Ruby laser-assisted hair removal reduces the coarseness of regrowing hairs: fallacy or fact?

British Journal of Plastic Surgery (1999), 52, 3 384 1999 The British Association of Plastic Surgeons Ruby laser-assisted hair removal reduces the coarseness of regrowing hairs: fallacy or fact? S. H. Liew, K. Ladhani, A. O. Grobbelaar, D. T. Gault, R. Sanders, C. J. Green and C. Linge RAFT Institute of Plastic Surgery, Mount Vernon Hospital, Northwood, Middlesex, UK SUMMARY. There have been anecdotal reports that hairs that regrow after ruby laser-assisted hair removal are finer in appearance. If true, this phenomenon adds to the improved aesthetic effect of laser treatment of unwanted hair. It is the aim of this study to determine whether this phenomenon indeed occurs, and if so, assess its permanence and its mode of action. In this prospective clinical study, 71 patients with 94 treatment sites were treated with the Chromos 694 Depilation Ruby Laser. Hair diameter was measured pre-treatment, and at 3 and 7 months post-treatment. In addition, ex vivo scalp skin was used to assess if the ruby laser selectively damaged coarser hairs. Laser-treated and matched untreated skin samples were histologically assessed and the diameters of hair shafts (normal or obviously damaged) were measured. Results of this study were analysed using Kruskal Wallis one-way analysis. There was no statistically significant difference between the hair diameter of non-lasered specimens and the hair diameter of the normal hair in lasered specimens. However, a statistically significant difference was seen between the hair diameter of nonlasered specimens and diameters of damaged hair in lasered specimens (P < 0.05). There was a statistically significant difference (P < 0.05) between pre-treatment and 3 month hair diameters, but no statistically significant difference was found between pre-treatment and 7 month hair diameters. In conclusion, ruby laser-assisted hair removal results in a temporary reduction in hair diameter of regrowing hair. This is not due to the selective targeting of larger hair follicles. Keywords: ruby laser, hair removal. Ruby laser-assisted hair removal has been used recently with encouraging results. 1 8 While the exact mechanism of hair destruction is not known, it is generally accepted that melanin in hair follicles acts as a chromophore, which is selectively targeted by the ruby laser with a wavelength of 694 nm through the process of selective photothermolysis. 9 While the ruby laser has shown promising results in removing unwanted body hair, it has become clear that, at least after a single treatment, absolute hair removal is not achieved in every patient treated. In addition to reducing hair density, several investigators have also observed that the regrowing hairs in ruby laser-treated sites appear thinner. 1,10 This has significant clinical implications, because despite regrowth of hair, finer hair often improves the appearance of hirsutism visually. 11 The mechanisms underlying this phenomenon and also its durability are currently unknown. This phenomenon could be caused in several ways. It is possible that laser-assisted hair removal targets coarser hairs. This theory is plausible for two reasons. First, coarser hairs often have higher melanin content (i.e. of the laser chromophore) than finer hairs. Secondly, coarser hairs lose heat more slowly than finer hairs because of their smaller surface area-tovolume ratio. Thus heat would be concentrated adjacent to the viable components of the hair follicle for longer, rather than being rapidly dissipated into the surrounding skin and blood supply. Given that the theoretical mechanism of laser-assisted hair removal is selective photothermolysis, one can see how these two factors could result in greater destruction of coarser hairs. Alternatively, regrowing hair in laser-treated sites may become thinner due to permanent destruction of some of the progenitor hair cells that give rise to new hair formation. It may be, however, that regrowing hairs simply appear thinner if they are still in an early stage of development before they have achieved their full mature diameter. It was the aim of this study to determine scientifically if ruby laser irradiation results in decreased diameter of regrowing hair, and if so, if this effect was permanent. This study also sought to determine if there is selective destruction of coarser hairs by ruby laser-assisted hair removal. A prospective clinical study and an ex vivo study on scalp skin were performed to achieve these goals. Materials and methods For the ex vivo study, fresh scalp-skin specimens were obtained from facelift operations. Six pieces of scalp skin were obtained from six different Caucasian patients with dark-coloured hair. Each piece of scalp 3

Effect of ruby laser-assisted hair removal on hair diameter 381 skin was divided into three pieces and used within 1 h of removal from the patients. One piece was not lasered, one piece was lasered at 14 J cm 2 and one lasered at J cm 2. The samples were fixed immediately in formal saline after laser treatment and processed for routine wax impregnation histology. The specimens were sectioned tangentially at 0 µm from the skin surface and were stained with the Modified Sacpic staining technique. 12 This staining enabled easy identification of both normal and obviously damaged hair in the specimens. Hair diameters were measured with the aid of a dedicated programme using an image-analysis system (Seescan, UK) that was calibrated by a stage micrometer. The images of the sections were projected onto the Seescan. The investigator identified normal and laser-damaged hairs and the transverse diameters of the hairs were measured. The distribution of hair diameters for the non-lasered specimens, distribution of hair diameters of the normal hair in the lasered specimens and damaged hair in the lasered specimens were plotted and analysed statistically using Kruskal Wallis one-way analysis. For the clinical study, 71 healthy volunteers were recruited. All volunteers were Caucasians with Fitzpatrick skin types I III 13 and had dark-coloured, unwanted hair on various body sites. All volunteers were treated with the Chromos 694 Depilation Ruby Laser (SLS/Biophile, Wales, UK). It had a wavelength of 694 nm and was capable of a repetition rate of 1 Hz. The laser was delivered via a flexible fibre-optic cable with a uniform intensity over a 7 mm diameter circle. The range of fluence (energy density) delivered was 8 14 J cm 2. The fluence used was calibrated before each treatment. Both the operator and the patients during the treatment wore protective spectacles. The lowest fluence that caused visible hair destruction (i.e. singeing) without causing excessive skin redness at the time of treatment was used. Hair was collected from treatment sites before treatment by trimming close to the surface of the skin. Twenty hairs were collected from each patient. These were then mounted onto a glass slide using doublesided tape and examined under the microscope. The most proximal hair diameter was measured with the aid of a dedicated programme using an image-analysis system (Seescan, UK) that was calibrated using a stage micrometer. The hair samples were projected onto the screen of the image analyser and the total area of each hair shaft was measured. The average diameter of that section of hair as seen on the screen was calculated by dividing its total area by the length of the hair. The length of hair projected on the screen was 890 µm. The patients were followed up at 3 and 7 months, when hairs were collected and measured as described above. Results In the ex vivo study, six normal, non-lasered scalp skins, and 12 lasered scalp skins (six lasered at 14 J cm 2 and six lasered at J cm 2 ) were examined. Figure 1 Transverse section of lasered scalp skin at 0 µm from the skin surface. The obviously laser-damaged hair follicle is indicated by an arrow. The adjacent two smaller diameter hair follicles appear normal, and their bright-red internal root sheaths indicate that they are in a growing phase. Modified Sacpic staining technique, (keratinised structures stain yellow, internal root sheath bright red, outer border brush end orange, outer root sheath pale green, collagen blue, smooth muscle and erythrocytes green and nuclei blue/black). In laser-treated specimens, some hair follicles were obviously damaged, with charred material around the periphery of the hair shafts and cracking and distortion of the shaft matrix (Fig. 1). However, not all hair follicles were damaged in this way. Damaged follicles appeared in a random pattern throughout the measured areas. This study failed to show any significant differences between treating the scalp skins at 14 J cm 2 and at J cm 2 in terms of the proportion of hair damaged after laser treatment. The percentage of damaged hairs was calculated by dividing the number of obviously damaged hairs as seen in the scalp specimens by the total number of hair follicles. The mean percentage of damaged hairs in the scalp skin when treated at 14 J cm 2 was 46.1% (SD = 28.6) and with J cm 2 was 52.1% (SD = 26.2). There was no statistically significant difference between the two values when assessed with the Wilcoxon Signed Rank test (P = 1.00). This study also failed to show any significant differences between treating the scalp skins at 14 J cm 2 and at J cm 2 in terms of the distribution of normal and damaged hair diameters. The mean diameter of the damaged hairs when the scalp skin was lasered at 14 J cm 2 was 121.11 µm (SD = 41.39) and at J cm 2 was 126.28 µm (SD = 50.01). The results were analysed using Student s t-test, and no statistically significant difference was noted between the two groups (t = 0.234, P = 0.818). Therefore the results of treatment at 14 J cm 2 and J cm 2 were pooled together. The mean hair diameter for non-lasered specimens was 89.32 µm (SD = 49.04), normal hair in lasered specimens was 85. µm (SD = 23.90), and damaged hair in lasered specimens was 121.11 µm (SD = 41.39) (Fig. 2). The results were analysed using the Kruskal Wallis one-way analysis of variance on ranks, and pairwise multiple comparisons were made using

382 British Journal of Plastic Surgery 2 2 0 1 1 Normal Normal (laser-treated) Abnormal (laser-treated) Figure 2 Scatter plot illustrating the distribution of mean hair diameters for six samples of non-lasered scalp skin, and the normal hair and obviously damaged hair present in 12 samples of lasered skin. The bars indicate the standard deviations of the means. 1 Chin Lip Chest Arm Bikini Cheek Back Abdomen Leg Figure 3 Graph showing the distribution of hair diameters in different parts of the body. The bars indicate the standard deviations of the means. Dunnett s method. There was no statistically significant difference between the hair diameter of nonlasered specimens and the hair diameter of the normal hair in lasered specimens. However, a statistically significant difference was seen between the hair diameter of non-lasered specimens and the diameter of damaged hair in lasered specimens (P < 0.05). In the clinical study, 71 subjects with 97 treatment sites were examined. There were 58 females and 13 males, with a mean age of 37 (range 15 66). The areas treated included face, chest, abdomen, back, arms and legs. Sixty-eight sites were followed up for 3 months and sites were followed up for 7 months. The mean diameters of hairs before laser treatment of different parts of the body were plotted (Fig. 3). Hair from the chin had the largest diameter (mean = 99.6 µm, SD = 27.6), followed by hair from the groin (mean = 95.0 µm, SD = 14.8). Hair from the abdomen had the smallest diameter (mean =.7 µm, SD = 12.4). The mean hair diameter before treatment was 82.05 µm (SD = ), at 3 months 70.17 µm (SD = 17.18), and at 7 months 77.46 µm (SD = 17.98) (Fig. 4). The results were analysed using Kruskal Wallis one-way analysis of variance on ranks, and pairwise multiple comparisons were made using Dunnett s method. There was a statistically significant difference between pre-treatment hair diameter and hair diameter at 3 months (P < 0.05) and between hair diameter at 3 months and hair diameter at 7 months (P < 0.05). However, no statistically significant difference was found between pre-treatment hair diameter and hair diameter at 7 months. Discussion Pre-treatment 3 months 7 months Figure 4 Scatter plot showing the distribution of mean hair diameters at pre-treatment, 3 months and 7 months after laser treatment. The bars indicate the standard deviations of the means. After ruby laser irradiation, not all hair follicles were obviously damaged and the selection seemed to be quite random. This is consistent with another study that showed heterogeneous damage to hair follicles after ruby laser irradiation. 1 The average proportion of hair follicles obviously damaged by ruby laser irradiation was 46.1% and 52.0% when treated with 14 J cm 2 and J cm 2, respectively. No significant difference was seen between these two groups. This implies that the fluence used to remove hair may not be as important as the inherent susceptibility of the hair follicles to target selection. It is not yet known why certain hair follicles are more susceptible to laser destruction than others. Larger hair follicles contain more melanin and, according to the theory of selective photothermolysis, 9 will absorb more radiant energy during laser

Effect of ruby laser-assisted hair removal on hair diameter 383 irradiation and dissipate heat less readily. It is therefore feasible that these larger follicles may be preferentially destroyed. However, from the results of our study, this does not seem to be the case. If larger hairs were selectively damaged during ruby laser irradiation, the remaining normal hairs, which were not affected by the treatment, would have a smaller mean hair diameter. On the contrary, the normal hair follicles in the lasered specimens had the same mean hair diameter as the hair follicles in non-lasered specimens, whereas the obviously damaged hairs present in lasered specimens had larger average diameters. The results suggest that the larger diameter of damaged hair follicles is an artefact and may be due to expansion of hair shafts and follicular canals following heat absorption. These observations imply that there is no selective targeting of coarser hairs by laser treatment. This conclusion is supported by the results of our clinical study, which shows that although laser treatment significantly reduced the mean diameter of regrowing hairs at 3 months, the diameter reverted to normal after 7 months. This reduction in hair diameter seen after ruby laser-assisted hair removal is encouraging, since it may result in improvement of hirsutism visually. 11 However, this phenomenon does not seem to be permanent. The thinner hair after laser irradiation is more likely to arise from the immaturity of the regrowing hair before it reaches its full mature diameter. Alternatively, it may also be due to the partial destruction of progenitor cells that give rise to new hair, which recover their full hair-regrowing potential after several months. It is clinically difficult to differentiate regrowing hairs that have survived laser treatment from hairs that were previously dormant and hence escaped being irradiated. Other methods of hair removal have also been shown to affect the hair diameter. Peereboom-Wynia 15 showed that electrolysis resulted in a 0.8% reduction in hair diameter in 11 patients 5 months after electrolysis. Repeated plucking of hair often damages hair follicles, leading to production of finer or thicker hair. 16 Hair may also become thinner after repeated waxing. 16 Shaving, contrary to popular belief, does not affect the hair diameter. 17 In addition, hormonal treatment with dianette (blocking the androgen receptors in hair follicles), was found to reduce hair diameter by up to 39% after 8 12 months treatment. 18 Our observations also suggest that factors other than hair follicle diameter are important in affecting the efficacy of ruby laser-assisted hair removal. Current suggestions are the relative eu- and pheomelanin content of hair follicles, or the growth phases of hair follicles, which are closely associated with melanogenesis. 14 To date, these theories remain unsubstantiated and the true rates of permanent hair destruction after repeated treatments are unknown. We are currently assessing the validity of the correlation between treatment efficacy and the melanin content and growth phases of hair follicles and the feasibility of permanent hair destruction after repeated treatments at quick successive intervals. In conclusion, our results show that in addition to permanent destruction of some of the hair follicles, laser-assisted hair removal temporarily (for up to 3 months) reduces the coarseness of regrowing hairs. Both effects of laser-assisted hair removal can result in the improved appearance of hirsutism. Notwithstanding this, the reason why some hair follicles are affected by laser treatment and not others remains unknown. It is hoped that a better understanding of the mechanisms which underlie this treatment, and thus the factors which affect its efficacy, will lead to the development of ideal treatment regimes for different subsets of hair follicles and patients. Acknowledgements This study was supported initially by a charitable donation from MEHL/Biophile International Corporation and thereafter by the Restoration of Appearance and Function Trust (RAFT) Registered Charity No. 299811. The laser was kindly supplied by SLS Biophile (Wales, UK). References 1. Grossman MC, Dierickx C, Farinelli W, Flotte T, Anderson RR. Damage to hair follicles by normal-mode ruby laser irradiation. J Am Acad Dermatol 1996; 35: 889 94. 2. Grossman MC, Dierickx C, Farinelli B, Geronemus RG, Anderson RR. Permanent hair removal with a normal mode ruby laser. Lasers Surg Med 1997; S9: 146. 3. Dierickx C, Grossman MC, Farinelli WA, Manuskiatti W, Anderson RR. Long-pulsed ruby laser hair removal. Lasers Surg Med 1997; S9: 167. 4. Adrian RM, Tanghetti E. Clinical evaluation of a high-energy long-pulsed ruby laser for the treatment of unwanted body hair. Lasers Surg Med 1997; S9: 166. 5. Sommer S, Burd R, Sheenan-Dare R. Ruby laser treatment for facial hirsutism: early clinical response and patient tolerance. Lasers Surg Med 1997; S9: 161. 6. Lask G, Elman M, Slatkine M, Waldman A, Rozenberg Z. Laser-assisted hair removal by selective photothermolysis. Preliminary results. Dermatol Surg 1997; 23: 737 9. 7. Gold MH, Bell MW, Foster TD, Street S. Long-term epilation using the epilight broad band, intense pulsed light hair removal system. Dermatol Surg 1997; 23: 909 13. 8. Bjerring P, Zacharia H, Lybecker H, Clement M. Evaluation of the free-running ruby laser for hair removal: a retrospective study. Acta Derm Venereol (Stockh) 1998; 78: 48 51. 9. Anderson RR, Parrish JA. Selective photothermolysis: precise microsurgery by selective absorption of pulsed radiation. Science 1983; 2: 524 7. 10. Ueda S, Imayama S. Normal mode ruby laser for treating congenital nevi. Arch Dermatol 1997; 133: 355 9. 11. Rushton H, James KC, Mortimer CH. The unit area trichogram in the assessment of androgen-dependent alopecia. Br J Dermatol 1983; 109: 429 37. 12. Nixon AJ. A method for determining the activity state of hair follicles. Biotechnic Histochem 1993; 68: 316 25. 13. Fitzpatrick TB. The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 1988; 124: 869 71. 14. Slominski A, Paus R. Melanogenesis is coupled to murine anagen: toward new concepts for the role of melanocytes and the regulation of melanogenesis in hair growth. J Invest Dermatol 1993; 101: 90s 97s. 15. Peereboom-Wynia JDR. The effect of electrical epilation on the beard hair of women with idiopathic hirsutism. Arch Derm Res 1975; 254: 15 22. 16. Richards RN, Uy M, Meharg G. Temporary hair removal in patients with hirsutism: a clinical study. Cutis 1990; 45: 199 2. 17. Lynfield YL, Macwilliams P. Shaving and hair growth. J Invest Dermatol 1970; 55: 170 2. 18. Barth JH, Cherry CA, Wojnarowska F, Dawber RPR. Cyproterone acetate for severe hirsutism: results of a doubleblind dose-ranging study. Clin Endocrinology 1991; 35: 5 10.

384 British Journal of Plastic Surgery The Authors Se Hwang Liew FRCS, Surgical Research Fellow, RAFT Institute of Plastic Surgery, Kaetan Ladhani BSc, Research Assistant, RAFT Institute of Plastic Surgery, Addie O. Grobbelaar FRCS, Consultant Plastic Surgeon, David T. Gault FRCS, Consultant Plastic Surgeon, Roy Sanders BSc, MB, BS, FRCS, Director of Research, RAFT Institute of Plastic Surgery, Colin J. Green PhD, DSc(MED), FRCVS, FRCPath, FRCS(Hon) Acdn of Ukraine Academy of Science, Director of Biomedical Science, RAFT Institute of Plastic Surgery, Claire Linge PhD, Group Leader in Cell Biology, RAFT Institute of Plastic Surgery, Mount Vernon Hospital, Northwood, Middlesex HA6 2RN, UK. Correspondence to Mr Se Hwang Liew. Paper received 2 June 1998. Accepted 13 October 1998.