Open

Preface

This risk-benefit analysis (RBA) is part of Forever Healthy's "Rejuvenation Now" initiative that seeks to continuously identify new therapies and systematically evaluate them on their risks, benefits, procedures and potential application.

Special thanks are extended to the whole Rejuvenation Now team at Forever Healthy for their friendly contributions.

Section 1: Overview 

Motivation

Low-level light therapy (LLLT) is the use of low incident levels of photon energy at a particular wavelength, targeting tissue to achieve a clinically useful local or systemic effect without the creation of heat (athermal) or damage (atraumatic) (Calderhead & Tanaka, 2017). LLLT has shown dramatic effects when used for wound healing, pain management, and various musculoskeletal conditions.

This review focuses on its potential use in skin rejuvenation. It has been shown that upon exposure to light, chromophores in the skin (mitochondrial cytochrome C, melanin, and protoporphyrins) absorb photons which lead to downstream alterations in physiology such as changes in cell proliferation, differentiation, migration, inflammatory mediators, and collagen production. It is supposed that these photobiomodulative effects have beneficial effects on the skin, leading to a more youthful appearance through increased collagen and elastin production, and a reduction in age spots and wrinkles. 

Key Questions 

This analysis seeks to answer the following questions:

• Which benefits with regard to skin rejuvenation result from LLLT? 

• Which risks are involved in using LLLT for skin rejuvenation (general and method-specific)?

• What are the potential risk mitigation strategies?

• Which method/device or combination of methods/devices is most effective for skin rejuvenation using LLLT?

• Which of the available devices/methods are safe for use? 

• What is the best therapeutic protocol available at the moment?  

Impatient readers may choose to skip directly to Section 6 for the conclusion and tips on practical application. 

Section 2: Methods

Analytic model

The RBA has been prepared based on the principles outlined in A Comprehensive Approach to Benefit-Risk Assessment in Drug Development (Sarac et al., 2012). 

Literature search

A literature search was conducted on Pubmed, Google Scholar, and the Cochrane Library using the search terms shown in Table 1. Titles and abstracts of the resulting studies were screened and relevant articles downloaded in full text. The references of the full-text articles were manually searched in order to identify additional trials that may have been missed by the search terms.

Inclusion criteria: We chose to include any dermatological trial conducted in humans using LLLT (any type of light source) that specified the mode of action as athermal, photobiomodulation.

Exclusion criteria: Trials that used ablative, fractional, and nonfractional lasers were excluded unless the study specified that the laser was used at settings that were nonthermal and atraumatic. We also excluded any photodynamic therapy trials that did not have a light therapy treatment only arm. 

Table 1: Literature Search 

Recommended Reading

The following sites offer information on LLLT at a consumer level and are useful as an introduction to the topic:

•  

  • Red Light Therapy Benefits - healthline.com

  • The Ultimate Guide to Red Light Therapy and Near-Infrared Light Therapy - energyblueprint.com

  • Definitive Guide to Red Light Therapy (Photobiomodulation) - trustedsupplementreviews.com
    
    

  • Yes, LED Facial Treatments Do Work (But Not In A Flash) - huffingtonpost.ca 

  • What to know about red light therapy - medicalnewstoday.com

Abbreviation list

Section 3: Existing evidence

Summary of results from clinical trials (humans)

We screened the titles/abstracts of 1836 papers identified by our search terms and included 57 clinical trials in our analysis. The remainder were excluded due to duplication or lack of relevance. The included studies used a variety of light sources (gas discharge lamps, LED) with widely differing treatment parameters making comparison difficult. Many of the studies were written by authors with a declared conflict of interest (n= 18). Several studies were small and were not placebo-controlled. The overall quality of the evidence on this topic is low, with most studies relying on subjective patient reporting or blinded photographic assessment by clinicians. 

Although our focus was on skin rejuvenation, we also included trials on other dermatological conditions treated by LLLT in order to identify potential risks and benefits beyond the condition studied. These included: 14 trials on the treatment of acne, 4 trials on actinic keratosis (a result of photoaging), 5 trials on psoriasis, and 1 trial on melasma. The trials conducted on acne were of higher quality and declared fewer conflicts of interest (2/14) than most other trials. The majority of the other disease trials were funded by device companies (actinic keratosis 4/4, psoriasis 3/5). Most trials irradiated the skin on the face only. A few trials irradiated skin on the back, arms or legs, and one trial used "full-body" irradiation. 

Table 2: Clinical Trials 

Section 4: Risk-Benefit Analysis

Decision Model

Risk and benefit criteria

The decision profile is made of up risk and benefit criteria extracted from the outcomes of the above-mentioned papers. The benefit criteria are organized by category and include the type, magnitude, and duration of the benefit as well as its perceived importance to the patient. The risk criteria are organized by category, type, severity, frequency, detectability, and mitigation. All are assigned numerical values: 

1 = low

2 = moderate

3 = high

The numerical values for both risk and benefit criteria are then summarized serving as the justification for the weighting in the following column.

Weight

The criteria are weighted on a value scale to enable comparison (based on the relative importance of a difference). Each benefit and risk criteria is assigned a weight/importance of 1 (low) 2 (medium) or 3 (high).

Weighting is independent of data sets and the final weights are based on consensus with justification based on the preceding columns of the table.

Score

Each category is assessed according to the performance of LLLT against the comparator (physiological aging) whereby a numerical value is assigned for each criterion -1 (inferior), 0 (equivalent or non-inferior) and +1 (superior) to the comparator.

Uncertainty

Uncertainty is determined according to the amount and quality of the evidence, whether it came from human or animal studies and whether methodological flaws, conflicting studies, or conflicts of interest (funding) by the authors are present. Human evidence is initially assigned a score of "1", evidence from rodent studies, "2", and in vitro or lower animal studies, "3". The uncertainty score is then adjusted by upgrading or downgrading using the above-mentioned criteria. 

Weighted score

The weights and scores are multiplied to produce weighted scores that enable direct comparison (-3 → +3) and then adjusted using the uncertainty score. Weighted scores may be upgraded where the uncertainty of the evidence is low or downgraded where the uncertainty of the evidence is high. 

Benefit assessment 

Our analysis identified a total of 28 benefits that have been documented in clinical trials to date. 

Table 3: Benefit assessment  

Benefits 

Reduction in pore size

A reduction in pore size following LLLT was reported by two studies (Lask et al., 2005; Fournier et al., 2006). Treatments were conducted with LED or metal halide lamp sources using a combination of blue and red or infrared light. The studies showed significantly reduced pore size in 85-90% of patients (after 8 treatments over 4 weeks) that lasted at least 3 months (Lask et al., 2005; Fournier et al., 2006).

Reduction in hyperpigmentation/melanin levels

Six trials reported beneficial effects on hyperpigmentation (Lee et al., 2007; Barolet, 2018; Weiss et al., 2006; Weiss et al., 2005; Ji et al., 2014; Shaoul & Mulholland, 2011) resulting from treatment with a variety of wavelengths (yellow, red, and infrared). In one trial, melanin levels actually increased with treatment with blue light but decreased with red or combined blue-red light treatment (Lee et al., 2007). A study on melasma showed significant pigment reduction that was maintained at least 12 months in 75% of patients (Barolet et al., 2018). The same study also showed an improvement on the untreated side, indicating a possible systemic effect of LLLT. Between 60 and 90% of patients benefited from improvements in pigmentation (Weiss et al., 2006; Weiss et al., 2005). 

Skin tone/complexion/radiance

Five trials reported improvements in skin tone/complexion (Lee et al., 2007; Sadick 2008; Wunsch & Matuschka, 2014; Fournier et al., 2006; Bhat et al., 2005). However, the intensity of the improvement was not assessed. The light sources used were LED or gas discharge lamps at wavelengths ranging from 415 to 890 nm. All but one of the trials that achieved this result used combination treatment. The likelihood of experiencing this improvement ranged from 58 to 87% and the duration of the benefit was at least moderate. Follow-up assessments (conducted between 5 and 12 weeks post-treatment) showed continuing benefits. 

Skin Texture/smoothness

Eleven trials reported an improvement in skin texture (Lee et al., 2007; Snehal et al., 2006; Weiss et al., 2006; Weiss et al., 2005; Ji et al., 2014; Sadick 2008; Shaoul & Mulholland, 2011; Wunsch & Matuschka, 2014; Migliardi et al., 2009; Barolet et al., 2009; Bhat et al., 2005). The light sources used were LED or gas discharge lamps with a variety of wavelengths that included blue, yellow, red and infrared both alone and in combination. Most studies relied purely on subjective assessments or at best, blinded image grading and did not attempt to quantify the change in skin texture. The likelihood of experiencing an improvement in skin texture ranged from 58.3% to 97%. Improvement in texture was graded by clinical assessment in one trial as "excellent" (in 25% of participants) or "considerable" (in 73% of participants). Skin texture either continued to improve or was maintained for at least 12 weeks post-treatment in several studies. 

Periocular wrinkles

Four trials specifically assessed the reduction of ocular wrinkles (Nam et al., 2017; Snehal et al., 2006; Weiss et al., 2005; Russell et al., 2005). These trials were all performed using LED systems as the light source and a variety of wavelengths including yellow, red, infrared, and white light. Objective measurement using a skin microrelief of the periorbital area showed a significant reduction in periocular wrinkles, particularly deep wrinkles. However, in the same trial, blinded dermatologists failed to find a significant reduction in eye wrinkles (Nam et al., 2017). The likelihood of experiencing a reduction in eye wrinkles was between 66 and 95%. 

In two articles that reported results from the same trial, participants received 9 treatments over 5 weeks and results were assessed through periorbital skin reliefs, blinded clinical assessment, and subjective improvement reports (Snehal et al., 2006; Russell et al., 2005). The authors found that although maximum furrow depth did not change significantly the density of furrows was reduced at the nine-week follow-up. Subjective improvement scores also peaked at 9 weeks with improvements reported by 83.9% of the subjects. This improvement was maintained at 12 weeks in 80.6% of subjects indicating at least a moderate duration of the benefit. 

The final trial used yellow light (8 treatments over 4 weeks) and participants showed an average improvement of 10% over baseline on profilometry. The benefit was experienced by 62% of participants two months post-treatment. The blinded assessment showed a reduction of at least one wrinkle class by the majority of subjects. Again, results peaked at 4 months post-treatment (Weiss et al., 2005).

Global photoaging scores

Four trials using LED light sources with yellow, red, and/or infrared wavelengths reported a decrease in "global photoaging" as evaluated by blinded assessment (Snehal et al., 2006; Weiss et al., 2005; Ji et al., 2014; Russell et al., 2005). In two articles reporting results from the same trial, photoaging assessment scores were significant at all followups with 51.6% of the population experiencing a 25-50% improvement and 12.9% experiencing a 50-75% improvement at 3 months post-treatment (Snehal et al., 2006; Russell et al., 2005).

A trial that used a yellow light LED system found that more than 85% of patients showed at least a 25% improvement in the global photoaging score at 4 months post-treatment. This effect was reduced at 12 months but still significantly above the baseline (Weiss et al., 2005). 

The light-only arm of a trial that compared PDT (2 sessions) with LED red light therapy (3 treatments per week for 4 weeks) for the treatment of photoaging found that both therapies significantly reduced photoaging (Ji et al., 2014). This trial irradiated forearm skin and dermoscopic evaluation two weeks after treatment showed marked improvement in skin appearance and elimination of photodamage signs in both groups (although it was more prominent in the PDT group). 

Firmness

Three trials using LED systems with combined red and infrared light wavelengths and one on red light alone, reported increased skin firmness (Snehal et al., 2006; Sadick 2008; Migliardi et al., 2009; Bhat et al., 2005). The magnitude of the benefit was not measured but by subjective assessment, between 52% and 68% of participants (Snehal et al., 2006; Sadick 2008; Bhat et al., 2005 ) reported increased skin firmness. One trial used a satisfaction survey and reported 10% of participants were very satisfied, 50% satisfied and 40% slightly satisfied with the results on skin firmness (Migliardi et al., 2009). The results were retained at the 12-week follow-up. 

Collagen production

An increase in collagen following LLLT therapy was reported by seven trials (Snehal et al., 2006; Weiss et al., 2005; Wunsch & Matuschka, 2014; Lee et al., 2007; Barolet et al., 2009; Barolet et al., 2005; Nikolis et al., 2016), as assessed through biopsy and ultrastructural analysis or ultrasonography of collagen density. The trials used LED and gas discharge lamps as light sources with a variety of wavelengths (yellow, red, infrared, and broad-spectrum). 

Electron microscopy of five biopsied samples showed evidence of an increased number of thicker, new collagen fibrils after treatment (Snehal et al., 2006). In another study, all biopsies (10) showed significant collagen production as well as collagen deposition in the papillary dermis and staining with antibodies revealed a 28% average increase in collagen density (Weiss et al., 2005). 

Ultrasound evaluation of collagen density also found significant increases in groups that had been exposed to red or broad-spectrum light (Wunsch & Matuschka, 2014). 

A double-blind, placebo-controlled, split-face, clinical trial that compared red, infrared and combination treatment found that in all three groups, the amount of collagen was significantly increased at 2 weeks post-treatment in the entire dermis and that each collagen bundle was more packed and better organized (Lee et al., 2007). 

An increase in type 1 procollagen production was observed in the supernatants of human reconstructed skin in LED-treated samples (mean increase of 31%) (Barolet et al., 2009). Biochemical analysis of suction blisters raised from patients that had received 12 high-fluence and high-irradiance LED treatments using a sequentially pulsed mode demonstrated a 50% increase in type III procollagen production rate compared to control sites whereas patients that received treatment using a continuous wave mode showed an increase of 29%. Patients that received low-fluence and irradiance treatments exhibited a 76% increase in type III procollagen production rate in the sequential pulsing mode group vs. 38% in the continuous wave mode group (Barolet et al., 2005). 

Postauricular biopsies performed at weeks 0 and 12 revealed an average increase of 287% in collagen levels (4 treatments over 4 weeks) (Nikolis et al., 2016).

Under-eye bags

One trial reported the subjective results of infrared LED therapy on under-eye bags (Stirling & Haslam, 2007). 41% of subjects reported an improvement of the saggy tissue under their eye and of these, all but one correctly identified the eye that had received the treatment. 

Fibroblast number

A study that examined the morphological ultrastructure of the dermis following treatment with a red LED light source identified an increase in the number of fibroblasts (Takezaki et al., 2005). The results also demonstrated a treatment-dependent enhancement in metabolism evidenced by the presence of more mitochondria and vimentin in the cytoplasm. Additionally, lymphocytes were induced to move into the treated area suggesting a mild, nonthermally mediated inflammatory reaction.

Photoprotection

In vitro, it has been shown that non-coherent, near-infrared radiation protects human dermal fibroblasts from solar UV toxicity (Menezes et al., 1998). In humans, LED treatment reduced UV-B induced erythema in 85% of subjects with evidence of a dose-related pattern in results. A sun protection factor SPF-15-like effect and a reduction in post-inflammatory hyperpigmentation were observed on the pre-treated side (Barolet & Boucher, 2008). 

Nasolabial fold

Only one small trial reported an improvement in the appearance of the nasolabial folds (Baez & Reilly, 2007). It used alternating therapy with red and infrared wavelengths from LED light sources (9 treatments over 5 weeks). The results for softening of the nasolabial fold peaked at the 9-week followup when 25% of subjects reported a moderate response (50% softening) and 75% reported a slight response (25%). 9% of participants rated the overall effect on the nasolabial fold as good, 55% moderate, 27% no effect and 9% poor. 

Fine lines/wrinkles

Ten trials reported a reduction of fine lines and/or wrinkles (Weiss et al., 2006; Shaoul & Mulholland, 2011; Wunsch & Matuschka, 2014; Stirling & Haslam, 2007; Migliardi et al., 2009; Fournier et al., 2006; Lee et al., 2007; Barolet et al., 2009; Bhat et al., 2005; Nikolis et al., 2016). A variety of light sources and wavelengths were used in various combinations. The magnitude of the benefit was often not measured. In the cases in which it was measured, it was about a 30% reduction (although some trials reported up to a 50% reduction of wrinkles in some patients).

Subjective assessments showed stronger effects than clinical assessment across several studies. The likelihood of experiencing a reduction in fine lines and wrinkles was very high with 90% of participants reporting this benefit in most cases. The duration of the benefit is at least moderate, lasting for 12 weeks post-treatment at a minimum. 

In a retrospective analysis of 300 patients treated by an LED system with a pulsed 590nm wavelength, the reduction in wrinkles ranged from significant to subtle. However, an improvement was noted by 90% of participants. In the same study, 60% of patients were found by clinical assessment to demonstrate an improvement in fine lines (Weiss et al., 2006). Similarly, blinded photographic assessment in another study found 59% of subjects showed clinical improvement in fine lines and wrinkles (Bhat et al., 2005). 

A study using a handheld device with red/infrared wavelengths reported that 94% of patients showed considerable or excellent improvement in fine lines/wrinkles at the 3-month follow-up (Shaoul & Mulholland, 2011). 

A study found that about 70% of patients improved, 15% remained the same and 15% worsened following 30 treatments with either red or broad-spectrum light (Wunsch & Matuschka, 2014).

Many volunteers thought that they could identify a difference between the treated and untreated sides (75%) but only 57% correctly identified the improved side as the one which had received treatment (Stirling & Haslam, 2007) raising the possibility of a strong placebo effect. 

A study that compared LED light combined with glycolytic peels with peels alone (8 treatments over 4 weeks) found that LED treatment significantly improved the results. Peak results were achieved one month after the last treatment when 50% of patients and 40% of blinded observers reported a reduction in wrinkles. By 3 months post-treatment, these values had declined to 45% and 20% respectively, suggesting that benefits begin to fade by 3 months (Fournier et al., 2006). 

A study that compared LED with RF alone and combined with RF (8 LED sessions, one every 5 days) found that all patients were either globally satisfied or very satisfied. LED was superior to RF in terms of wrinkle reduction and the combination of both methods led to the greatest improvements (Migliardi et al., 2009).

A head-to-head study that compared various wavelengths on wrinkle reduction found that the combination of 830 and 633 nm was most effective, achieving a reduction of 36% at the 3-month follow up while the lowest reduction was seen in the red light only treatment arm (26%) with infrared-only between (Lee et al., 2007).

One study found that >85% of subjects obtained an improvement in at least one subtype of the Fitzpatrick Classification System. Blinded observers found the degree of improvement was mild to moderate (Barolet et al., 2009). Another study found that subjective analysis of wrinkles favored the use of LED treatment-only over photodynamic therapy (Nikolis et al., 2016). 

Background erythema

Two studies that used a 590 nm pulsed LED device as a light source reported improvements in background erythema (Weiss et al., 2006; Weiss et al., 2005). Redness was improved at one-week post-treatment in about 25% of subjects. This improvement continued to increase over time and peaked at 4 months post-treatment when >40% of patients showed more than a 25% improvement. The benefits began to decline thereafter and by 12 months, only about 12% of patients continued to exhibit a >25% improvement. 

Matrix metalloproteinases/TIMPs

Skin aging is associated with a downregulation in collagen synthesis and an elevation in MMP expression. Two studies, one in vitro (on human reconstructed skin) (660 nm) and the other in vivo (590 nm) reported that treatment with LED resulted in an average decrease in MMP-1 of 18% and 4% respectively (Weiss et al., 2005; Barolet et al., 2009). A third study did not identify a change in levels of MMPs but did find a noticeable increase in TIMP-1 and TIMP-2 which varied according to the wavelength of light tested (Lee et al., 2007). TIMP-1 was most increased in the combined red/infrared treatment group while TIMP-2 increased most with red light treatment.

Stratum corneum hydration & transepidermal water loss

One small trial measured an approximately 15% increase in stratum corneum hydration and a 30% decrease in transepidermal water loss following LED treatment with 630 nm red light (Ji et al., 2014). 

Elastin

Two trials reported an increase in the amount of elastin in response to treatment with combined red and/or infrared light (Ji et al., 2014; Lee et al., 2007). There was a significant increase in the number of elastic fibers in the upper to mid-reticular dermis at 2 weeks post-treatment (8 treatments over 4 weeks) (Lee et al., 2007). Punch biopsies from 5 patients showed an improvement in the denaturation of elastic fibers with an increase in content and more regular distribution (Ji et al., 2014). 

Solar elastosis

Solar elastosis is diagnosed clinically (thickened yellow skin with bumps, wrinkles or furrowing) or microscopically (accumulation of irregularly thickened elastic fibers that degrade to form disorganized structures). It occurs as a result of long term sun exposure and is a feature of photoaging. One trial reported that the typical signs of solar elastosis were improved by 12 sessions of red LED light (Ji et al., 2014) but doesn't provide any indication of the magnitude of the effect. 

Elasticity

Two trials that used combined red/infrared LED systems reported an increase in skin elasticity (Sadick 2008; Lee et al., 2007). 26% of participants reported improved elasticity at the 6-week follow up. This number increased to 47% by 9 weeks and was maintained until the 12-week follow up (Sadick 2008). A second study used a Cutometer to measure changes in elasticity following three different types of LED treatment: red, infrared and combined (Lee et al., 2007). The study found the best results with infrared only treatment (19% average increase) at 3 months post-treatment. The final increase in the red light group was 14% and the combined group, 16%. 

On the other hand, not only did another trial did not find a significant increase in elasticity it identified an initial drop in elasticity. This was followed by an increase (that did not reach significance) to a maximal effect at 8 weeks that dropped again by the 12-week assessment (Bhat et al., 2005). 

Precancerous lesion clearance

Four trials (three that used LED systems with red wavelengths and one that used a broadspectrum light source) reported the clearance of actinic keratosis (AK) in the "light-only" treatment arm of photodynamic therapy studies (Szeimies et al., 2009; Szeimies et al., 2010; Reinhold et al., 2016; Dirschka et al., 2011). Although the PDT treatment arms (light sensitizing substance + light) had much better results than the light only treatment arms, a number of lesions resolved in the light only groups. Between 17.1% and 33% of AKs resolved in various trials. However, many AKs resolve spontaneously and spontaneous complete field regression rates reportedly range from 0-21%. 

Psoriatic plaques

5 trials assessed the use of LLLT in the treatment of psoriatic plaques using blue, red and infrared wavelengths (Kleinpenning et al., 2011; Ablon 2010; Pfaff et al., 2015; Weinstabl et al., 2011; Maari et al., 2003). The first trial compared blue and red light (12 treatments over 4 weeks) and found that both led to clinical improvement with no significant difference between blue (34%) or red (27%) (Kleinpenning et al., 2011). However, the lesions treated with red light did not continue to improve after the 6th session whereas those treated with blue light continued to improve until the end of the study. 

A small trial of the treatment of recalcitrant psoriatic lesions using a combination of red and infrared LED light showed clearance rates of 60-100% and an extremely high level of patient satisfaction (Ablon 2010).

A trial that examined the effect of high-intensity and low-intensity settings on a blue light home treatment device found a significant improvement at both intensities (Pfaff et al., 2015). Blue light (daily for 4 weeks) treatment also resulted in a statistically significant improvement in treated groups as compared to control groups (Weinstabl et al., 2011) in one study while another study on blue light found no significant improvement after exposure to blue light from a fluorescent panel 3 times per week for 4 weeks (Maari et al., 2003). 

Acne

The evidence for the efficacy of LLLT in treating acne, particularly for inflammatory lesions, is substantial (Kwon et al., 2013; Lee et al., 2007; Goldberg & Russell, 2006; Morton et al., 2005; Papageorgiou et al., 2000; Sadick, 2009; Tremblay et al., 2006; Kawada et al., 2002; Sigurdsson et al., 1997; Nestor et al., 2016; Alba et al., 2016; Akaraphanth et al., 2007). The trials used a variety of light sources/devices and wavelengths in all parts of the spectrum with highly differing treatment parameters.

The magnitude of the reduction in inflammatory lesions ranged from 25-78%. The effects on noninflammatory lesions were less pronounced ranging from nonsignificant up to a reduction of 58% (Kwon et al., 2013; Lee et al., 2007; Papageorgiou et al., 2000; Sadick, 2009; Kawada et al., 2002; Sigurdsson et al., 1997; Nestor et al., 2016; Alba et al., 2016).

A decrease in the size and output of sebaceous glands was also reported (Kwon et al., 2013).

Two trials reported that LLLT treatment (blue or flashed broad-spectrum light) increased the rate of healing of acne lesions (Gold et al., 2011; Sadick et al., 2010). Trial results were relatively similar with one day as the average time to improvement in the first trial and 29 hours in the second trial (vs.1.5 days and 45 hours in the placebo arm) (Sadick et al., 2010). The time to clearance was 99 hours in the active group vs. 122 hours in the placebo group (Gold et al., 2011). 

These benefits occurred in the vast majority of patients, were of at least medium duration and are of high importance to patients suffering from acne. 

Risk assessment 

Our analysis identified 14 risks that have appeared in clinical trials to date, mostly mild and of a transient, self-resolving nature. More side effects are associated with blue light treatment than other wavelengths and were reported in disease treatment trials rather than in skin rejuvenation trials. 

Table 4: Risk assessment  

Individual risks that have appeared in clinical trials

Erythema

Erythema is the most common side effect of treatment with LLLT. It was identified as a side effect in 13/57 trials we analyzed (Kwon et al., 2013; Lee et al., 2007; Goldberg & Russell, 2007; Morton et al., 2005; Akaraphanth et al., 2007; Szeimies et al., 2009; Dirschka et al., 2011; Snehal et al., 2006; Sanclemente et al., 2010; Sadick, 2008; Tijoe et al., 2003; Shaoul & Mulholland, 2011; Nikolis et al., 2016). It occurred with several types of light sources in 10 different devices at all tested wavelengths including blue, yellow, red, and infrared. It was most often mild in intensity but affected anywhere from 3-40% of treated subjects. However, all cases of erythema were transient and resolved within hours to a couple of days. 

Hyperpigmentation

Hyperpigmentation is one of the more serious potential side effects of LLLT. It occurred in 12 of the trials in our analysis (Lee et al., 2007; Porges et al., 1988; Rosen et al., 1990; Pathak, 1962; Ramasubramaniam et al., 2011; Sanclemente et al., 2010; Kollias & Baqer, 1984; Kleinpenning et al., 2011; Pfaff et al., 2015; Weinstabl et al., 2011; Kleinpenning et al., 2010; Mahmoud et al., 2010) again with a variety of light sources and different devices. Many of the trials used blue light wavelengths as at least a component of the treatment. The hyperpigmentation was mild and affected anywhere from 8-80% of subjects within a study. The duration was moderate, resolving over weeks to months. 

One trial on visible light (400-700 nm) showed that it can induce a dark brown pigmentation surrounded by erythema in darker skin types that is both darker and more sustained (lasting over 2 weeks) than that induced by UV radiation (Mahmoud et al., 2010). This effect did not occur in a lighter skin type even at the highest doses tested (480 J/cm2). Polychromatic light (390-1700 nm) was also shown to induce pigmentation changes (Kollias & Baqer, 1984). Another trial found that exposure to visible light resulted in immediate pigmentation (that resolved within 24 hours) as well as a delayed tanning reaction (that lasted up to 10 weeks) (Porges et al., 1988). The threshold dose for immediate pigmentation was between 40-80 J/cm2 while that for persistent pigmentation was 80 J/cm2. 

Increased dryness of skin

Skin dryness was reported as a side effect of LLLT in 4 acne trials (Kwon et al., 2013; Morton et al., 2005; Papageorgiou et al., 2000, Kawada et al., 2002). These trials used different light sources but all used wavelengths in the blue part of the spectrum so it is likely that this side effect is related to this specific wavelength. The dryness was mild in intensity, affected about 5% of study participants and resolved without treatment in all cases. 

Pruritus

Pruritus was reported by one acne trial (Morton et al., 2005) that used an LED blue light source. 50% of participants were affected but it was mild and self-resolving. 

Rash

Rash was reported by one acne trial (Papageorgiou et al., 2000) that used fluorescent blue and red light sources. The rash was mild and resolved without treatment. 

Worsening of acne

Aggravation of acne was reported in two trials (Papageorgiou et al., 2000, Kawada et al., 2002) by approximately 10% of participants and was serious enough to lead to their withdrawal from the study in some cases. Again, the trials used different light sources but both had blue light as a component of the treatment.

Desquamation

Desquamation was reported in two trials (Akaraphanth et al., 2007; Sanclemente et al., 2010), one that used blue light and the other, red light. It was mild in intensity and affected 2% of participants in one trial and 60% in the other, smaller trial. The higher incidence was likely related to the blue light and underlying skin disorder (acne). 

Pain

A very low level of pain (1/10) was reported by participants in three trials (Szeimies et al., 2009; Reinhold et al., 2016; Dirschka et al., 2011). The trials involved used red light and between 20-50% of participants were affected. The pain resolved on its own or local anesthetic was applied. Pain occurred in the light treatment-only arm of actinic keratosis trials and was likely due to endogenous activation of protoporphyrin containing compounds.

Burning

A burning sensation was reported by participants in three trials (Szeimies et al., 2009; Dirschka et al., 2011; Kleinpenning et al., 2011). The trials used blue, red, and infrared light and were on actinic keratosis and psoriasis, and the burning was likely due to endogenous activation of protoporphyrin containing compounds. 

Ocular glare, blurred vision, and ocular floaters

Vision-related symptoms were reported by one trial that used white and red light (Nam et al., 2017). Ocular glare appeared with both types of light (6 cases) while blurred vision and ocular floaters occurred in one participant treated with a red light. Symptoms were more frequent in the group treated with red light possibly due to the relatively higher lux of the red light. Ocular glare and blurred vision resolved without treatment while the participant who experienced floaters dropped out of the study. The floaters resolved over a period of 3 weeks without treatment. Lack of skill in the use of eye-protection was observed during the initial period of study in the subjects who eventually developed ocular symptoms. Complete eye protection and comprehensive instructions for using LED devices are necessary to mitigate this risk.

Edema

Edema was reported in the "light only" treatment arm of two photodynamic therapy trials (Sanclemente et al., 2010; Nikolis et al., 2016), one that used blue light and the other, red light. Edema occurred in about 5% of participants, was short-lived and resolved without treatment.

Vesiculation

Vesiculation was reported in the light only treatment arm of one photodynamic therapy trial (Sanclemente et al., 2010). It was transient, mild and resolved without treatment.

Worsening of telangiectasia

One patient reported a worsening of preexisting telangiectasia (Wunsch & Matuschka, 2014). 

Reactivation of scar tissue

One patient reported reddening of a 40-year-old knee scar that resolved on its own after the study (Wunsch & Matuschka, 2014). This likely occurred because of the activation of fibroblasts in the dermis (Table 5).

Risks from preclinical trials

Carcinogenesis

LLLT has been shown to stimulate growth and progression of cancer cells in vitro (Sroka et al., 1999; Sperandio et al., 2013) and in vivo (Frigo et al., 2009; de C. Monteiro et al., 2011). The effect may be biphasic and dose-dependent. In an LLLT study on melanoma in mice (660 nm), low dose treatment reduced tumor size while high dose treatment increased the tumor size (Frigo et al., 2009). LLLT (660 nm, 424 mW/cm2) every other day for 4 weeks in hamsters has been shown to increase the progression and negatively influence the histology of existing tumors (de C. Monteiro et al., 2011). 

IR radiation has also been shown to cause a condition called erythema ab igne that leads to an elevated risk of various types of skin cancer. The irradiances and cumulative dose in all of these cases were very high (Barolet et al., 2016). 

Photoaging

Similar to UVA and UVB, IR radiation has been shown to activate MAP-kinases and to induce gene expression via a retrograde signaling response that differs from that of UV irradiation (Schroeder et al., 2008). One major consequence of this retrograde signaling response is increased expression of MMP-1 (a protease that degrades collagen and elastin fibers) without an increase in its inhibitor (TIMP-1). As the upregulation of MMP-1 and downregulation of TIMP-1 have been shown to play a role in the pathogenesis of photoaging, this indicates that IR radiation may contribute to photoaging of the skin.

IR radiation was shown to increase the actinic damage caused by UVA in guinea pigs (Kligman,1982). IR radiation has also been shown to significantly decrease the antioxidant content of the skin following 5 minutes of exposure (Schroeder et al., 2008). However, in order to get the same dose with the sun, one would have to be exposed for 12 hours in the tropics (Barolet et al., 2016). 

Skin angiogenesis plays an important role in photoaging and near-IR both induces dermal angiogenesis and alters the balance between VEGF and it's inhibitor TSP-2. Acute exposure of human skin to IR radiation also has been shown to increase mast cell number and tryptase expression in human skin in vivo (Cho et al., 2009). 

Most studies showing negative effects of near-infrared radiation used artificial light sources far above the solar irradiance threshold (Barolet et al., 2016). 

Mechanism of action and treatment parameters

Mechanism of action

Once absorption occurs it induces a cascade of intermediate reactions (signal transduction and amplification) that lead to the creation of ATP.

Exposure to visible light causes a response that is primarily photochemical, with the main photoreceptor being cytochrome c oxidase (CCO). CCO has an action spectrum from 580-700 nm that peaks at 630-635 nm.

Exposure to infrared light at 830 nm induces a different initial response that is photophysical in nature (vibrational and rotational changes in the electrons of the atoms in the molecules making up the membranes of the cells) and activates membrane transport mechanisms and cellular exchange. The energy requirements for this are very high so the mitochondria are recruited to create ATP.

Both processes lead to a photoactivated cell that then has three possible modes of action: repair, functional improvement or cell recruitment. Additionally, compromised cells have been shown to respond much more to LLLT than normal cells do (Calderhead & Tanaka, 2017).

Wavelength

Wavelength is the most important parameter in LLLT. Absorption must occur before there can be any reaction. Different wavelengths behave differently in the body because they have different chromophore targets and penetration depths. In general, as wavelength increases, penetration increases. The shorter wavelengths of visible light (blue, green and yellow) penetrate skin poorly in vivo. The difference between yellow (590nm) to red (633nm) is only 43nm but penetration of red light is almost 3.5 orders of magnitude (over 1000 times) higher than that of yellow light. 

The natural chromophores in the skin are melanin, blood, and water. Because of the strong affinity of shorter wavelengths for melanin (mostly in the epidermis) and blood, most of the light is absorbed superficially. At wavelengths beyond 830nm water starts to predominate as a chromophore. 

Table 5 summarizes the effects of various wavelengths on cell types found in the skin. 

Table 5: Effects on cell types by wavelength 

(Adapted from Calderhead & Tanaka, 2017; Calderhead, 2007) 

Light source

The main devices currently in use for LLLT are LED-based systems and filtered polychromatic non-laser light sources such as xenon and incandescent lamps. We identified 36 different devices used in dermatological conditions all of which exhibited positive effects. 

In filtered lamps, the desired color can be obtained by using a narrow bandwidth "cut-on cut-off" filter to obtain a 10-20 nm band at the desired wavelength. However, this dramatically reduces the irradiance. A second possibility is to use a cut-off filter to eliminate shorter wavelengths. The emitted light is still polychromatic and not suitable for any indication requiring wavelength selectivity for the target chromophore. These lamps have been used effectively in many studies but require long exposure times to achieve the desired dose. 

LEDs are quasimonochromatic, emitting all of their photon energy at approximately the rated wavelength and no color filter is required. They can be semi-collimated to decrease the angle of divergence and increase the irradiance but cannot be perfectly focused like a laser. They are inexpensive and can be mounted in large planar arrays to irradiate a large area in a hands-free manner. Due to all of these considerations, LED systems are the preferred light source.

Irradiance (power density) and fluence (energy density)

LLLT is characterized by a biphasic dose-response in which lower doses are often more beneficial than higher ones (Chung et al., 2012). Too low a dose can result in reduced effectiveness and too high a dose can lead to tissue damage. Many studies include negative results that could stem from inappropriate dosing rather than a defect in the treatment itself.

The power density (PD) describes the actual power incident on the tissue per unit area. If the light intensity is lower than the physiological threshold value for a given target, it does not produce photostimulatory effects even when irradiation time is extended. Additionally, photoinhibitory effects may occur at higher doses (Barolet et al., 2016). Fluence (energy density) is the energy in joules per unit area (cm2) and can be regarded as the "dose". 

Distance

Precise positioning is mandatory to ensure optimal beam delivery intensity to achieve maximum effects. LED and filtered lamp systems deliver a divergent cone of light which makes it important to know the distance the manufacturer used to calculate the irradiance. Calculation of irradiance is difficult with LED systems because intersecting LED beams create a phenomenon known as photon interference. Interference is less likely to occur with filtered light sources than with LED systems because many of them incorporate a polarizer in the lens and a significant drop off in intensity occurs with increasing distance.

A greater photon intensity is delivered at a certain distance away from the surface of the LEDs in the array. Therefore, mask-type or handheld devices do not achieve maximum effects. There will be some absorption but it will not be as effective as when the array is placed some centimetres away. One study that measured irradiance at various distances found that the intensity at 10 cm was higher than at the surface and remained high up to a distance of 17cm. Distances of 20 cm and 3 cm resulted in the same irradiance (Calderhead & Tanaka, 2017). 

LED systems using red and infrared wavelengths combine the benefit of photon interference with a powerful scattering effect. Due to these effects, the highest photon intensity is beneath the surface, exactly where it is needed (Calderhead & Tanaka, 2017).

Treatment time

Once an ideal dose in J/cm2 has been determined, then the irradiation time necessary to achieve that dose can be calculated for any system once the irradiance is known, by dividing the dose in J/cm2 by the irradiance expressed in W/cm2. 

Cellular responses take place at different rates. It has been shown that increases in ATP can be observed directly following treatment while cell proliferation requires at least 24 hours (Barolet, 2008). 

Treatment frequency

The effects of different treatment intervals are underexplored although there is enough evidence to suggest it is an important parameter (Chung et al., 2012). In culture, fibroblasts show cyclical patterns of collagen and MMP activity that can be emphasized by treatments every 48 hours (Barolet, 2008).

Treatment duration

Optimal treatment duration and the optimal number of treatments are also unknown. The highest number of treatments in the trials we analyzed was 60 while most studies utilized a total of 8-12 sessions. Determination of the optimal number of sessions is important to avoid overdosing and increased risk of photoaging and possible carcinogenesis. It has been shown that while a single IR treatment increases type I procollagen expression, multiple irradiations reduced its expression (Cho et al., 2009). Also, while one session of IR irradiation didn't induce MMP-1, multiple irradiations did and in mice, irradiation with 30 J/day for 15 weeks was sufficient to induce wrinkle formation. 

Pulsed wave vs continuous wave mode

The influence of continuous wave versus pulsed wave mode as well as precise pulsing parameters (ie. duration, interval, pulse per train, pulse train interval) on cellular response has not been fully studied and comparative studies have shown conflicting results (Barolet, 2008). Pulses may travel deeper into tissues than continuous-wave radiation because the first part of a powerful pulse may contain enough photons to occupy all chromophores in the upper tissue layers, leaving the remainder to pass further into the tissue. 

State of cells and tissues

The magnitude of PBM depends on the condition of the cell at the moment of irradiation. Compromised cells respond more readily to LLLT. For example, light stimulates cell proliferation only if cells are growing poorly at the time (Barolet, 2008).

Safety

Many LLLT devices have been commercialized without FDA or other medical regulatory approval because the light output is below a nominal hazard level. 

Table 6: Skin benefits by device