Back

Methodological issues in visible LED therapy dermatological research and reporting

Grimes, D. R.

2024-09-13 dermatology
10.1101/2024.09.12.24313560 medRxiv
Show abstract

The advent of mass-market Light Emitting Diodes (LEDs) has seen considerable interest in potential dermatological applications of LED light photobiomodulation (PBM) for a range of conditions, with a thriving market for direct-to-consumer LED treatments, including red light, blue light, and yellow light wavelengths. Evidence of efficacy for many conditions is however decidedly mixed, with starkly different outcomes reported by different authors. Due to the wide range of irradiances and wavelengths used, interpretation, comparison, and even efficacy evaluation is often impossible or prohibitive, impeding evidence synthesis. This work establishes a framework for objectively cross-comparing patient dose in terms of fluence, and a model for contrasting received dose to typical solar dose at ground level to facilitate interpretation of results and evidence synthesis. This allowes direct cross-comparison of patient skin fluence from LED PMB treatments under different regimes, and a means for evidence synthesis. This was applied to LED PMB data from 27 clinical trials to examine fluences and patient-equivalent solar exposure from LED light-sources for dermatological conditions, including acne vulgaris, wrinkle-reduction, wound-healing, psoriasis severity, and erythemal index. The results of this analysis suggest that fluences, wavelengths, and solar exposure equivalent differed by orders of magnitude in he studies analysed, with effective doses often comparable to typical daily solar exposure. Better dose quantification and plausible biological justification for various wavelengths and fluences are imperative if LED therapy studies for dermatology are to be informative and research replicability improved.

Published in PLOS ONE (predicted rank #4) · training set

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.