Abstract
The use of low-light laser therapy to treat androgenetic alopecia is a promising modality to restore hair growth. However, the effect of skin color on response to laser therapy for hair growth has not been systematically explored in the literature. The objective of this study is to systematically assess through a comprehensive literature search of the MEDLINE database whether skin type data were collected in clinical trials and analyzed in each study and determine if we can estimate an effect. 10/22 studies have defined inclusion criteria as Fitzpatrick skin types I–IV. No studies mentioned effects on darker skin types, Fitzpatrick skin types V-VI. Only 5/10 studies had statistical data on efficacy depending on Fitzpatrick skin type, with four showing no effect and one showing a significant positive effect with darker skin types having faster rate of hair growth. There are not enough data to conclude whether skin type effects laser-induced hair growth in androgenic alopecia. The studies are severely lacking in sample size. One showed a potential effect. Importantly, there are no data on black or brown skin colors. Development of optimal laser irradiating wparameters through the prediction of personalized absorbance based on skin color measurement is needed.
Introduction
As one of the most common hair disorders in the population, androgenetic alopecia (AGA), also known as male or female pattern baldness, is characterized by transformation of terminal hair into vellus hair as mediated by dihydrotestosterone [1]. The psychological impact of AGA is profound and well documented in the literature [2‒10]. The prevalence of AGA is high in white men and women, approaching an incidence of 50% in men aged 40–49 and 50% in women throughout their lifetime [11‒13]. Few studies have investigated the incidence of AGA in other races, which cumulatively suggest that African Americans and Asian males have lower rates of AGA than Caucasians [14‒16]. Similar data are seen among women with AGA with higher rates of AGA being found in Caucasians [17, 18]. However, these studies did not include a large sample size and collected data from younger aged patients. There is an overall lack of data on AGA in darker skin tones compared to lighter skin tones.
Low-light laser therapy (LLLT) has demonstrated safety and efficacy in treating androgenic alopecia in both men and women [19]. However, the effect of skin color on hair regrowth using laser therapy has not been explored in the literature. LLLT has largely been indicated for use in patients with Fitzpatrick type I to IV according to FDA approval with no mention of darker skin types [20, 21]. While side effects are uncommon in patients using LLLT, patients with darker skin types are theoretically at increased risk for undesired side effects of laser therapy due to melanin’s wide absorption spectrum [22]. One study found that high LLLT doses delivered with a strong 3B laser (200 mW) do increase skin temperature significantly in darker skin types versus lighter skin, and these photothermal effects can induce intolerable pain in patients with dark skin color [23]. However, this study used higher irradiation doses than those used to treat AGA. The objective of this study is to systematically assess through a comprehensive literature search of the MEDLINE database whether skin type data were collected and analyzed in clinical trials using LLLT technology and if we can estimate an effect skin color has on laser therapy.
Methods
A comprehensive literature search of the MEDLINE database was performed from the database’s inception through June 1, 2022. Search terms included “laser” or “light” and “hair.” The full text was assessed if the title and abstract did not include enough information for exclusion and inclusion criteria to be applied. The full text of the remaining articles was then evaluated for inclusion.
Eligibility Criteria
Results were filtered to include only clinical trials. The titles and abstracts of all articles were initially screened for potential inclusion. Articles were included if they reported data on the use of laser or light-based therapies to improve the growth of hair. Exclusion criteria included abstracts and review articles. No date or language restrictions were applied. The full text was then evaluated for description of Fitzpatrick skin type (FST) or race in patient sample or inclusion criteria.
Search Strategy
The final search terms were “laser” OR “light” AND “hair.” This search resulted in 194 articles. After screening of titles and abstracts, 172 articles were eliminated due to being unrelated to AGA or LLLT, basic research articles, and reviews. The full text of the remaining 22 articles was reviewed by two authors for mention of FST as an inclusion criteria. This process excluded 12 of these articles and left 10 for analysis.
Results
Ten of the 22 included studies reported inclusion criteria as FST I–IV (Table 1). Twelve studies did not mention skin type or color in their study [24‒35]. No studies mentioned effects on darker skin types, FST V-VI. All studies demonstrated a positive effect of laser therapy on treatment of AGA (Table 1). Of the 10 studies that reported FST, only 5 included a statistical model (although not primary analysis) that accounted for the effect of FST on laser efficacy, with 4 showing no difference among darker skin types and 1 showing a significant positive effect with darker skin types having faster rate of hair growth (Table 2). Five of 10 studies analyzed reported no side effects during trial runs. Five trials reported side effects including headache, pruritus, burning, eczema, acne, irradiation-induced inflammation, scalp tenderness, and paresthesia at treatment site (Table 2) [36‒40].
Summary of clinical trials evaluating LLLT to treat androgenic alopecia with FST type included as inclusion criteria
| Study . | N . | Intervention and control . | Follow-up time . | Statistical test . | Outcomes . |
|---|---|---|---|---|---|
| Blum et al. [41] | 119 | X5 HairLaser with 4.5-mw diodes emitting 650 nm versus placebo | 26 weeks | RM-ANOVAPost hoc linear trend test | Significant increase in mean hair counts (F = 10.60; p < 0.0001)Increased number of hair count in darker skin type (FSK II and IV) versus lighter skin types (FSK I) |
| Faghihi et al. [36] | 50 | 20 drops of topical minoxidil 5% solution plus LLLT (using LDU 8024PN/8024BN with a 10–50 mw power and a 785-nm wavelength versus only topical minoxidil 5% solution | 12 months | Paired t testRM-ANOVAIndependent-sample t test | Significantly higher mean increase in hair count (78.3% vs. 51.3%; p < 0.001) and hair diameter (45.4% vs. 32.3%; p = 0.002) at 12 monthsSignificant improvement in patient satisfaction with the treatment (p < 0.001) |
| Mai-Yi Fan et al. [37] | 100 | iRestore ID-520 (WELLMIKE Technology Corp., New Taipei City, Taiwan) with a 650/660 nm light versus sham device | 24 weeks | Paired t testWilcoxon signed-rank test | Significantly higher hair count on LLLT-treated side (105.6 vs. 99.0 [p < 0.001])Significantly higher hair coverage on LLLT-treated side (14.2% vs. 11.8%; p < 0.001)Significant increase in hair thickness (2.3±4.3 μm vs. −1.3±3.5 μm [p < 0.001]) |
| Ferrara et al. [44] | 21 | LLLT (using a modified Capellux® delivering 5 mW of 660 nm light), followed by topical minoxidil 5% application versus topical minoxidil 5% only | 6 months | Paired t testWilcoxon signed-rank testKruskal-Wallis test | No statistical difference in hair growth noted between treatment and control groups (p > 0.05) |
| Freidman and Schnoor [43] | 44 | Novel device containing 272, 5-mW diode lasers and operating at 650 nm versus sham device | 17 weeks | ANOVAWilcoxon signed-rank testUnequal variance t test | Significant increase in terminal hair counts (51% increase in treatment group [p < 0.001]) |
| Jimenez et al. [38] | 334 | HairMax Lasercomb® (6, 7, or 9 beams at 655 nm or 12 beams at 635 nm) versus sham device | 26 weeks | ANCOVACochran-Mantel-Haenszel row mean score testCochran’s Q testDerSimonian-Laird approach for meta-analysis | Statistically significant improvement in terminal hair density with LLLT compared with sham treatment (p < 0.0001 in both female trials, p = 0.0017, p = 0.0249, and p = 0.0028, for the 7-, 9-, and 12-beam lasercombs in males, respectively)Least-squares mean difference of change in terminal hair density of 15.27 (standard error 1.781) at 26 weeks (p < 0.0001) |
| Lanzafame et al. [42] | 41 | “TOPHAT655” unit (Apira Science, Inc, Boca Raton, FL) containing 20.5-mW lasers, and 31 LEDS both operating at 655 nm versus sham device | 16 weeks | ANOVAWilcoxon signed-rank testUnequal variance t test | Significantly increased hair count in treatment group (67.2% increase from baseline in the treatment group vs. 32.3% in control, p = 0.003) |
| Lanzafame et al. [45] | 42 | “TOPHAT655” unit (Apira Science, Inc, Boca Raton, FL, USA) containing 21.5-mW lasers and 30 LEDS both operating at 655 nm versus sham device | 16 weeks | ANOVAWilcoxon signed-rank testUnequal variance t test | Significantly increased hair count in treatment group (48.07% increase from baseline in the treatment group vs. 11.05% in control, p < 0.001) |
| Leavitt et al. [39] | 123 | HairMax LaserComb® (emulates 9 beams at a wavelength of 655 nm) versus sham device | 26 weeks | Two-sample t testsPearson’s χ2 testCochran-Mantel-Haenszel row mean score testANCOVAFisher’s exact test | Significantly greater increase in mean terminal hair density in LLLT-treated scalp (+19.8 hairs/cm2 vs. −7.6 hairs/cm2 in control)Improved subject perception of new hair growth in LLLT-treated scalp versus control (p = 0.01)The ANCOVA was 18.6 hairs/cm2 (hair density) |
| Pestana et al. [40] | 9 | 2% topical minoxidil solution and UVB with six 20-watt (FS20) bulbs or 2% topical minoxidil solution and PUVA (0.1% topical psoralen solution followed by UVA light 14 fluorescent UVA bulbs that delivered 8 mW/cm z at 25 cm 1 h later) versus 2% topical minoxidil solution only | 4 months | Mann-Whitney rank-sum testKendall’s correlation coefficient | Significant increase in nonvellus hair growth in PUVA + minoxidil-treated scalp (227.5 vs. 177.3)No significant increase in overall clinical response between PUVA and UVB (p = 0.25) and UV light therapy versus minoxidil alone (p = 0.59) |
| Study . | N . | Intervention and control . | Follow-up time . | Statistical test . | Outcomes . |
|---|---|---|---|---|---|
| Blum et al. [41] | 119 | X5 HairLaser with 4.5-mw diodes emitting 650 nm versus placebo | 26 weeks | RM-ANOVAPost hoc linear trend test | Significant increase in mean hair counts (F = 10.60; p < 0.0001)Increased number of hair count in darker skin type (FSK II and IV) versus lighter skin types (FSK I) |
| Faghihi et al. [36] | 50 | 20 drops of topical minoxidil 5% solution plus LLLT (using LDU 8024PN/8024BN with a 10–50 mw power and a 785-nm wavelength versus only topical minoxidil 5% solution | 12 months | Paired t testRM-ANOVAIndependent-sample t test | Significantly higher mean increase in hair count (78.3% vs. 51.3%; p < 0.001) and hair diameter (45.4% vs. 32.3%; p = 0.002) at 12 monthsSignificant improvement in patient satisfaction with the treatment (p < 0.001) |
| Mai-Yi Fan et al. [37] | 100 | iRestore ID-520 (WELLMIKE Technology Corp., New Taipei City, Taiwan) with a 650/660 nm light versus sham device | 24 weeks | Paired t testWilcoxon signed-rank test | Significantly higher hair count on LLLT-treated side (105.6 vs. 99.0 [p < 0.001])Significantly higher hair coverage on LLLT-treated side (14.2% vs. 11.8%; p < 0.001)Significant increase in hair thickness (2.3±4.3 μm vs. −1.3±3.5 μm [p < 0.001]) |
| Ferrara et al. [44] | 21 | LLLT (using a modified Capellux® delivering 5 mW of 660 nm light), followed by topical minoxidil 5% application versus topical minoxidil 5% only | 6 months | Paired t testWilcoxon signed-rank testKruskal-Wallis test | No statistical difference in hair growth noted between treatment and control groups (p > 0.05) |
| Freidman and Schnoor [43] | 44 | Novel device containing 272, 5-mW diode lasers and operating at 650 nm versus sham device | 17 weeks | ANOVAWilcoxon signed-rank testUnequal variance t test | Significant increase in terminal hair counts (51% increase in treatment group [p < 0.001]) |
| Jimenez et al. [38] | 334 | HairMax Lasercomb® (6, 7, or 9 beams at 655 nm or 12 beams at 635 nm) versus sham device | 26 weeks | ANCOVACochran-Mantel-Haenszel row mean score testCochran’s Q testDerSimonian-Laird approach for meta-analysis | Statistically significant improvement in terminal hair density with LLLT compared with sham treatment (p < 0.0001 in both female trials, p = 0.0017, p = 0.0249, and p = 0.0028, for the 7-, 9-, and 12-beam lasercombs in males, respectively)Least-squares mean difference of change in terminal hair density of 15.27 (standard error 1.781) at 26 weeks (p < 0.0001) |
| Lanzafame et al. [42] | 41 | “TOPHAT655” unit (Apira Science, Inc, Boca Raton, FL) containing 20.5-mW lasers, and 31 LEDS both operating at 655 nm versus sham device | 16 weeks | ANOVAWilcoxon signed-rank testUnequal variance t test | Significantly increased hair count in treatment group (67.2% increase from baseline in the treatment group vs. 32.3% in control, p = 0.003) |
| Lanzafame et al. [45] | 42 | “TOPHAT655” unit (Apira Science, Inc, Boca Raton, FL, USA) containing 21.5-mW lasers and 30 LEDS both operating at 655 nm versus sham device | 16 weeks | ANOVAWilcoxon signed-rank testUnequal variance t test | Significantly increased hair count in treatment group (48.07% increase from baseline in the treatment group vs. 11.05% in control, p < 0.001) |
| Leavitt et al. [39] | 123 | HairMax LaserComb® (emulates 9 beams at a wavelength of 655 nm) versus sham device | 26 weeks | Two-sample t testsPearson’s χ2 testCochran-Mantel-Haenszel row mean score testANCOVAFisher’s exact test | Significantly greater increase in mean terminal hair density in LLLT-treated scalp (+19.8 hairs/cm2 vs. −7.6 hairs/cm2 in control)Improved subject perception of new hair growth in LLLT-treated scalp versus control (p = 0.01)The ANCOVA was 18.6 hairs/cm2 (hair density) |
| Pestana et al. [40] | 9 | 2% topical minoxidil solution and UVB with six 20-watt (FS20) bulbs or 2% topical minoxidil solution and PUVA (0.1% topical psoralen solution followed by UVA light 14 fluorescent UVA bulbs that delivered 8 mW/cm z at 25 cm 1 h later) versus 2% topical minoxidil solution only | 4 months | Mann-Whitney rank-sum testKendall’s correlation coefficient | Significant increase in nonvellus hair growth in PUVA + minoxidil-treated scalp (227.5 vs. 177.3)No significant increase in overall clinical response between PUVA and UVB (p = 0.25) and UV light therapy versus minoxidil alone (p = 0.59) |
ANCOVA, analysis of covariance; ANOVA, analysis of variance.
Summary of reported statistical effect of FST type on results in selected clinical trials
| Study . | FST inclusion criteria . | Distribution of participants by FST in experimental group (N) . | FST-based outcome . | Description of adverse effects present/stratified by FST . |
|---|---|---|---|---|
| Blum et al. [41] | FST I–IV | LLLT-treated cohortFST I: 7FST II: 20FST III: 27FST IV: 16 | Among the 4 FST, FST IV exhibits the strongest pattern of hair counts increasing linearly over time with LLLT treatmentThe more sensitive the skin type is, the less average hair counts are, at the baseline (FST I being the most sensitive and FST IV being the least sensitive)FST I: F = 0.66 (p = 0.6599)FST II: F = 5.19 (p = 0.0003), slope = 1.791 (R2 = 0.0083)FST III: F = 2.49 (p = 0.0346), slope = 0.953 (R2 = 0.0025)FST IV: F = 4.37 (p = 0.0008), slope = 1.846 (R2 = 0.0169) | Study reported no severe adverse events among patients |
| Faghihi et al. [36] | FST I–IV | Does not report distribution of participants by FST | Study reported headache, burning, and itching with treatment but did not stratify by FST | |
| Mai-Yi Fan et al. [37]* | FST I–IV | Total studyFST I: 0FST II: 0FST III: 31FST IV: 69 | No formal statistical analysis on effect of FST was reported | Study reported pruritus, eczema, and acne after treatment but did not stratify by FST |
| Ferrara et al. [44] | FST I–IV | LLLT-treated cohortFST I: 0FST II: 1FST III: 13FST IV: 5ControlFST I: 0FST II: 1FST III: 13FST IV: 5* | No formal statistical analysis on effect of ST. Authors state in discussion that they hypothesize a higher proportion of patients with higher skin pigmentation (FST III and IV) possibly caused interference with light penetration | Study reported no adverse events among patients |
| Freidman and Schnoor [43] | FST I–IV | LLLT-treated cohortFST I: 0FST II: 5FST III: 17FST IV: 0ControlFST I: 0FST II: 4FST III: 17FST IV: 1 | When included in a third sensitivity model, FST was not significant (p = 0.397) | Study reported no severe adverse events among patients |
| Jimenez et al. [38] | FST I–IV | LLLT-treated cohortFST I: 8FST II: 48FST III: 61FST IV: 32ControlFST I: 2FST II: 23FST III: 34FST IV: 17 | FST was included as a categorical variable in ANCOVA. No formal statistical analysis on effect of FST was reported. The homogeneity assessment results were nonsignificant (p = 0.6188) | Study reported dry skin, pruritus, irradiation, scalp tenderness, and warm sensation with treatment but did not stratify by FST |
| Lanzafame et al. [42] | FST I–IV | Does not report distribution of participants by FST | Study reported no adverse events among patients | |
| Lanzafame et al. [45] | FST I–IV | LLLT-treated cohortFST I: 4FST II: 6FST III: 12FST IV: 2ControlFST I: 3FST II: 3FST III: 12FST IV: 0 | When included in a third sensitivity model, FST was not significant (p = 0.939) | Study reported no adverse events among patients |
| Leavitt et al. [39] | FST I-IV | Total studyFST I: 4FST II: 17FST III: 65FST IV: 37 | FST was included as a categorical variable in ANCOVA. No formal statistical analysis on effect of FST was reported. Mean change from baseline of terminal hair density was adjusted for study center, subject’s age, and FST | Study reported mild paresthesia and urticaria with treatment but did not stratify by FST |
| Pestana et al. [40] | FST II–III | UVB cohortFST I: 0FST II: 0FST III: 5FST IV: 0PUVA cohortFST I: 0FST II: 3FST III: 2FST IV: 0 | No formal statistical analysis on effect of FST was reported | Study reported erythema, tenderness, dry skin with treatment but did not stratify by FST |
| Study . | FST inclusion criteria . | Distribution of participants by FST in experimental group (N) . | FST-based outcome . | Description of adverse effects present/stratified by FST . |
|---|---|---|---|---|
| Blum et al. [41] | FST I–IV | LLLT-treated cohortFST I: 7FST II: 20FST III: 27FST IV: 16 | Among the 4 FST, FST IV exhibits the strongest pattern of hair counts increasing linearly over time with LLLT treatmentThe more sensitive the skin type is, the less average hair counts are, at the baseline (FST I being the most sensitive and FST IV being the least sensitive)FST I: F = 0.66 (p = 0.6599)FST II: F = 5.19 (p = 0.0003), slope = 1.791 (R2 = 0.0083)FST III: F = 2.49 (p = 0.0346), slope = 0.953 (R2 = 0.0025)FST IV: F = 4.37 (p = 0.0008), slope = 1.846 (R2 = 0.0169) | Study reported no severe adverse events among patients |
| Faghihi et al. [36] | FST I–IV | Does not report distribution of participants by FST | Study reported headache, burning, and itching with treatment but did not stratify by FST | |
| Mai-Yi Fan et al. [37]* | FST I–IV | Total studyFST I: 0FST II: 0FST III: 31FST IV: 69 | No formal statistical analysis on effect of FST was reported | Study reported pruritus, eczema, and acne after treatment but did not stratify by FST |
| Ferrara et al. [44] | FST I–IV | LLLT-treated cohortFST I: 0FST II: 1FST III: 13FST IV: 5ControlFST I: 0FST II: 1FST III: 13FST IV: 5* | No formal statistical analysis on effect of ST. Authors state in discussion that they hypothesize a higher proportion of patients with higher skin pigmentation (FST III and IV) possibly caused interference with light penetration | Study reported no adverse events among patients |
| Freidman and Schnoor [43] | FST I–IV | LLLT-treated cohortFST I: 0FST II: 5FST III: 17FST IV: 0ControlFST I: 0FST II: 4FST III: 17FST IV: 1 | When included in a third sensitivity model, FST was not significant (p = 0.397) | Study reported no severe adverse events among patients |
| Jimenez et al. [38] | FST I–IV | LLLT-treated cohortFST I: 8FST II: 48FST III: 61FST IV: 32ControlFST I: 2FST II: 23FST III: 34FST IV: 17 | FST was included as a categorical variable in ANCOVA. No formal statistical analysis on effect of FST was reported. The homogeneity assessment results were nonsignificant (p = 0.6188) | Study reported dry skin, pruritus, irradiation, scalp tenderness, and warm sensation with treatment but did not stratify by FST |
| Lanzafame et al. [42] | FST I–IV | Does not report distribution of participants by FST | Study reported no adverse events among patients | |
| Lanzafame et al. [45] | FST I–IV | LLLT-treated cohortFST I: 4FST II: 6FST III: 12FST IV: 2ControlFST I: 3FST II: 3FST III: 12FST IV: 0 | When included in a third sensitivity model, FST was not significant (p = 0.939) | Study reported no adverse events among patients |
| Leavitt et al. [39] | FST I-IV | Total studyFST I: 4FST II: 17FST III: 65FST IV: 37 | FST was included as a categorical variable in ANCOVA. No formal statistical analysis on effect of FST was reported. Mean change from baseline of terminal hair density was adjusted for study center, subject’s age, and FST | Study reported mild paresthesia and urticaria with treatment but did not stratify by FST |
| Pestana et al. [40] | FST II–III | UVB cohortFST I: 0FST II: 0FST III: 5FST IV: 0PUVA cohortFST I: 0FST II: 3FST III: 2FST IV: 0 | No formal statistical analysis on effect of FST was reported | Study reported erythema, tenderness, dry skin with treatment but did not stratify by FST |
ANCOVA, analysis of covariance.
*Study conducted half-head study (one half received treatment while other half acted as control).
The only [41] study reporting primary outcome stratified by FST (Table 2) found statistically significant differences among the 4 documented skin types (FST I–IV) both in baseline hair counts and improvement at study visits. At baseline, lighter skin types such as FST I, also termed “more sensitive” types, were found to have lower average hair counts than darker, or “less sensitive,” types (Table 2). FST IV patients were found to have the strongest positive correlation between average hair count and length of treatment (F = 4.37; p = 0.0008) in repeated measures analysis of variance and slope = 1.846 (R2 = 0.0169; p = 0.0004) in post hoc linear trend. Contrarily, FST I was found to have no significant change in hair counts or linear trend over time. These results suggest that darker FST may have less severe disease at baseline and are more likely to respond to LLLT therapy than lighter FST.
Study Characteristics
Ten total studies were selected for analysis including 8 clinical trials and 2 pilot studies. LLLT-delivering devices included the TOPHAT655 [42, 43], HairMax LaserComb [38, 39], X5 HairLaser [36], iRestore ID-520 [37], Capellux®[44], LDU 8024PN/8024BN [36], and a novel device [43]. Eight out of ten studies utilized wavelengths between 635 and 660 nm with 655 nm being the most common in 4 of those studies. One study utilized 785-nm wavelength [36] and another utilized UV light of unknown wavelength [40]. Follow-up and treatment length ranged from 16 weeks to 12 months. The majority of studies had treatment and follow-up for 16–26 weeks [36, 37, 38, 39, 42, 43, 45].
Objective response to therapy was evaluated using a variety of methods including global scalp photography, phototrichogram assessment, the investigator’s global assessment of hair regrowth, hair clippings, and computer-aided hair counts. Subjective evaluation of hair growth by author and patient was also utilized. A variety of outcomes were assessed with the most common being hair count or density and hair thickness or diameter.
Conclusion
In this review, we evaluated the prevalence of data concerning FST in studies concerning LLLT treatment of AGA. LLLT appears to be an effective treatment for androgenic alopecia [19]. Skin phototype is often a key factor in choosing whether laser therapy is an appropriate option for patients due to the structural differences in skin with different pigmentation levels. Melanin in the epidermis functions as a photoreceptive chromophore, allowing for higher retention of light therapy which can result in increased side effects in patients with more pigmented skin [22, 46, 47]. The FST scale is a widely used tool in dermatology that rates melanin concentration from FST I (very fair) to FST VI (dark skin). This scaling system is a subjective evaluation based on the individual’s genetic disposition, reaction to sun exposure, and tanning habits [48].
In our analysis, 10 out of 22 articles selected for analysis included FST as an inclusion criterion in their studies. All 10 studies selected for lighter complected patients with inclusion criteria requiring FST I–IV. No study included patients with FST V or VI. Out of these 10 studies, 8 described the distribution of FST among participants. Four studies found no statistical significance of FST on results, and one study demonstrated that higher FST had a greater response to LLLT therapy, even though their baseline disease was less severe (or maybe their baseline hair count was higher) [41]. However, the sample size for the specific FST was variable and low. These results do not support to hypotheses that darker skin type would have a decreased effect due to decreased penetration of light therapy [44]. Instead, they might be indicative that more melanin causes increased absorption of light therapy [49, 50]. Further studies looking into the effect of LLLT on darker skin, including FST V or VI, are needed to better illustrate this relationship.
Most studies to date, including those used in this study (Table 1), have shown a beneficial effect of an array of low-light laser therapies on treatment of AGA [19]. It is likely that these light therapies will be used on skin of darker colors than tested in the studies of this review. Therefore, it is pertinent for clinicians and researchers to be aware of the potential side effects of LLLT on skin of color. For other laser therapies, such as early laser hair removal therapy, the increased risk of complications in darker skin types (FST III–VI) was mitigated by modulating the wavelength of the laser used while maintaining hair removal efficacy [22, 51‒56]. While the reported adverse effects of LLLT are minor, they may lead to poor patient compliance with therapy. Therefore, it is essential clinicians and researchers develop the optimal laser parameters in hair regrowth strategies to help control efficacy and side effects. Our analysis showed that no trial stratified the prevalence of side effects by FST (Table 2). Importantly, it is necessary to test the effectiveness of LLLT on patients with darker skin types with rigorous trials in order to provide the best evidence-guided treatment of AGA to patients of color.
Statement of Ethics
All work involved in the generation of this manuscript abide by all ethical guidelines as internationally accepted standards for research practice and reporting.
Conflict of Interest Statement
The authors have no conflicts of interest to declare.
Funding Sources
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Author Contributions
N.B. conceived the review and coordinated writing. D.C., N.B., and C.C.M. were involved in literature search and writing and had final approval of the submitted and published versions of the manuscript.
Data Availability Statement
All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.
