Ostrin · JAMA ophthalmology 2026 · Laboratory bench safety evaluation · n=4 devices

Safety Evaluation of 4 Red Light Therapy Devices for Myopia.

Cited 8 times in the scientific literature.

Level 5 - mechanism / opinion, no new human data

Laboratory bench evaluation of optical devices without human clinical data

PubMed 41642586 · doi:10.1001/jamaophthalmol.2025.5660 · record verified 2026-08-27

What was done

Researchers conducted a laboratory-based optical evaluation of 4 commercially available red light therapy devices marketed for myopia management (Sky-n1201, Future Vision, EyeRising, and AirDoc) between November 2023 and April 2024. Radiometric power was measured with an integrating sphere radiometer at 1-cm and 10-cm distances through a 7-mm aperture. Retinal irradiance was calculated across modeled pupil diameters ranging from 2 mm to 7 mm. Devices were classified according to ANSI Z80.36-2021 and ANSI Z136.1-2022 safety standards to assess the exposure time needed to reach group 1 safety limits.

What was found

For a 7-mm pupil, the Sky-n1201 and EyeRising laser instruments reached ANSI group 1 safety limits within 2.8 seconds and 1.4 seconds of exposure, respectively, and were classified as Class 1 and Class 2M laser devices. The Future Vision device reached group 1 limits at extended exposures of 253 seconds or longer (Class 1 laser). The LED-based AirDoc device produced diffuse illumination requiring 22,761 seconds to reach the group 1 limit, classifying it as group 1. Typical clinical protocols recommend 180-second treatment sessions.

Why it matters

Multiple commercial laser-based red light therapy devices deliver retinal irradiance exceeding established ANSI optical safety limits in under 3 seconds, well below standard 180-second clinical exposures, highlighting potential ocular safety risks before widespread pediatric use.

Limits

This was an in vitro bench study measuring optical outputs and calculated retinal irradiance using standardized optical models rather than direct measurements in living human eyes. It did not directly assess biological tissue damage or account for individual variations in ocular anatomy, accommodation, or eye movements. Only 4 specific device models were evaluated.

Cited by