Ocular Risks of Topical Atropine Prescriptions Among Taiwanese Children

Importance: Myopia is a global health concern, with high myopia elevating the risk of ocular complications. While atropine can affect myopia progression, its long-term safety profile remains uncertain. Objective: To investigate the ocular risks associated with atropine prescription in clinical practice for myopia control in Taiwanese children. Design, Setting, and Participants: This was a retrospective, population-based cohort study using longitudinal data from Taiwan's National Health Insurance Research Database (2000-2021). Children aged 8 to 15 years with newly diagnosed myopia and matched nonmyopic control participants in the period between 2001 and 2015 were included. Data analysis was performed from March 8 to May 30, 2024. Exposure: Data on atropine prescription were collected, and participants were categorized by cumulative duration and dose. Main Outcomes and Measures: Diagnoses of cataracts, primary open-angle glaucoma, and maculopathy were recorded during the observation period. The incidence of ocular complications after a minimum 5-year follow-up period was reported. Hazard ratios (HRs) were derived using multivariable Cox regression models to evaluate atropine-related risks. Results: A total of 1 213 846 Taiwanese children (mean [SD] age at myopia diagnosis, 10.4 [1.9] years; 633 440 [52.2%] female) were included. Among 606 923 children with myopia, 406 383 (67.0%) were prescribed atropine. The incidence of ocular complications (cataracts, glaucoma, and maculopathy) was higher in the myopia group (1.54 per 10 000 person-years) compared with the nonmyopia group (0.96 per 10 000 person-years; adjusted HR [aHR], 1.49; 95% CI, 1.36-1.64). In children with myopia, the incidence of ocular complications was not different between atropine users and nonusers (both 1.54 per 10 000 person-year; aHR, 1.05; 95% CI, 0.93-1.18). An increased risk of ocular complications was observed in children with cumulative duration of atropine prescription exceeding 3 years (aHR, 1.51; 95% CI, 1.17-1.94). However, this trend was not present when conditioned on children with high myopia (aHR, 1.10; 95% CI, 0.56-2.19). No increased risk was found among children with the highest quartile of cumulative atropine dose (aHR, 1.05; 95% CI, 0.89-1.25). Conclusions and Relevance: While the incidence of cataracts, glaucoma, or maculopathy was higher in children with myopia, atropine prescription was not associated with these risks. While these findings offer insights for weighing risks and benefits of atropine use in myopia control, they are limited to atropine prescriptions implying atropine use, and they generate hypotheses but cannot imply cause-and-effect relationships.