Abstract
Background and Objective: Acquired dermal melanocytosis (ADM; acquired bilateral nevus of Ota-like macules) is known for its recalcitrance compared to Nevus of Ota, and one of the reasons seems to be a high rate of post-inflammatory hyperpigmentation (PIH) after laser treatments.
Methods: Topical bleaching treatment with 0.1% tretinoin aqueous gel and hydroquinone ointment (4-6 weeks) was initially performed to discharge epidermal melanin. Subsequently, Q-switched ruby (QSR) laser was irradiated to eliminate dermal pigmentation. The both steps were repeated until patients' satisfaction was obtained. This treatment was performed in 19 patients with ADM. Skin biopsy was performed in 6 cases at baseline, after the bleaching pretreatment, and at the end of treatment.
Results: All patients showed good to excellent clearing after 2 to 3 sessions of QSR laser treatments. Total treatment period ranged from 3 to 13 (mean = 8.3) months. PIH was observed in 10.5 % of the cases. Histologically, epidermal hyperpigmentation was observed in all specimens, and was dramatically improved by the topical bleaching pretreatment.
Conclusion: QSR laser combined with the topical bleaching pretreatment appeared to consistently treat ADM with low occurrence rate of PIH and lessen the number of laser sessions and total treatment period, and may also be applied to any other lesions with both epidermal and dermal pigmentation.

Introduction
Nevus of Ota, first described by Ota and Tanino as nevus fuscocaeruleus ophthalmo-maxillaris in 19391, is usually unilaterally located in the area innervated by the ophthalmic and maxillary branch of the trigeminal nerve. A typical nevus of Ota is a flat blue-black or slate-gray macule intermingled with small, flat, and brown spots. Pigmented macules are also often present in ocular, oral, and nasal mucous membrane. On the other hand, acquired bilateral nevus of Ota-like macules (Hori's nevus) is an acquired pigmented lesion involving bilateral grayish-brown facial macules that was first reported by Hori et al. in 19842. This condition and its similar conditions have been referred to as acquired dermal melanocytosis (ADM) by some Japanese dermatologists3,4, and we call this condition "ADM" in this article. ADM onsets later in life most after 20 years of age, represents bilateral involvements, with the malar regions almost always affected; and a lack of mucosal and optic involvement2. It is rare in Caucasians but relatively common in Asian females, and is seen much more frequently than nevus of Ota 5. There is a report that it occupied 7.5 % of cosmetic skin complaints in Japan6. Clinically, it can be easily distinguished from nevus of Ota by bilateral presentation, spot distribution, and difference in color, whereas, in some atypical cases, it can be rarely misdiagnosed as melasma.
There are two significant differences in histology between nevus of Ota and ADM; 1) melanocytes are diffusely distributed throughout the entire dermis in nevus of Ota, while they are located only in the upper dermis in ADM, and 2) epidermal hyperpigmentation is not seen in nevus of Ota, while it is always prominent in ADM; the latter was not well documented before, but confirmed in a series of our samples. Difference in color between nevus of Ota and ADM is due to these histological differences. Although nevus of Ota responds Q-switched lasers very well7-11, ADM is known for its recalcitrance to conventional treatments12-15, and one of the reasons seems to be a high rate of post-inflammatory hyperpigmentation (PIH) seen after laser treatments12-14. The authors assumed that it was mainly due to the presence of epidermal hyperpigmentation and property of epidermal melanocytes in ADM.
The authors previously described an aggressive and optimal use of tretinoin along with hydroquinone for various kinds of skin hyperpigmentation15-17. This topical bleaching treatment was very effective for removal of epidermal pigmentation. Therefore, we tried the bleaching pretreatment before Q-switched ruby (QSR) laser for ADM in order to eliminate epidermal hyperpigmentation and decrease the risk of PIH after QSR laser treatment. This combination therapy was applied for patients with ADM, and its efficacy and usefulness was evaluated.

Patients and Methods
Preparation of Ointments: Tretinoin aqueous gels (tretinoin gel) at 3 different concentrations (0.1, 0.2, and 0.4 %) were originally prepared at the Department of Pharmacy, University of Tokyo, Graduate School of Medicine. The precise regimen of tretinoin aqueous gel was described before16. An ointment including 5% hydroquinone and 7% lactic acid (HQ-LA ointment), and an ointment including 5% hydroquinone and 7% ascorbic acid (HQ-AA ointment) were also prepared. Plastibase (petrolatum polyethylene ointment base, Taisho Pharmacology, Osaka, Japan) was used as the ointment base of the HQ-LA ointment, while the hydrophilic ointment was used for the HQ-AA ointments. Because tretinoin gel, HQ-LA, and HQ-AA ointments (especially tretinoin gel) are pharmacologically unstable, fresh ointments were prepared at least once a month and stored in a dark and cool (4oC) place.
Evaluations of results:
Photographs were taken for every patient at baseline and after treatment with a high-resolution digital camera (Canon EOS-D30). The percentage of pigmentary clearance was evaluated via the photographs by two experienced plastic surgeons who did not perform this treatment. The mean data of the pigmentary clearance of each patient were classified into 4 categories: "excellent" (80% or more clearance), "good" (50% to less than 80% clearance), "fair" (0% to less than 50% clearance), and "poor" (no change or worse).
Patients: Nineteen Japanese patients with ADM were treated with our treatment protocol. All patients were women. Patient age at the start of the treatment ranged from 24 to 50 years old (37.9 ±8.7; mean ± S.D.). The age of onset of ADM was 15-40 years old (24.7 ±7.5; mean S.D.). Pigmented lesions were located bilaterally on the malar area in all 19 cases (100%), on the nose in 4 cases (21%), temporal area in two cases (11%), and forehead in one case (5%). The follow-up time after the final laser session ranged from 3 to 13 months (8.3 ±3.7; mean S.D.).
Treatment Methods: Our treatment was composed of two steps. The first step was a topical bleaching treatment using tretinoin gel and hydroquinone ointment (6-8 weeks) and the second step was QSR laser irradiation. The both steps were repeated until patients' satisfaction was obtained.
1) Topical bleaching treatment: The purpose of this treatment is to improve epidermal pigmentation by accelerating discharge of epidermal melanin (by tretinoin) and suppressing new epidermal melanogenesis (by hydroquinone). The two-phased (bleaching and healing) treatment was performed as following.
a) Bleaching phase: 0.1 % tretinoin gel and HQ-LA ointment were applied initially to the skin lesions twice a day. Tretinoin gel was carefully applied only on pigmented spots using a small cotton-tip applicator, while HQ-LA ointment was widely applied with fingers (e.g. all over the face). The way of ointment application is critical in this aggressive treatment in order to obtain maximal bleaching effects with minimal irritant dermatitis. In case in which severe irritant dermatitis was induced by HQ-LA ointment, HQ-AA ointment was used instead. Patients were requested to visit our hospital at 1, 2, 4, 6 and 8 weeks after starting this treatment, and every 4 weeks afterwards. When the appropriate skin reaction (that is, mild erythema and scaling) was not observed at 1 week, the concentration of tretinoin was changed to 0.4%. In most cases, it took 4 to 6 weeks to finish this phase.
b) Healing phase: After 4-6 weeks' bleaching phase, the application of tretinoin gel and HQ-LA ointment was discontinued, and application of HQ-AA ointment was used in order to prevent post-inflammatory hyperpigmentation (PIH) until the redness was sufficiently reduced. It usually took 2-4 weeks to complete this phase. Topical corticosteroids were not employed in either the bleaching or healing phase.
2) QSR laser treatment: In all patients, topical anesthesia (lidocaine patch; PenlesR, Wyeth Lederle Japan Inc., Tokyo, Japan) was applied 60-120 minutes before the laser treatment. For QSR 694.5 nm laser (Model IB101, Niic Co. LTD, Tokyo, Japan) treatment, spot size of 5 mm, 1 Hz repeat rate, pulse duration of 20 ns, and fluences ranged from 4.0 to 5.0 J/m2 were used. After laser treatment, topical antibiotics ointment was applied twice a day until a scale or crust disappeared (usually for 5-7 days). At 2 weeks after laser treatment, application of HQ-AA ointment was started.
At 4 weeks after each laser treatment, the topical bleaching treatment with 0.1% tretinoin gel and HQ-AA ointment was started as a pretreatment of the next laser irradiation, and also as a treatment of post-laser PIH in some cases. In most cases, the bleaching phase for 2 weeks was long enough, and we can usually estimate the clinical result at 8 weeks after each laser treatment. When some hyperpigmentation remained, we can go for the next session. An example of typical time course was demonstrated in Fig. 1.
Skin biopsy of pigmented regions with a diameter of 2 mm was performed in 6 cases at baseline, just after the 1st bleaching treatment, and at the end of the treatment. Sections were stained with Fontana-Masson staining for visualization of melanin granules.

Results
Clinical results: In the topical bleaching treatment, erythema was usually seen in a few days, followed by continual scaling during the first week. Erythema and scaling were usually seen continually throughout the bleaching phase. On the other hand, erythema gradually declined with time in the healing phase. The difference in color of the macules was usually observed between before and after the first topical bleaching treatment; for example, a color change from brown to gray-brown, or from gray-brown to bluish black, suggesting clearance of epidermal pigmentation.
All patients showed "good" to "excellent" clearing after 2 or 3 laser treatments without any complications such as scarring and persistent depigmentation. Fifteen of 19 cases were evaluated as "excellent" and the other 4 cases as "good" (Table 1). No cases were regarded as "fine" or "poor". QSR laser treatments were performed twice in 7 of 19 cases, and 3 times in the other 12 cases (Table 2). Although PIH apparently occurred in 2 of 19 cases (10.5 %) after the 1st laser treatment, PIH was not clearly seen in any cases after the second and third laser treatments. The average treatment period was 24.8 ± 3.6 (mean ± S.D.) weeks, and the average number of QSR laser treatment was 2.63 ± 0.5 (mean ± S.D.) times. Although patients had unpleasing irritant dermatitis during the topical bleaching treatment, all achieved sufficient satisfaction with the final results and they were followed up for 8.3 ± 3.7 (mean ± S.D.) months (3-13 months) without any evidence of recurrence. The representative 4 cases are shown in Figs. 2-4.
Histological results: At baseline, not only dermal melanocytosis but also epidermal melanosis around basal layers was seen in all 6 samples (Fig. 5). In the upper dermis, elongated, slender, and pigment-bearing melanocytic cells dispersed between the collagen fibers were observed. In addition, all 6 specimens showed disappearance of rete ridges. In most cases, epidermal melanin granules were significantly cleared after the initial bleaching treatment, while dermal pigmentation appeared not to change at all (Fig. 6).

Discussion
The first-reported treatment of ADM was cryotherapy19, but it showed an unpredictable result with a high risk of permanent scarring and hypopigmentation. Kunachak et al.20 treated ADM using dermabrasion with successful results. Despite its advantage as a single session procedure, this approach is perceived by some patients as an invasive procedure21. Therefore, QS lasers are considered to be main treatments of ADM today as well as nevus of Ota.
Although previous reports of QS laser treatments showed good clearance of ADM12-15,21, it has been pointed out that PIH and hypopigmentation are frequently observed after the laser treatments. In ADM, it is known that PIH occurs 2-4 weeks after laser irradiation in higher degrees and frequency than in nevus of Ota. Kunachak et al.21 employed repetitive treatment sessions at only 1-2 weeks intervals. They performed the second laser session before PIH appeared and reported successful clearance of ADM but relatively high (5.7 %) risk of hypopigmentation. Polnikorn et al.12 and Kunachak et al. 14, both used QS Nd:YAG laser to treat ADM, and reported that the rate of PIH was 71 % and 50 %, respectively. Polnikorn et al.12 waited for disappearance of PIH before the next session of laser treatment, and that was 3-6 months. Lam et al.13 used QS alexandrite laser with the mean session number of 7, and most patients showed post-laser PIH. In our own experience using QSR laser for ADM without any pretreatments, PIH was almost always observed 2-4 weeks after the first laser treatment.
The typical color of ADM is grayish-brown2-4. It is more brownish than typical nevus of Ota (blue-black, or slate-gray). The reasons for the difference in color seem to be the existence of epidermal hyperpigmentation and the location of dermal melanocytes. In ADM, dermal melanocytes locates more superficial (the upper dermis2) when compared to nevus of Ota (throughout the entire dermis22). Although few reports have mentioned about the epidermal hyperpigmentation of ADM in the past, we confirmed epidermal hyperpigmentation in all specimens examined in this study (Fig. 5). We consider that the existence of epidermal (basal) melanosis is the reason for the difference in response to laser treatment and incidence rate of PIH between ADM and nevus of Ota. In addition, all biopsy specimens showed disappearance of rete ridges, whereas surrounding intact skin had normal-like rete ridges in some cases. This finding may clinically mean suppression of epidermal turnover and discharge of epidermal melanin, and may be related to the epidermal hyperpigmentation seen in ADM, whereas the reason for epidermal hyperpigmentation in ADM is unknown and epidermal melanocytes in ADM are presumably abnormal like melasma.
In this study, we confirmed histologically that accumulated melanin granules around the basal layer were cleared up after treatment with tretinoin and hydroquinone, but the melanin deposits (dermal melanocytes) in the dermis appeared not to change in ADM (Fig. 6). Taken together with our previous studies18,23-25, this finding supports our previous hypothesis for mechanism of this topical bleaching therapy; Tretinoin acts as a discharger of epidermal melanin by accelerating epidermal turnover and promoting keratinocytes proliferation, while hydroquinone suppresses new melanin production by epidermal melanocytes.
It is considered that the present combination therapy with QSR laser and the aggressive bleaching treatment has the following advantages: 1) High efficiency of the QSR laser treatment in improving dermal pigmentation; After the pretreatment removing epidermal pigmentation (basal melanosis), the laser radiation can be expected to more efficiently get to the dermis because the laser energy is thought not to be much absorbed by epidermal melanins 2) Decreasing the rate of PIH; We assume that, if there is a significant amount of epidermal pigmentation, considerable inflammation would be induced in the entire epidermis, resulting in occurrence of PIH usually 2-4 weeks after laser irradiation. In this sense, therefore, the pretreatment to discharge epidermal melanins seems to be quite important. Indeed, with the bleaching pretreatment, the frequency of PIH after initial laser treatment was as low as 10.6 %. It was significantly lower than other studies. In addition, PIH was not clearly detected after the second or third laser treatment.
Nevus of Ota, which can usually be well treated by several sessions of QSR laser, have predominantly dermal pigmentation. This is because, unlike ADM, it does not have significant epidermal pigmentation, which induces PIH after laser treatments and makes it more difficult to treat consistently. Therefore, we prefer the topical bleaching therapy with tretinoin and hydroquinone for epidermal pigmentation, and QS lasers for dermal pigmentation, with the exceptions of hyperkeratotic lesions with such as solar lentigines on extremities and trunks that we treat with QSR laser. It may be desirable to perform laser treatments after pretreatment of epidermal pigmentation for lesions with both epidermal and dermal pigmentation, such as ADM and hyperpigmentation after atopic dermatitis. The topical bleaching therapy can treat almost any kinds of epidermal hyperpigmentation without hyperkeratosis including PIH and melasma, which can not be treated with lasers.
For treatment of PIH after laser treatments, topical tretinoin and hydroquinone appeared to be best as we and others13 did, though the bleaching protocols are not the same. Otherwise, we can wait for spontaneous clearance of PIH, but the clearance is not guaranteed and intervals between laser sessions become much longer such as 3-6 months12. PIH is one of the easiest pigmented lesions to treat with the topical bleaching treatment17, and, in this study, a mild treatment with tretinoin for only 2 weeks was usually sufficient, while hydroquinone was used continually for over 1 month. Even if the pretreatment is performed, intervals between laser treatments can be shortened up to 6-8 weeks, therefore leading to shortening the total treatment period compared with methods waiting for voluntary disappearance of PIH.

Figures

Fig. 1. A representative time course of the combined treatment. Tretinoin is used for 4 weeks in the initial bleaching pretreatment, and for 2 weeks in the following pretreatments. QSR laser treatment is performed 3 times, and the total treatment period is 30 weeks.

Fig. 2. Case1. A, B) Baseline photos of a 24-year-old woman with ADM. C, D) Just after the bleaching treatment with tretinoin and hydroquinone. The color change of the macules was moderate, but the histological change was apparent as shown in Fig. 6. E, F) Six months after the 3rd QSR laser treatment. Note the complete clearance of pigmentation.

Fig. 3. Case2. A) A baseline view of a 29-year-old woman with ADM. B) Ten months after the 3rd QSR laser treatment. The clinical result was evaluated as "excellent".

Fig. 4. Case3. A) A baseline view of a 46-year-old woman with ADM. B) Three months after the 3rd QSR laser treatment. The result of the clearance was evaluated as "good".

Fig. 5. Histology of ADM before treatments. Both sections demonstrated epidermal hyperpigmentation around the basal layer, melanocytosis in the upper dermis, disappearance of rete ridges, and slight thinning of the epidermis. (Masson-Fontana staining; 100X)

Fig. 6. Histology of ADM in case 1. (A) at baseline, and (B) just after the topical bleaching pretreatment. A) At baseline, the section demonstrated epidermal hyperpigmentation as well as the scattered dermal melanocytes featuring a highly pigmented, elongated dendritic appearance. B) Just after the topical bleaching pretreatment, epidermal pigmentation was significantly improved, while the dermal menlanocytosis appeared not to change at all. (Masson-Fontana staining; 200X)

References

1. Ota M, Tanino H. Naevus fuscocaeruleus ophthalmo-maxillaris and melanosis bulbi. Tokyo Iji Shinshi 1939;63:1243-5.
2. Hori Y, Kawashima M, Oohara K, Kukita A. Acquired, bilateral nevus of Ota-like macules. J Am Acad Dermatol 1984;10:961-4.
3. Kunachak S, Leelaudomlipi P. Q-switched Nd:YAG laser treatment for acquired bilateral nevus of ota-like maculae: a long-term follow-up. Lasers Surg Med 2000;26:376-9.
4. Polnikorn N, Tanrattanakorn S, Goldberg DJ. Treatment of Hori's nevus with the Q-switched Nd:YAG laser. Dermatol Surg 2000;26:477-80.
5. Lam AY, Wong DS, Lam LK, et al. A retrospective study on the efficacy and complications of Q-switched alexandrite laser in the treatment of acquired bilateral nevus of Ota-like macules. Dermatol Surg 2001;27:937-41.
6. Kuroki T, Noda H, Ichinose M, et al. Review of patients with aquired bilateral nevus Ota-like macules. Journal of Japan Society of Aesthetic Plastic Surgery 1999;21:29-37.
7. Goldberg DJ, Nychay SG. Q-switched ruby laser treatment of nevus of Ota. J Dermatol Surg Oncol 1992;18:817-21.
8. Geronemus RG. Q-switched ruby laser therapy of nevus of Ota. Arch Dermatol 1992;128:1618-22.
9. Watanabe S, Takahashi H. Treatment of nevus of Ota with the Q-switched ruby laser. N Engl J Med 1994;331:1745-50.
10. Alster TS, Williams CM. Treatment of nevus of Ota by the Q-switched alexandrite laser. Dermatol Surg 1995;21:592-6.
11. Chan HH, Alam M, Kono T, Dover JS. Clinical application of lasers in Asians. Dermatol Surg 2002;28:556-63.
12. Hidano A. Acquired, bilateral nevus of Ota-like macules. J Am Acad Dermatol 1985;12:368-9.
13. Sun CC, Lu YC, Lee EF, Nakagawa H. Naevus fusco-caeruleus zygomaticus. Br J Dermatol 1987;117:545-53.
14. Mizoguchi M, Murakami F, Ito M, et al. Clinical, pathological, and etiologic aspects of acquired dermal melanocytosis. Pigment Cell Res 1997;10:176-83.
15. Lee GY, Kim HJ, Whang KK. The effect of combination treatment of the recalcitrant pigmentary disorders with pigmented laser and chemical peeling. Dermatol Surg 2002;28:1120-3.
16. Yoshimura K, Harii K, Aoyama T, et al. A new bleaching protocol for hyperpigmented skin lesions with a high concentration of all-trans retinoic acid aqueous gel. Aesthetic Plast Surg 1999;23:285-91.
17. Yoshimura K, Harii K, Aoyama T, Iga T. Experience with a strong bleaching treatment for skin hyperpigmentation in Orientals. Plast Reconstr Surg 2000;105:1097-108.
18. Yoshimura K, Momosawa A, Watanabe A, et al. Cosmetic color improvement of the nipple-areola complex by optimal use of tretinoin and hydroquinone. Dermatol Surg 2002;28:1153-8.
19. Hori Y, Takayama O. Circumscribed dermal melanoses. Classification and histologic features. Dermatol Clin 1988;6:315-26.
20. Kunachak S, Kunachakr S, Sirikulchayanonta V, Leelaudomniti P. Dermabrasion is an effective treatment for acquired bilateral nevus of Ota-like macules. Dermatol Surg 1996;22:559-62.
21. Kunachak S, Leelaudomlipi P, Sirikulchayanonta V. Q-Switched ruby laser therapy of acquired bilateral nevus of Ota-like macules. Dermatol Surg 1999;25:938-41.
22. Hirayama T, Suzuki T. A new classification of Ota's nevus based on histopathological features. Dermatologica 1991;183:169-72.
23. Yoshimura K, Tsukamoto K, Okazaki M, et al. Effects of all-trans retinoic acid on melanogenesis in pigmented skin equivalents and monolayer culture of melanocytes. J Dermatol Sci 2001;27suppl1:68-75.
24. Yoshimura K, Uchida G, Okazaki M, et al. Differential expression of heparin-binding EGF-like growth factor (HB-EGF) mRNA in normal human keratinocytes induced by a variety of natural and synthetic retinoids. Exp Dermatol, in press.
25. Yoshimura K, Momosawa A, Aiba E, et al. Clinical trial of bleaching treatment with 10 % all-trans retinol gel. Dermatol Surg, in press.