Myopia progression after cessation of atropine in children: a systematic review and meta-analysis

Purpose: To comprehensively assess rebound effects by comparing myopia progression during atropine treatment and after discontinuation. Methods: A systematic search of PubMed, EMBASE, Cochrane CENTRAL, and ClinicalTrials.gov was conducted up to 20 September 2023, using the keywords "myopia," "rebound," and "discontinue." Language restrictions were not applied, and reference lists were scrutinized for relevant studies. Our study selection criteria focused on randomized control trials and interventional studies involving children with myopia, specifically those treated with atropine or combination therapies for a minimum of 6 months, followed by a cessation period of at least 1 month. The analysis centered on reporting annual rates of myopia progression, considering changes in spherical equivalent (SE) or axial length (AL). Data extraction was performed by three independent reviewers, and heterogeneity was assessed using I2 statistics. A random-effects model was applied, and effect sizes were determined through weighted mean differences with 95% confidence intervals Our primary outcome was the evaluation of rebound effects on spherical equivalent or axial length. Subgroup analyses were conducted based on cessation and treatment durations, dosage levels, age, and baseline SE to provide a nuanced understanding of the data. Results: The analysis included 13 studies involving 2060 children. Rebound effects on SE were significantly higher at 6 months (WMD, 0.926 D/y; 95%CI, 0.288-1.563 D/y; p = .004) compared to 12 months (WMD, 0.268 D/y; 95%CI, 0.077-0.460 D/y; p = .006) after discontinuation of atropine. AL showed similar trends, with higher rebound effects at 6 months (WMD, 0.328 mm/y; 95%CI, 0.165-0.492 mm/y; p < .001) compared to 12 months (WMD, 0.121 mm/y; 95%CI, 0.02-0.217 mm/y; p = .014). Sensitivity analyses confirmed consistent results. Shorter treatment durations, younger age, and higher baseline SE levels were associated with more pronounced rebound effects. Transitioning or stepwise cessation still caused rebound effects but combining optical therapy with atropine seemed to prevent the rebound effects. Conclusion: Our meta-analysis highlights the temporal and dose-dependent rebound effects after discontinuing atropine. Individuals with shorter treatment durations, younger age, and higher baseline SE tend to experience more significant rebound effects. Further research on the rebound effect is warranted.