values of more than 0 mean differences favor the column-indicated treatment. Black shading represents various interventions, gray shading represents statistically significant differences between two interventions. CI: confidence interval

Fig. 5

Results of network meta-analysis using observation as referent intervention in terms of distance and near angle of deviation changes: (A) distance angle of deviation changes, (B) near angle of deviation changes. Each non-surgical interventions was compared with the control treatment, which was the reference group. CrI = confidence interval

Fig. 6

Net league table of head-to-head comparisons for different non-surgical interventions in reducing distance and near angle of deviation: (Lower left) mean difference in distance angle of deviation changes and (Upper Right) mean difference in near angle of deviation changes. The treatment comparisons should be read from left to right. The estimate is shown in the shared cell between the treatment column and row. Values of more than 0 mean differences favor the column-indicated treatment. Black shading represents various interventions, gray shading represents statistically significant differences between two interventions. PD: prism diopter, CI: confidence interval

Fig. 7

Rank probabilities of all regimes in terms of outcomes: distance and near control and exodeviation changes