Red Light Instruments for Myopia Exceed Safety Limits

Abstract

Purpose:

Low level red light (LLRL) therapy has recently emerged as a myopia treatment in children, with several studies reporting significant reduction in axial elongation and myopia progression. The goal of this study was to characterise the output and determine the thermal and photochemical maximum permissible exposure (MPE) of LLRL devices for myopia control.

Methods:

Two LLRL devices, a Sky-n1201a and a Future Vision, were obtained. Optical power measurements were made using an integrating sphere radiometer through a 7 mm diameter aperture, in accordance with ANSI Z136.1–2014, Sections 3.2.3–3.2.4. Retinal spot sizes of the devices were obtained using a model eye and high resolution beam profiler. Corneal irradiance, retinal irradiance and MPE were calculated for an eye positioned at the oculars of each device.

Results:

Both devices were confirmed to be Class 1 laser products. Findings showed that the Sky-n1201a delivered laser light as a point source with a 654 nm wavelength, 0.2 mW power (Ø 7 mm aperture, 10 cm distance), 1.17 mW/cm² corneal irradiance and 7.2 W/cm² retinal irradiance (Ø 2 mm pupil). The MPE for photochemical damage is 0.55–7.0 s for 2–7 mm pupils and for thermal damage is 0.41–10 s for 4.25–7 mm pupils. Future Vision delivered the laser as an extended source subtending 0.75 × 0.325°. It has a 652 nm wavelength, 0.06 mW power (Ø 7 mm aperture, 10 cm distance), 0.624 mW/cm² corneal irradiance and 0.08 W/cm² retinal irradiance (Ø 2 mm pupil). MPE for photochemical damage is 50–625 s for 2–7 mm pupils.

Discussion:

For both of the LLRL devices evaluated here, three minutes of continuous viewing approached or surpassed the MPE, putting the retina at risk for photochemical and thermal damage. Clinicians should be cautious with the use of LLRL therapy for myopia in children until safety standards can be confirmed.

Myopia prevalence has been increasing worldwide, and with it, the incidence of associated and potentially blinding pathologies and socioeconomic burden are also expected to increase.¹ Low level red light (LLRL) therapy has emerged as a treatment for myopia.² LLRL therapy is also referred to as repeated low-level red-light (RLRL) therapy or photobiomodulation. Tabletop devices and wearable headsets that emit red laser light for the purpose of myopia control are being made by several different manufacturers. They are most widely available in China, primarily through eye hospitals, but have also been found on online marketplaces. In addition to China, one device, the Eyerising Myopia Management Device, (eyerisinginternational.com) has been recently approved for use in the UK, Ireland, Australia and New Zealand. For LLRL therapy, children are instructed to look into a red light-emitting instrument for three minutes, twice a day (separated by at least 4 hours), five days a week, for the duration of the treatment period, which could be on the order of years. Results from clinical trials show excellent efficacy of LLRL therapy in children, significantly reducing axial elongation and myopia progression.^3–5

Effects of narrowband light (i.e., long wavelength “red” and short wavelength “blue”) on myopia were first investigated in animal models. Collectively, these studies have shown that the spectral properties of ambient illumination influence eye growth, refractive development and experimental myopia.^6–10 Early studies in chicks and guinea pigs found that rearing animals in blue light was protective for myopia, whereas red light increased myopia in chicks and green light increased myopia in guinea pigs.^{6, 7} Later studies in tree shrews and rhesus monkeys showed the opposite effect; rearing animals in red light prevents experimental myopia and promotes hyperopia.^{9, 11} In juvenile and adolescent tree shrews, animals that were exposed to red light (peak wavelength 624–628 nm, 527–749 lux) demonstrated more hyperopic refractions and shorter vitreous chamber depth than tree shrews raised in standard broadband fluorescent lighting.^{8, 11} Similarly, rhesus monkeys reared in red light (630 nm, 274 lux) with unrestricted vision showed hyperopic shifts in refraction, while rhesus monkeys undergoing experimental myopia showed a reduced myopic response.⁹ In these experiments, narrowband light was produced by ceiling- or cage-mounted light emitting diodes (LEDs), and the animals were exposed to 12–14 hours of red light per day for the duration of the treatment period, being for 21 days for chicks, 56 days for guinea pigs, 13 days for tree shrews and 125 days for monkeys. The temporal properties of red light on emmetropisation were investigated in tree shrews.¹⁰ The authors found that the hyperopic effects of red light increased non-linearly with duration; the average hyperopic shift to red light rose exponentially with duration. While 14 hours of red light exposure per day was most effective, even one hour per day of red light exposure produced some hyperopia.¹⁰

More recently, clinical trials have been undertaken to investigate red light as a treatment for myopia in children.^{2, 12} Unlike previous studies in animal models, in which red light was produced by LEDs, studies in children use instruments that emit red light from semiconductor lasers. In animal studies, the duration of red light exposure was generally for the entire waking period (12–14 hours per day), while protocols in children use three minutes of red light exposure twice a day. Since 2021, over 24 papers have been published on red light therapy for myopia,^{2–5, 12–26} including three review papers.^27–29 Eyerising International, manufacturer of the Eyerising Myopia Management Device, advertises the instrument for use in children as young as 3 years old, with online claims that over 100,000 children have been or are currently being treated. Published studies universally report high effectivity for myopia control, ranging from 65–87% treatment effect for reduced axial elongation and myopia progression. In some cases, axial shortening has been reported.^{3, 25} Wang et al. reported that 22% of children experienced axial shortening greater than 5 microns after 12 months of LLRL therapy.³

The mechanism of LLRL therapy for myopia is not well understood. Early studies suggested that effects may be related to chromatic aberration.⁶ Retinal signalling, cone sensitivities and dopamine secretion have also been thought to play a role.³⁰ However, the mechanism may be more similar to that observed in photobiomodulation for other tissues, such as skin. Red light acts directly on mitochondria.³¹ In dermatological applications, red light therapy has been shown to stimulate collagen production and increase fibroblast proliferation, increase mitochondrial cytochrome C, blood circulation and oxygenation, reduce inflammation and speed wound healing.^{32, 33} Given key similarities between skin and sclera, both consisting of fibroblasts in an extracellular matrix containing collagen, speculation exists whether LLRL therapy for myopia may be acting on the choroid to increase blood flow or directly on the sclera to prevent remodelling and subsequent axial elongation.

There have been no serious adverse events reported as a result of the treatment,¹⁵ except for one published case study of a 12 year old patient who sustained retinal damage after five months of LLRL therapy.³⁴ In this case, best corrected acuity decreased from 6/6 to 6/9. Fundus photos revealed bilaterally darkened foveae with a hypoautofluorescent plaque in autofluorescence images. Optical coherence tomography (OCT) imaging showed bilateral foveal ellipsoid zone disruption and interdigitation zone discontinuity. Multifocal electroretinography revealed moderately and mildly decreased responses in the macula and paramacula, respectively. Visual acuity improved to 6/7.5 three months after LLRL therapy was terminated.

The safety profiles of red light laser devices for myopia have not been fully investigated. Published studies using LLRL therapy reported that the devices are Class 1^2,3 or Class 2²² laser products. Class 1 lasers are low powered devices that are considered safe from potential hazards. Class 2 laser products are described as low power (< 1mW), visible light lasers that could possibly cause damage to the eyes. The goal of this study was to characterise the output and determine thermal and photochemical maximum permissible exposure time for two different LLRL devices being used for myopia control.

Methods

LLRL tabletop devices from two different manufacturers were obtained (Figure 1), including a Sky-n1201a (Beijing Akihito Vision, Vision Technology Co., Ltd.) and a Future Vision (Hunan Medical Technology Co., Ltd.). The Sky-n1201a consists of an on/off button, a music button and an adjustment for pupillary distance. When turned on, red light is emitted for 3 minutes, at which time the instrument automatically turns off. The tag on the instrument is labelled, “Class 1, 0.6 ± 0.2 mW.”

Figure 1:

The two devices evaluated here, Sky-n1201a and Future Vision. A) Oculars, B) operating panel and C) treatment laser for Sky-n1201a

Future Vision has three functions: 1) red laser function, 2) black-and-white afterimage function and 3) multi-colour light flash function. The manual for this instrument describes it as a comprehensive treatment for amblyopia and myopia, indicated for children ages 7–18 years. The specifications state that it is Class 1 and uses a 650 nm red semiconductor laser with a power of 0.2–1 mW to “promote dopamine secretion, increase choroidal thickness, and supply sufficient oxygen to the sclera to inhibit myopia.” It has a pupillary distance adjustment and a forehead sensor, so that the red laser function is only turned on when the sensor is covered. The instrument plays music for the duration of the three minutes, at which time the red laser automatically turns off.

For each device, optical power measurements were made using an integrating sphere radiometer (S142C sensor + PM100D console, thorlabs.com) through a 7 mm diameter aperture. The 7 mm aperture was placed directly on the input port of the sensor, and the sensor was located directly at or 10 cm away from the oculars of the device. A 7 mm diameter aperture and 10 cm distance was used in accordance with American National Standards Institute (ANSI) Z136.1–2014,³⁵ Sections 3.2.3–3.2.4 and Table 9, which describe the radiometric measurement parameters for assessment of Class 1 laser classification. Corneal irradiance $\left(E_c\right)$ was determined with the sensor at the oculars and calculated based on the irradiance measured within the same 7 mm aperture:

$$E_c=\frac{\mathrm{\Phi }\left(mW\right)}{0.385{cm}^2}$$

where $\mathrm{\Phi }$ is the radiant power at the aperture and 0.385 cm² is the area of a 7 mm diameter circular aperture. Retinal irradiance $\left(E_r\right)$ was calculated further assuming a user having fully constricted (i.e., 2 mm) pupils during device use:

$$E_r=\frac{E_c\left(mW/{cm}^2\right)\times 0.0314{cm}^2}{5.11\times {10}^{-6}{cm}^2}$$

where 0.0314 cm² is the area of a 2 mm diameter circular pupil and 5.11 × 10⁻⁶ cm² is the irradiated retinal area (25.5 μm diameter circle) when the pupil of a normal human eye is irradiated with a point source (e.g., a laser). Retinal spot size was determined for each device using a model eye and high resolution laser beam profiler (BP209-VIS, thorlabs.com, Figure 2). The model eye consisted of a 19 mm focal length achromatic doublet lens (AC127–019-B, thorlabs.com) with an iris aperture placed directly at the front surface of the lens. The BP209-VIS beam profiler was placed in the focal plane of the lens. The beam profiler has a spatial resolution of 0.43 μm and spot sizes were measured with the iris aperture set to a 2 mm diameter.

Figure 2:

Experimental setup and output for beam profiling of LLRL instruments. A) Overview of the setup, including the instrument under test (Sky-n1201a), B) Close up of the model eye and beam profiler placed at the focal plane, C and D) Retinal irradiance profiles of the Sky-n1201a and Future Vision, respectively, measured on the beam profiler (full frame 0.9 × 0.9 mm²)

A laser device is Class 1 if the output power measured through a specified limiting aperture at a distance of 10 cm is below the maximum permissible exposure (MPE) multiplied by the area of the limiting aperture. For retinal exposures to visible lasers greater than 10 seconds, the limiting aperture diameter is 7 mm (0.385 cm²), and the maximum permissible exposure is 1 mW/cm² (ANSI Z136.1–2014, Table 5b),³⁵ which gives a measurement limit of 0.385 mW for Class 1 classification.

ANSI Z136.1–2014 standards for laser class and luminance dose restrictions (ANSI Z136.1–2014, 8.3.3)^{35, 36} were applied to calculate the MPE.³⁷ ANSI Z136.1–2014, Section 8.3.3 applies to special conditions, such as exposures >100 s in ophthalmic applications. It states that any retinal radiant exposure over 100 seconds must be less than 5 J/cm²/V(λ) to avoid photochemical damage to the retina, where V(λ) is the Commission Internationale de l’Éclairage (CIE) photopic luminous efficiency function.

Results

Both the Sky-n1201a and Future Vision were confirmed to be Class 1 laser products. Measured wavelength, power and illuminance are shown in Table 1. Both devices have a wavelength ~650 nm, so a value of V(λ) = 0.1 (ANSI Z136.1–2014, Appendix I)³⁵ was used to determine retinal radiant MPE, in this case 50 J/cm². Results from beam profiler were used in the calculations of retinal irradiance and MPE. The Sky-n1201a produced a circular spot 11 × 11 μm² and can be considered a point source. The Future Vision produced an elliptical spot 250 × 125 μm², making it a 0.75 × 0.325° extended source.

Table 1:

Measured wavelength, power and illuminance for the red light devices used here, a Sky-n1201a and a Future Vision, and specifications provided in published studies

Device and Manufacturer	Wavelength	Power	Illuminance	Laser Class
Devices tested here
Sky-n1201a, Beijing Akihito Vision, Vision Technology Co. Ltd.	654 nm	0.2 mW (Ø7 mm aperture at 10 cm)	310 lux (at the cornea)	Class 1
Future Vision, Hunan Medical Technology Co. Ltd.	652 nm	0.06 mW (Ø7 mm aperture at 10 cm)	100 lux (at the cornea)	Class 1
Devices used in published papers (not tested in the current study) †
Sky-n1201, Beijing Ming Ren Shi Kang, Science and Technology Co. Ltd.38‡	650 nm	0.37 mW, 0.6 mW, 1.2 mW		Class 1
Eyerising, Suzhou Xuanjia Optoelectronics Technology (eyerisinginternational.com)2, 3, 5, 12, 14, 16, 17, 19–21, 24, 25	650 nm	0.29 mW (through 4mm pupil), 1.63 mW, 2 mW	1600 lux (from pupil to fundus)	Class 1
New Vision, Hunan Medical Technology18§	650 nm	0.16 mW for 4mm pupil	1,600 lux	Class 1
Myopia and Amblyopia Treatment Device, Hunan EnVan Technology15, 26	650 nm	n/a	n/a	Class 1
Optoelectronic Co, (vyoptics.com)22	635 nm	0.4 mW	n/a	Class 2
Longda, Jilin Longda Optoelectronics Technology, Jilin, China23	635 nm	0.35 mW	n/a	n/a
Ya Kun Optoelectronics Co. Ltd.13	650 nm	2 mW	n/a	n/a
LD-A, Jilin Longda Optoelectronics Technology, Jilin, China4	650 nm	0.35 mW	400 lux	n/a

^†these values are provided in published papers; further details regarding measurement protocols are not provided

^‡same manufacturer as the Sky-n1201a tested here

^§same manufacturer as the Future Vision tested here

Retinal irradiance with pupil diameter is shown in Figure 3, and MPE for photochemical and thermal damage with pupil diameter are shown in Figure 4. The Sky-n1201a has a wavelength of 654 nm, power of 0.2 mW (Ø 7 mm aperture, 10 cm distance), corneal irradiance of 1.17 mW/cm² and retinal irradiance of 7.2–88.2 W/cm² (Ø 2–7 mm pupils). The MPE for photochemical damage, defined as the time needed to reach a retinal radiant exposure of 50 J/cm², is 0.55–7 s for 2–7 mm pupils. The MPE for thermal damage is 0.41–10 seconds for 4.25–7 mm pupils. Viewing the laser with pupil diameters less than 4.25 mm does not put the user at risk of thermal damage.

Figure 3:

Retinal irradiance (W/cm²) with pupil diameter (mm) for the Sky-n1201a (solid symbols) and the Future Vision (open symbols)

Figure 4:

A) Photochemical and B) thermal maximum permissible exposure (seconds) with pupil diameter for the Sky-n1201a (solid symbols) and the Future Vision (open symbols)

The Future Vision has a wavelength of 652 nm, power of 0.06 mW (Ø 7 mm aperture, 10 cm distance), corneal irradiance of 0.624 mW/cm² and retinal irradiance of 0.08–0.97 W/cm² (Ø 2–7 mm pupils). The MPE for photochemical damage is 50–625 s for 2–7 mm pupils. The Future Vision does not put the retina at risk for thermal damage.

Discussion

We examined two different LLRL devices, the Sky-n1201a and the Future Vision, and confirmed both instruments are Class 1 laser products, as defined by International Electrotechnical Commission standard 60825–1:2014.2. Based on measurements in our laboratory, Future Vision is within the safety limits for thermal damage. However, the Sky-n1201a puts the retina at risk for thermal damage if the pupil diameter is 4.25 mm or larger. Three minutes of continuous viewing of either device surpasses the luminance dose MPE, putting the retina at risk for photochemical damage. Spectral and photometric measures for the two devices tested here were similar in wavelength and power to other LLRL devices described in published papers (Table 1). As highlighted here, LLRL devices from different manufacturers have different parameters and, therefore, different MPEs.

Class 1 lasers are low powered devices that are considered safe from all potential hazards when viewed accidentally and briefly. Examples of class 1 lasers are laser printers, CD players and digital video disc (DVD) devices. Class 1 lasers are not intended to be viewed directly for extended periods of time. Therefore, additional standards have been developed for lasers utilised for ophthalmic applications, where the light is meant to be directed at the retina (Section 8.3.3).³⁵ These standards for ophthalmic devices were used to determine the MPE and luminance dose restrictions in the current study. Of note, for the Eyerising and New Vision instruments used in clinical trials (see Table 1), claims of Class 1 do not appear to be supported by their published measurement data. The Eyerising instrument reports a power of 0.29 mW over a 4 mm pupil, which corresponds to an irradiance of 2.31 mW/cm² and, when averaged over a 7 mm aperture, per ANSI guidelines, gives a power of 0.889 mW (> MPE_{(Class 1)} = 0.385 mW). The New Vision instrument reports a power of 0.16 mW over a 4 mm pupil, which corresponds to an irradiance of 1.27 mW/cm², and when averaged over a 7 mm aperture, per ANSI guidelines, gives a power of 0.489 mW (> MPE_{(Class 1)} = 0.385 mW). Papers using other devices do not provide enough information to determine the MPE.

Thermal ocular injury from a laser can occur with exposures at any wavelength when the temperature change of the retina is greater than 10° C, resulting in denaturation of proteins. With thermal damage, the lesion is typically less than the size of the beam diameter, and the resultant scotomas are permanent. As measured here, the Sky-n1201a, but not the Future Vision, put the retina at risk for thermal damage when viewed for three minutes if the pupils did not constrict to less than 4.25 mm. Photochemical damage can occur with exposures for all wavelengths in the visible spectrum. Photochemical damage occurs when light hitting the retina interacts with molecules, including photoreceptor photopigments, retinal pigment epithelial melanin or lipofuscin granules, to cause a chemical change.³⁹ Damage may be due to the photo-oxidation of lipofuscin.⁴⁰ With photochemical damage, the minimal lesion size is typically equal to the beam diameter, and scotomas may be reparable. In the current study, luminance dose restrictions suggested that the MPE for direct viewing is 0.55 to 7 seconds for the Sky-n1201a and 50 to 625 seconds for the Future Vision, depending on pupil size. Therefore, for average pupil diameters of 4 mm, three minutes exceeds the MPE. Importantly, ANSI recommendations state that after reaching the MPE, light of similar intensity should not be viewed for at least 48 hours. However, in the case of LLRL therapy, children are instructed to view this high intensity laser light twice within 24 hours and repeated daily for potentially years.

ANSI standards were developed to protect individuals from hazardous light exposures in occupational situations. This assumes that individuals have an aversion response to bright light and are not looking directly at the light for extended periods of time. Therefore, guidelines for ophthalmic instruments were added to the ANSI standards in 2008, when luminance dose restrictions were established.³⁶ This Standard was added following a study which demonstrated light-induced retinal damage using lasers within the visible spectrum that were previously believed to be safe. High resolution autofluorescent imaging of the retinal pigment epithelium was performed in vivo in rhesus monkeys exposed to 0.5–2° of 568- or 830-nm light for 15 minutes at various intensities of 3 to 150 μW, followed by adaptive optics scanning laser ophthalmoscope imaging of the photoreceptors. Imaging was also performed intermittently for up to 165 days after exposure. The authors noted that even when exposing the retina to light that was nearly 2.3 times lower than the MPE, retinal damage was evident. For powers above 55 μW, dramatic changes at the photoreceptor and retinal pigment epithelial level were noted immediately and were still apparent 165 days after exposure. In the case of LLRL therapy, children are instructed to look at the laser continuously for three minutes. Our measurements were calculated with the assumption that children were foveating the light; it is possible that fixational eye movements may lessen the safety hazards as described here.

No published studies have carried out comprehensive investigations of structure and function during and after LLRL therapy. Retinal damage has only been reported for one case associated with LLRL therapy.³⁴ Clinical trials have typically used best corrected visual acuity as a measure of retinal function; some trials also used OCT imaging to assess structure. Jiang et al. defined adverse events as sudden vision loss of equal to or greater than two lines, a scotoma, decrease in best corrected visual acuity or structural damage seen on OCT scans.² Eyerising International recommend that patients should contact their clinician if the afterimage of the red laser lasts more than 5 minutes. These metrics are too gross to detect retinal damage at a cellular level. To thoroughly establish safety, more sensitive tests are required, such as multifocal ERG and single cell resolution adaptive optics imaging. In a published case of laser-induced maculopathy, Vitellas et al. used adaptive optics imaging to show that the inner and outer segments of cone photoreceptors were lost following direct viewing of a laser pointer.⁴¹ While the laser pointer was a higher risk class (Class 3A) than LLRL devices for myopia, the case illustrates the need for more sensitive testing to evaluate potential retinal damage.

In conclusion, based on measurements in our laboratory, it is recommended that clinicians strongly reconsider the use of LLRL therapy for myopia in children until safety standards can be confirmed. Instrument manufacturers must take into account ANSI standards for lasers used in ophthalmic applications to ensure that the retina is not at risk for photochemical damage. Future studies should utilise high resolution imaging and more advanced functional testing to assess retinal integrity.

Key Points.

• Two devices for low level red light therapy for myopia control evaluated here were shown to approach or exceed the maximum permissible exposure, depending on pupil size.

• The three-minute protocol for low level red light therapy may put the retina at risk for photochemical and thermal damage.

• Clinicians should be cautious with the use of low level red light therapy for myopia control in children until safety standards can be confirmed.