Ruby Laser Hair Removal: Evaluation ofLong-Term Efficacy and Side Effects

Valeria B. Campos, MD,1 Christine C. Dierickx, MD,1 William A. Farinelli, BS,1

Tai-Yuan D. Lin, MD,1 Woraphong Manuskiatti, MD,2 and R. Rox Anderson, MD1*

1Wellman Laboratories of Photomedicine, Department of Dermatology, Massachusetts GeneralHospital, Harvard Medical School, Boston, Massachusetts 02114

2Department of Dermatology, Siriraj Hospital, Mahidol University, Bangkok 10700, Thailand

Background and Objective: Although several studies on laser-assistedhair removal have been published, data on long-term follow-up are few.The present study investigated the long-term efficacy and safety of nor-mal-mode ruby laser pulses on hair removal.Study Design/Materials and Methods: The normal-mode ruby laser (Epi-laser; 694 nm, 3 msec) was used to treat a wide range of body sites in 51volunteers. The mean follow-up after the last treatment was 8.37months.Results: Sixty-three percent of the patients had sparse regrowth. Themean fluence used was 46.5 J/cm2 in patients who had sparse hair re-growth and 39.3 J/cm2 in patients who had moderate hair regrowth (P= 0.0127). Transient pigmentary changes occurred most frequently inpatients with skin type 4.Conclusion: The normal-mode ruby laser is an efficient and safe methodfor long-term hair reduction, especially in fair-skinned individualswith dark hair. Higher fluences produce greater long-term efficacy.Adverse effects are minimal and transient. Lasers Surg. Med. 26:177–185, 2000. © 2000 Wiley-Liss, Inc.

Key words: fluence; hirsutism; hypertrichosis; treatment

INTRODUCTION

Unwanted hair is a major cosmetic problem.Methods of hair removal such as shaving, waxing,and tweezing frequently cause ingrown hairs, fol-liculitis, and temporary hair removal. Electroly-sis, although tedious and painful, may be perma-nent in 15–50% of cases [1].

Laser-assisted hair removal has recently re-ceived attention because of its noninvasive na-ture, fast results, and potential for permanence.The Q-switched ND:YAG laser (1,064 nm) can beused in conjunction with a topical carbon–mineraloil suspension, with the carbon acting as a chro-mophore. This system has shown minimal effec-tiveness for long-term hair removal [2,3]. Long-term results have been described with ruby (694nm) [4,5] and Alexandrite (755 nm) [6] lasers, di-ode lasers [7,8] and noncoherent, filtered, pulsedflashlamps [9,10], which use melanin as a chro-mophore for selective photothermolysis. Selective

thermal damage of pigmented target structuresoccurs when sufficient fluence at a wavelength,preferentially absorbed by the target, is deliveredduring a time equal to or less than the thermalrelaxation time of the target [11]. Dierickx et al.[5] described hair removal 2 years after a singletreatment with normal ruby laser pulses (694 nm,0.3 msec, 6-mm beam diameter).

The purpose of this study was to describe ourclinical experience in a consecutive series of 51patients treated with a long pulsed ruby laser forhair removal in different anatomic areas at differ-ent fluences and numbers of treatments.

Contract grant sponsor: Palomar Medical Technologies, Inc.;Contract grant sponsor: Wellman Laboratories.

*Correspondence to: R. Rox Anderson, MD, Wellman Labora-tories of Photomedicine, 50 Blossom Street, Bartlett Exten-sion 6, Boston, MA 02114.E-mail: anderson@helix.mgh.harvard.edu

Accepted 14 July 1999

Lasers in Surgery and Medicine 26:177–185 (2000)

© 2000 Wiley-Liss, Inc.

MATERIALS AND METHODS

Subjects

Fifty-one adults (35 women, 16 men; agerange 4 18–49 years, mean 4 33.7 years) weretreated. Thirty (59%) patients had Fitzpatrick’sskin type 2, 15 (29%) had skin type 3, and six(12%) had skin type 4. All treated areas hadbrown or black terminal hair. Patients with pho-tosensitivity, recent sun exposure, pregnancy,and history of keloid formation or immunosup-pression were excluded. Informed consent was ob-tained from all patients.

Laser Apparatus

A normal-mode, flashlamp-pumped, 694-nmruby laser, emitting a 3-msec pulse duration and7- and 10-mm spot sizes was used, with fluencesof 20–70 J/cm2 when using the 7-mm spot size and10–60 J/cm2 when using the 10-mm spot size(model RD 1200-H, Palomar Medical Technolo-gies, Beverly, MA). The beam was deliveredthrough an articulated arm into an actively cooledhandpiece designed to maximize delivery of lightinto the reticular dermis. The arm consisted of aplanoconvex sapphire lens actively cooled to 10°C,with a 19-mm convex radius and a focal length of25 mm in air, to provide a convergent beam at theskin surface.

Laser Treatment

Patients were instructed to refrain fromtweezing or waxing for 6 weeks before laser treat-ment.

The most common areas treated were theface (25.4%) and back (25.4%). Table 1 shows thedifferent treatment sites.

Before treatment, the area to be treated wasshaved; a topical anesthetic cream (EMLA, AstraUSA, Inc., Westborough, MA) was applied andkept under occlusion for 1 hour before laser treat-ment. The tolerated fluence used at each treat-ment depended on the patient’s skin color. Start-ing fluences when using the 7-mm spot size were40 J/cm2 for skin type 2, 30 J/cm2 for skin type 3,and 20 J/cm2 for skin type 4. If any sign of epi-dermal injury, such as immediate whitening, ve-siculation, or forced epidermal separation (Ni-kosky’s sign), was observed during the procedure,the fluence was decreased by 20–30%. Fluencewas increased or decreased in subsequent treat-ments depending on the patient’s tolerance andpresence of side effects. The average fluence usedwas 44 J/cm2 (range 4 10–70 J/cm2). Retreat-

ment was performed when hair started growingback. The average number of treatments was 2.7(range 4 1–6).

The handpiece was held firmly against theskin surface before, during, and after each laserpulse; no gel was applied. Occasionally an icepack was applied immediately after laser treat-ment. Patients were instructed to apply a topicalantibiotic ointment if scaling or crusting devel-oped and to avoid sun exposure for a few weeksafter treatment. Clinical evaluation was done andphotographs were obtained during each follow-upvisit.

Response Evaluation

The removal of hair was assessed in ablinded fashion by three dermatologists not in-volved in the study by comparing pretreatmentphotographs with photographs taken during thelast follow-up visit. The camera settings, lighting,and position of patients using anatomic land-marks were replicated at each follow-up to opti-mize the quality of the photographs. Efficacy oftreatment was graded according to hair regrowthas none (no hair regrowth), sparse (<25% of hairregrowth), moderate (25–75% of hair regrowth),or full (>75% of hair regrowth).

Side effects such as persistent erythema, hy-perpigmentation, hypopigmentation, texturalchange, and scarring were graded visually by theinvestigators.

Statistical Analysis

The effects of fluence, time of follow-up onhair regrowth, and pigmentary changes in differ-ent areas were analyzed with Fisher’s exact test.Student’s t-test was performed to compare the dif-ferences on mean fluence across groups. P < 0.05was considered statistically significant.

TABLE 1. Distribution of Treatments Sites

Anatomic area Number of patients (%)

Abdomen 1 (1.96)Arm 1 (1.96)Feet 1 (1.96)Shoulder 1 (1.96)Axillae 3 (5.89)Chest 3 (5.89)Thigh 4 (7.84)Bikini 9 (17.64)Back 13 (25.49)Face 15 (29.41)Total 51 (100)

178 Campos et al.

RESULTS

The median follow-up after the last treat-ment was 8 months (mean 4 8.37 months). Therewas no significant difference in hair regrowth inpatients followed for more than 6 months afterthe last follow-up when compared with patientsfollowed for 3–6 months (P 4 0.781). At the finalfollow-up, the majority of the patients (63%) hadsparse regrowth, whereas only 3.9% had full re-growth (Figs. 1, 2A,B).

Fluences of 40 J/cm2 or more were more ef-fective in producing hair loss compared with flu-ences less than 40 J/cm2 (P 4 0.0125; Fig. 3). Themean fluence used in patients who had sparsehair regrowth was 46.5 J/cm2, and the mean flu-ence in patients who had moderate hair regrowthwas 39.3 J/cm2 (P 4 0.0127). Multiple treatmentshad an additive effect (Fig. 4); all patients whoreceived five or six treatments had sparse re-

growth. Table 2 shows hair regrowth at final fol-low-up, mean fluence, and number of treatments.

In this study, the face and axillae had lesshair regrowth than did the back and bikini in thefinal follow-up. However, because of the limitedsample size, this difference was not statisticallysignificant.

Fifteen subjects (29.4%) had transient pig-mentary changes: 3.9% had hypopigmentation,15.7% had hyperpigmentation, and 9.8% hadboth. The incidence of hyperpigmentation andhypopigmentation differed according to the ana-tomic area (Table 3). Mean fluence used in pa-tients with pigmentary changes was not statisti-cally different from the mean fluence used inpatients with no side effects (P 4 0.61). Patientswith skin type 4 had the highest incidence of pig-mentary changes (P 4 0.012; Table 4, Fig. 5A,B).All pigmentary changes were transient, with a

Fig. 1. Hair regrowth rates at the final follow-up visit.

Ruby Laser Hair Removal 179

mean clearance time of 5.6 months (range 4 1–12months). No permanent pigmentary changes, tex-tural change, or scarring occurred in any subject.Table 5 shows hair regrowth at final follow-up,treatment site, mean fluence, and incidence ofside effects for each treatment site.

DISCUSSION

Long-pulsed ruby laser is an effective andsafe method of hair removal. The 694-nm wave-length deeply penetrates into the dermis, wherefollicular melanin is by far the dominant chromo-phore [12]. Absorption of laser light by melanin inthe hair shaft, epithelium, and matrix cells causesthermal damage of the hair follicles.

Two important target sites are thought to bethe germinative cells of the hair, matrix, and pa-

pilla or bulb and/or the follicular stem cells lo-cated near insertion of the arrector pili muscle orbulge [13,14].

Thermal conduction during the laser pulseheats a region around each microscopic site of op-tical energy absorption [11]. Therefore, the spa-tial scale of thermal confinement is related to la-ser pulse duration. For best epidermal cooling,pulse duration would occur between the relax-ation time for the epidermis (3–10 msec) and thatfor most hair follicles (∼10–100 msec for hair fol-licles 100–300 mm in diameter) [4,15,16]. In thisstudy, the pulse duration was 3 msec, which al-lowed thermal confinement even in fine-hair fol-licles, and is at or near the thermal relaxationtime of epidermis, allowing sparing of the epider-mis during conductive contact cooling.

In a small study of ruby laser hair removal,

Fig. 2. A: An axilla before treatment. B: The same axilla 20 months after the second treatment.

180 Campos et al.

Grossman et al. [4] showed that a single treat-ment produces a hair-growth delay, followed bypartial regrowth at 6 months’ follow-up. Dierickxet al. [5] later reported permanent hair reduction2 years after a single treatment with high-fluenceruby laser pulses (at least 30 J/cm2). The samestudy reported that “steady-state” loss wasachieved at 6 months’ follow-up. The mean follow-up time in our study was 8 months after the lasttreatment. Whether this follow-up time is longenough to claim permanency for different bodysites remains to be confirmed.

Treatment fluences, pulse duration, spotsize, and numbers of treatment are important fac-tors for achieving long-term hair reduction. Inthis study, the majority of patients (63%) hadsparse hair regrowth at a mean follow-up of 8months after the last treatment, at a mean flu-ence of 46.5 J/cm2 and a mean of 2.7 treatments.This has also been shown in other studies usingdifferent ruby lasers [17,18].

Complete hair removal (100% of all hair re-moved in the treated area) was not achieved inany patient. However, the regrowing hair after

ruby laser treatment was thinner and lighter andtherefore less visible, which contributed to theoverall and cosmetic improvement. Dierickx et al.[5] suggested that miniaturization of coarse ter-minal hair follicles to vellus like hair follicles isthe main mechanism by which ruby laser pulsescause permanent hair reduction, which is consis-tent with our observations.

Walther et al. [19] reported that hair re-moval with ruby laser could not be achieved with20 J/cm2, a spot size of 2 mm, and a pulse durationof 0.5 msec. Dierickx et al. [5] obtained long-termhair removal with a 0.3-msec pulse duration, amuch larger spot size (6 mm), and fluences up to60 J/cm2.

Almost one-third of our patients presentedwith temporary pigmentary changes, occurringmainly in the darker skin types. It is also impor-tant to note that pigmentary side effects occurredmainly in the bikini area and back (55.6% and46.2% of the patients treated in these areas, re-spectively). Only 6.7% of the patients treated onthe face had pigmentary changes.

In practice, the ideal patient for laser hair

Fig. 3. Influence of the fluence on efficacy.

Ruby Laser Hair Removal 181

removal has dark hair and fair, untanned skin.Such patients are usually best treated with highfluences (>39 J/cm2) at the 7- or 10-mm spot sizeand almost always achieve reduction of hair to asparse amount in 3–6 treatments, with minimalor no side effects. Darker skin types with dark

TABLE 4. Incidence of Pigmentary ChangesAccording to Skin Type

Skin typeIncidence of

pigmentary change (%)

2 6/30 (20)3 4/15 (26.7)4 5/6 (83.4)*Total 15/51 (29.4)

*Incidence was statistically significantly different from types2 and 3, P 4 0.012.

Fig. 4. Influence of the number of treatments on efficacy.

TABLE 2. Distribution of Subjects According to HairRegrowth at the Final Follow-up, Mean Fluence, andNumber of Treatments

Meanfluence

Number oftreatments

Hair regrowth at final follow-up

Sparse(%)

Moderate(%)

Full(%)

<40 J/cm2 1 0 0 02 0 4 (7.8) 03 3 (5.9) 3 (5.9) 04 0 1 (2) 0

5–6 0 0 0Subtotal 3 (5.9) 8 (15.7) 0 (0)$40 J/cm2 1 2 (4) 2 (4) 1 (2)

2 10 (19.6) 3 (5.9) 03 12 (23.5) 4 (7.8) 1 (2)4 1 (2) 0 0

5–6 4 (7.8) 0 0Subtotal 29 (56.9) 9 (17.6) 2 (3.9)Total 31 (62.8) 17 (33.3) 2 (3.9)

TABLE 3. Incidence of Pigmentary Changes forDifferent Areas

Anatomical areaIncidence of

pigmentary changes (%)

Abdomen 1/1 (100)Axillae 1/3 (33.3)Back 6/13 (46.2)*Bikini 5/9 (55.6)*Face 1/15 (6.7)*Thigh 1/4 (25)Total 15/51 (29.4)

*The face had significantly less pigmentary change than didthe bikini and back (P < 0.05).

182 Campos et al.

Fig. 5. A: Hypopigmentation and hyperpigmentation in a patient with skin type 4 1 month after treatment. Spontaneousrepigmentation and clearance occurred after 6 months. B: Hyperpigmentation in a patient with skin type 4 1 month aftertreatment. Spontaneous clearance occurred after 5 months.

Ruby Laser Hair Removal 183

hair also respond well, but with a higher inci-dence of side effects. To limit epidermal damagein dark skin types, the treatment fluence below 40J/cm2 was often necessary, which also reduces ef-ficacy and requires a higher number of treat-ments. We strongly recommend avoidance of sunexposure before and after the ruby laser hair re-moval treatment. Patients with light-colored hairand dark skin are generally poor candidates forlaser hair removal and should consider other con-ventional methods such as electrolysis.

ACKNOWLEDGMENTS

Woraphong Manuskiatti, MD, was a re-search fellow at Wellman Laboratories during thestudy. We thank Maria Beatrice Alora, MarcyMoore, and Kim Palli for the help with dataanalysis and Jose Luis Ferreira Duque for the sta-tistical analysis.

REFERENCES

1. Wagner RF. Physical methods for the management of hir-sutism. Cutis 1990;45:319–326.

2. Goldberg DJ. Various mechanisms of hair removal. Cos-met Dermatol 1997;10:36–38.

3. Nanni CA, Alster TS. Optimizing treatment parametersfor hair removal using a topical carbon-based solution

and 1064 nm Q-switched neodynium:YAG laser energy.Arch Dermatol 1997;133:1546–1549.

4. Grossman MC, Dierickx CC, Farinelli W, Flotte T, Ander-son RR. Damage to hair follicles by normal-mode rubylaser pulses. J Am Acad Dermatol 1996;35:889–894.

5. Dierickx CC, Grossman MC, Farinelli WA, Anderson RR.Permanent hair removal by normal-mode ruby laser.Arch Dermatol 1998;134:837–842.

6. Finkel B, Eliezri YD, Waldman A, Slatkine M. PulsedAlexandrite laser technology for non-invasive hair re-moval. J Clin Laser Med Surg 1997;15:225–229.

7. Dierickx CC, Grossman MC, Farinelli WA, Anderson RR.Hair removal by a pulsed, infrared laser system. LaserSurg Med 1998;22(Suppl 10):198.

8. Grossman MC, Dierickx CC, Quintana A, Geronemus R,Anderson RR. Removal of excess body hair with an 800nm pulsed diode laser. Laser Surg Med 1998;22(Suppl10):201.

9. Smith SR, Goldman MP, Fitzpatrick RE. Long-term re-sults of hair photo-epilation. Laser Surg Med 1998;22(Suppl 10):205.

10. Weiss RA, Weiss MA, Marwaha S, Harrington AC. Hairremoval with a non-coherent filtered flashlamp pulsedlight source. Laser Surg Med 1998;22(Suppl 10):190.

11. Anderson RR, Parrish JA. Selective photothermolysis:precise microsurgery by selective absorption of pulsed ra-diation. Science 1983;220:524–527.

12. Anderson RR, Parrish JA. The optics of human skin. JInvest Dermatol 1981;77:13–19.

13. Sun T, Cotsarelis G, Lavker RM. Hair follicular stemcells: the bulge-activation hypothesis. J Invest Dermatol1991;96(Suppl 5):77S–78S.

14. Lyle S, Christofidou-Solomidou M, Liu Y, Elder D, Al-belda S, Cotsarelis G. The C8/144B monoclonal antibodyrecognizes cytokeratin 15 and defines the location of hu-

TABLE 5. Distribution of Subjects According to Hair Regrowth, Treatment Sites, Average Fluence, and SideEffects (Pigmentary Changes)

Meanfluence

Anatomicarea

Sparse (%) Moderate (%) Full (%)

No sideeffects

Side effectspresent

No sideeffects

Side effectspresent

No sideeffects

Side effectspresent

<40 J/cm2 Axillae 0 0 0 0 0 0Back 0 2 (4) 2 (4) 1 (2) 0 0Bikini 0 0 1 (2) 0 0 0Chest 0 0 1 (2) 0 0 0Face 1 (2) 0 2 (4) 0 0 0Thigh 0 0 0 0 0 0Othera 0 0 0 0 0 0

Subtotal 1 (2) 2 (4) 6 (11.8) 1 (2) 0 (0) 0 (0)$40 J/cm2 Axillae 2 (4) 1 (2) 0 0 0 0

Back 3 (5.9) 1 (2) 2 (4) 1 (2) 0 1 (2)Bikini 3 (5.9) 4 (7.8) 0 0 0 1 (2)Chest 1 (2) 0 1 (2) 0 0 0Face 8 (15.7) 0 1 (2) 1 (2) 0 0Thigh 3 (5.9) 1 (2) 2 (4) 0 0 0Othera 0 1 (2) 3 (5.9) 0 0 0

Subtotal 20 (39.2) 8 (15.7) 9 (17.6) 2 (4) 0 (0) 2 (4)Total all fluences 21 (41.1) 10 (19.6) 15 (29.4) 3 (5.9) 0 (0) 2 (4)aOther includes abdomen, arm, feet, and shoulders (in four subjects).

184 Campos et al.

man hair follicles stem cells. J Cell Sci 1998;111:3179–3188.

15. Van Gemert MJC, Welch AJ. Time constants in thermallaser medicine. Lasers Surg Med 1989;9:405–421.

16. Anderson RR. Laser–tissue interactions. In: GoldmanMP, Fitzpatrick RE, editors. Cutaneous laser surgery:the art and science of selective photothermolysis. St.Louis: Mosby-Year Book; 1994. p 1–18.

17. Williams R, Havoonjian H, Isagholian K, Menaker G,

Moy R. A clinical study of hair removal using the longpulsed ruby laser. Dermatol Surg 1998;24:837–842.

18. Sommer S, Render C, Burd R, Sheehan-Dare R. Rubylaser treatment for hirsutism: clinical response and pa-tient tolerance. Br J Dermatol 1998;138:1009–1014.

19. Walther T, Baumler W, Wenig M, Landthaler M, Hohen-leutner U. Selective Photothermolysis of hair follicles bynormal-mode ruby laser treatment. Acta Dermatol Vene-reol 1998;78:443–444.

Ruby Laser Hair Removal 185