To evaluate the safety of repeated low-level red-light (RLRL) therapy in children, and the dynamic evolution of choroidal and retinal blood flow. This is a single-centre, randomised, single-blind, parallel-group clinical trial. Seventy myopic children were randomly assigned to either the intervention group [receiving RLRL therapy plus single-vision spectacle (SVS)] or the control group (wearing SVS). Participants underwent comprehensive ophthalmic examinations following their first irradiation, 9 months continuous RLRL therapy and stop of treatment. Quantitative analyses of choroidal and retinal microcirculation were analysed via optical coherence tomography angiography. Over 9 months of treatment, while the RLRL treatment demonstrated significantly less increases in refractive error and axial length compared with the SVS treatment (ps < 0.05), no abnormalities in fundus structure or visual function (mfERG, VEP and microperimetry) were detected (ps > 0.05). A single red-light exposure did not exert a significant influence on choroidal thickness (ps > 0.05). Upon continuous treatment, the RLRL group achieved peak values in these circulations at 9 months (ps < 0.05). Following cessation of exposure, all circulations exhibited a declining trend, reaching similar levels in both groups (ps > 0.05). As the frequency of red-light exposures intensified, there was a consistent surge in these circulations (ps < 0.05). Nine months of continuous RLRL exposure does not cause toxic side effects on retinal or optic nerve functions, and there is a time-dependent cumulative response in choroidal and retinal circulation.