Infrared Light Cannot Be Doing What You Think It Is Doing (re: DOI: 10.1089/photob.2018.4489)

To the Editor:

Hipskind et al. recently published an open series treating traumatic brain injury (TBI) with an array of light-emitting diodes (LEDs). ¹ Small animal models indicate that transcranial photobiomodulation may benefit in TBI and stroke; however, a fundamental problem in infrared phototherapy has been how to “scale up” the therapy for clinical use. Given that a mouse skull is 0.2 mm thick and the thickness of the entire mouse brain is 7 mm, scaling the treatment to effectiveness in humans with 10 mm of scalp and skull is problematic. First, in an array of LEDs, there is not a cumulative effect. Although the entire organ may be exposed to multiple 0.5 W LEDs, the impact on each cell is only that of the cylinder of light falling upon it—0.5 W—no more. ² Second, reflection, refraction, scattering, and absorption all reduce the transmission of light energy through tissue. ² Over 80% of light energy is lost in skin ^3,4 and we have shown that energy from 0.5 W LEDs does not penetrate human skin. ⁴ Lapchak et al. ⁵ measured penetration of 700 mW (800 and 970 nm) light through animal and human skull. Only 4.2% of the light penetrated human skull. ⁵ Third, we measured penetration of near infrared (NIR) light through human hair, using a 10 W 810 nm emitter. Approximately 98% of the NIR energy was absorbed or reflected by the hair. ² When LED-based therapy devices are placed over hair, undoubtedly very little NIR energy reaches the brain. If 98% of the energy from a 0.5 W LED is absorbed by hair, ² 80–90% of what remains is absorbed by skin, ^3,4 and 96% of what remains after that is attenuated by skull, ⁵ then the claims of neurophysiological benefits of LED-based devices become highly questionable.

We have noted similar limitations in previous LED-based treatment studies of TBI in humans. ^6,7 Naeser et al. found modest benefit ⁶ when exposing patients to infrared light from LED-based devices. Moreover, the benefits faded quickly if the treatment was not continued on a daily basis. ⁶ In contrast, we found robust and persistent benefit in patients treated with multi-Watt infrared laser and followed for years after a single course of treatment. ^4,7 Based on the incontrovertible realities of light physics, we have proposed alternative theories for the observed, albeit slight, benefit on LED-based therapies. The weak and transient clinical benefit of LED-based treatments, ^1,6 compared with multi-Watt infrared laser therapy, ⁷ argues in favor of systemic effects most likely mediated by the skin. ²

Further, this article warranted closer reviewer scrutiny to statistics. The hypothesis is LED therapy will result in improvement. This one-sided hypothesis dictates one-tailed t-test. More importantly, they failed to correct for multiple measures of the same subject. One sided t-tests yield lower p-values, as does correcting for multiple comparisons, yielding nonstatistical results. At 3 weeks post-treatment, only two of three demonstrated cerebral perfusion changes. Given the transient benefit of LED-based therapy, ⁶ we suspect these results would have been further curtailed if imaging after treatment was delayed, unlike the results with multi-Watt infrared laser therapy. ⁷