French National Agency for Food, Environmental and Occupational Health & Safety (ANSES) has completed, on May 2019, several years of expertise on potential hazard of LEDs, and published series of recommendations in a 424-pages report in French. One of the most significant recommendation is the decreasing of limit value for human exposure to blue light from LEDs. Yet, based on International Commission on Non-Ionizing Radiation Protection (ICNIRP) guidelines, European standards set the exposure limit value (ELV) to 2.2 J / cm2, which is 5 times lower than the retinotoxic value. In ANSES report, arguments in favor of decreasing ELV are relied, mainly, on:
– in vitro studies, which provide elements that contribute to the risk analysis but do not allow a direct extrapolation on the effects of LEDs on human eye health.
– a study carried out on primates whose physiological and biometric properties of the eye are close to those of the human eye; but in this study, illuminations, of the order of 7000 lux, are not representative of normal lighting situations.
-A study which highlighted that Retinal Pigmented Epithelium AutoFluorescence photobleaching appears at brightness levels actually 20 times lower than the exposure limit value for 568 nm wavelength, which is green-yellow light, not blue. Moreover, authors state that it is not known whether the RPE AF photobleaching is a marker of a benign reaction or a potentially retinotoxic event. However, on this base, ANSES report’s authors extrapolate the results and estimate that ELV for retinal toxicity of blue light could be 20 times greater than a really protective value.
– Some in vivo studies (several of them conducted under the auspices of INSERM, the french national institute of health and medical research) made on rodent model. On the relevance of rodent model as used in these studies, we have published several peer-reviewed articles or conference papers indicating the effects, on the result of retinal exposure calculation, of the uncertainty surrounding the biometric values (pupil size and focal length) of the rat eye. We also published data on differences in pupil size/ focal length ratio between the human eye and the rat eye, which imply that, for a given lamp, the retinal illumination (and thus the phototoxic risk ) is much lower for humans than for rats. As the cited studies fail, on one hand, to take into account the uncertainties surrounding the calculation of the rat retina exposure, and, on another hand, to take into account the differences in retinal illumination introduced by the differences between the rat eye and the human eye, they should not be used to call into question the limit value for human exposure to blue light from LEDs. However, it sounds like our critical papers on methodology used by most of studies based on rat model have never been considered nor discussed by ANSES working group.
The ANSES report does not only ignore our divergent opinion, it is also opposite, in some of its conclusions, to those of SCHEER, the European Scientific Committee on Health, Environmental and Emerging Risks, that concluded last year, in its Final Opinion on
potential risks to human health of Light Emitting Diodes (LEDs) that “there is no evidence of direct adverse health effects from LEDs in normal use by the general population. Cellular and animal studies showing adverse effects appeared to be conducted under exposure conditions that were difficult to relate to human exposures or used exposure levels in excess of internationally agreed exposure limits ”. In conclusion, the evidence for retinal phototoxicity of light produced by LEDs at low levels of illumination (“low dose effect”) is only suspected in rats, and the results are potentially tainted with uncertainties related to accuracy of fundamental parameter values such as the pupil size and the focal length of the rat eye. In addition, given the biometric differences between the rat eye and the human eye, a lamp generally creates higher retinal illuminations on the rat retina than on the human retina, and one cannot consider LEDs in normal use as dangerous for general population because it is a danger to rats. This is also the opinion of European experts of the SCHEER. So, ANSES report really sounds like a “French affair”, that proves nothing but the fact that it is perhaps time for a unique European Agency in charge of evaluating emerging risk, what probably could help to save money, but also to produce better expertise and to provide more reliable scientific information to European citizens.
ANSES, avis et rapport de l’Anses relatif aux effets sur la santé humaine et sur l’environnement (faune et flore) des systèmes utilisant des diodes électroluninescentes (LED), avril 2019.
Click to access AP2014SA0253Ra.pdf
SCHEER, Final Opinion on potential risks to human health of Light Emitting Diodes (LEDs), 5-6 June 2018
Click to access scheer_o_011.pdf
Point, S., Lambrozo, J. (2017). Some evidences that white LEDs are toxic for human at domestic radiance? Radioprotection.
https://www.radioprotection.org/articles/radiopro/abs/2017/04/radiopro170025/radiopro170025.html
Point S. 2018. Blue Light Hazard: are exposure limit values protective enough for newborn infants? Radioprotection.
https://www.radioprotection.org/articles/radiopro/abs/2018/03/radiopro170072/radiopro170072.html
Point, S., (2018) Lumière et santé, Yearbook 2018, revue ERS.
Click to access ers-312078-synthese_lumiere_bleue_et_sante-a.pdf
Point, S., Beroud, M. (2019) Blue light hazard : does rat retina make a relevant model for discussing exposure limit values applicable to humans?, Radioprotection.
https://www.radioprotection.org/articles/radiopro/abs/first/radiopro190006/radiopro190006.html
This post is also available in: FR (FR)DE (DE)