Jean-Paul Ortonne, MD, Department of Dermatology, Archet-2 Hospital, 151 Route St Antoine de Ginestiere, BP 3079, 06202 Nice Cedex 3, France, Tel.: +33 4 9203 6488, Fax: +33 4 9203 6560, e-mail: email@example.com
Please cite this paper as: Reflectance confocal microscopy for pigmentary disorders. Experimental Dermatology 2010; 19: 233–239.
Abstract: In vivo reflectance confocal microscopy (RCM) is a non-invasive, repetitive imaging tool that provides real-time images at nearly cellular histological resolution. Application of this technology to skin imaging during the last decade has been a great advance in dermatology. As melanin is the strongest endogenous contrast in human skin, pigmentary disorders caused by abnormal amounts of melanin in the skin could be the most suitable candidates for RCM examination. This article reviewed the RCM applications in the characterization and management of pigmentary disorders. The application of RCM in pigmentary disorders has been expanded to describe hyper- and hypopigmentary disorders as well as pigmented skin tumors. The great advantages of non-invasive and repetitive examination of RCM may provide its usefulness not only in the diagnosis and management of pigmentary disorders, but also in researching pathogenesis of pigmentary disorders.
In vivo reflectance confocal microscopy (RCM) is a non-invasive, repetitive imaging tool that provides real-time images at nearly cellular histological resolution. Application of RCM to skin imaging during the last decade has been a great advance in dermatology. RCM has been used mainly for the differentiation between benign and malignant skin lesions showing great potential for melanoma and melanocytic lesion assessment (1–4). RCM also showed potential for diagnosis in inflammatory skin conditions including psoriasis, contact dermatitis and bacterial and fungal infections (5). Furthermore, as RCM provides the opportunity to determine in vivo kinetics of the skin after the application of therapeutics such as anti-ageing agent, RCM may be a useful tool in developing cosmetics (5).
As melanin is the strongest endogenous contrast in human skin, pigmentary disorders caused by abnormal amounts of melanin in the skin could be the most suitable candidates for RCM examination (6). This article reviews the RCM applications in the characterization and management of hyper- and hypopigmentary disorders as well as pigmented skin tumors.
Basic principle of RCM
The basic concept of confocal microscopy is the selective collection of light from the focussed spot of the skin. A confocal microscope consists of a light source, condenser and objective lenses, and a detector (Fig. 1). The light source, near-infrared wavelength laser beam, illuminates a specific point within a skin and reflected light is collected through a pinhole of the detector. The present commercially available confocal laser microscope (Vivascope 1500; Lucid Inc, Rochester, NY, USA) that we use, is equipped with an 830 nm diode laser with a low power of 5–10 mW, which causes no tissue damage. The light source, the illuminated spot, and the pinhole are in optically conjugated focal planes, leading to the name ‘confocal’. The illuminated spot is then scanned horizontally, producing black and white images from epidermis to the upper reticular dermis with an imaging depth of up to 250–300 μm. To visualize a larger area, a two-dimensional sequence of images are captured and stitched in software to create a mosaic that displays up to 8 × 8 mm of tissue. Highly reflective skin components including melanin, collagen and keratin appear bright (white) in RCM images.
Visualization of melanin and melanocytes by RCM
When examining the skin, the corneal layer is the first image seen as a highly refractile layer because of back scattering of light at the immersion medium–tissue junction (Fig. 2). The granular layer appears next consisting of cells with bright grainy cytoplasm surrounding an oval dark nucleus. The spinous layer presents as a honeycombed pattern of smaller cells. The basal cell layer, is seen as a cobblestone pattern created by bright clusters of cells at the dermo–epidermal junction (DEJ) of 40–130 μm depth. The brightness is stronger in the basal layer than in the spinous layer, because of the abundant supranuclear melanin caps in the basal cells. When imaging a little deeper, the suprapapillary epidermis at the DEJ appears as bright rings of basal cells surrounding a dark dermal papillae, which show a papillary dermal vasculature (Fig. 2d). Below DEJ, a network of collagen fibres and bundles can be observed in the papillary dermis and superficial reticular dermis.
Different brightnesses correspond to endogenous variation among skin phototypes and the anatomic location (7). For examples, the brightness is stronger on the dorsal than on the volar forearm and particularly pronounced in heavily pigmented individuals. In skin phototype I, the basal keratinocytes have low refractility and dermal papillary rings are difficult to elucidate (7).
Typical dendritic melanocytes were hardly observed in the normal human skin, although supranuclear melanin caps were easily visible (8). It was suggested that this might be because melanocytes rapidly transfer the produced melanin to keratinocytes and do not accumulate it (8). On the contrary, the melanocytes are observed in the pigmented animal skin such as Weiser–Maples guinea pig and during the pigmentation process of human skin (e.g. UV exposure; Fig. 3) (8,9). As activated Langerhans cells appear as highly refractile dendritic cells in suprabasal epidermal layer, it should be differentiated from the dendritic melanocytes (10).
Melanophages can be distinguished in the superficial dermis of pigmented skin as large, fuzzy, bright cells with ill-defined cytoplasmic borders (Fig. 7c) (11).
Reflectance confocal microscopy has been used to show dynamic pigmentary changes in human or animal skin over time in response to UV irradiation (8,12–15). In the human skin after single UVA exposure, the brightness in the basal layer temporarily decreased and gradually increased from day 4. Activated melanocytes were clearly visualized 8 days after UV exposure (Fig. 3). The melanocytes then gradually disappeared and could no longer be observed in the RCM images. Instead, the supranucelar melanin caps became clear from day 29. Melanin appeared to be distributed more scattered than before UVA exposure. However, changes in the melanin immediately after the exposure, i.e. in immediate pigment darkening were hardly detectable, and RCM images only showed accelerated capillary flow patterns (15). There was also report showing a reduction of dermal reflectivity (dermal oedema) and an increase in vessel diameter as well as increased brightness of the basal layer 72 h after UVB irradiation (12).
RCM for pigmented skin tumors
As RCM has been used in dermatology, RCM characterization of melanocytic skin tumors is a major area with the potential to improve the diagnostic accuracy of melanoma and equivocal melanocytic lesions (16–26). In addition, RCM allowed the study of non-melanocytic pigmented tumors and in these tumors, pigmentation comes from pigmented keratinocytes (11) or dendritic melanocytes which can be easily identified within tumor islands (29–32) (Table 1).
Table 1. Reflectance confocal microscopy features of pigmentary disorders
Melanocytic pigmented skin tumors
Small, round to oval, bright refractile cells which are arranged as nests (cell clusters) along the rete-ridge (junctional nevi) and/or dermis (compound/dermal nevi).
Disruption or loss of the normal honeycomb or cobblestone pattern (epidermal disarray). Dermal papilla lacking a sharply demarcated bright rim (non-edged papillae), and sometimes even sheet-like structures. Scattered atypical, large round melanocytes, sometimes presenting dendritic branches above basal layers (pagetoid cells). Cytological atypia in basal layers.
Non-melanocytic pigmented skin tumors
Cerebriform architecture of epidermis. Well-defined, round, black areas filled with whorled refractile material (horn cysts). Enlarged papillary rings lined by brightly refractile cells (pigmented keratinocytes). Plump bright cells (melanophages) may be present in the dermis.
Pigmented basal cell carcinoma
Well-circumscribed tumor nests consisting of packed elongated cells with palisading, forming cord-like structures or nodules at the basal cell layer and into the papillary dermis. Bright dendritic cells both within the epidermis overlying the tumor and within tumor nests.
Pigmented eccrine poroma
Dendritic melanocytes intermingled with small uniform cells of tumor nest.
No bright cells in the epidermis and complete loss of dermal papillary rings at dermo–epidermal junction. Dendritic melanocytes at the basal layer in repigmented skin.
Hyperrefractile dermal papillary rings with ovoid to annular or polycyclic contours at dermo–epidermal junction. Plump bright cells (melanophages) may be present in the dermis.
Increased cobblestoning at basal cell layer. Dendritic melanocytes may be present at basal cell layer. Plump bright cells may be present in the dermis.
Normal honeycomb pattern and normal edged papilla at the dermo–epidermal junction. Numerous dendritic cells (Langerhans cells) located above the basal cell layer.
Common blue nevus
Small bright dendritic or elongated cells (melanocytes) could be identified between collagen bundles.
Melanocytic nevi and melanoma
Reflectance confocal microscopy features of melanocytic nevi and melanoma have been described in many papers (16–19). Nevomelanocytes are characterized by small, round to oval, bright refractile cells which are arranged as nests (cell clusters) along the rete-ridge (junctional nevi) and/or dermis (compound/dermal nevi). A regular epidermal architecture, showing a honeycombed or cobblestone pattern is shown in melanocytic nevi. The valuable RCM features to suspect melanoma include (i) disruption or loss of the normal honeycomb or cobblestone pattern (epidermal disarray), (ii) dermal papilla lacking a sharply demarcated bright rim (non-edged papillae), and sometimes even sheet-like structures, (iii) scattered atypical large round melanocytes, sometimes presenting dendritic branches above basal layers (pagetoid cells) and (iv) cytological atypia in basal layers. Invasive melanoma also contains melanocytes in the dermis, as single nucleated cells or more often cell clusters with loose, inhomogeneous or cerebriform morphology. Whereas regular rete-ridge, edged papillae and regular dense nests favour mostly benign lesions. RCM features of melanoma have shown good correlation to histological examination (21). Several studies showed diagnostic significance of RCM features in melanocytic tumors with sensitivity and specificity of approximately 90% and 86% (22–26). This is compared very favourably with dermoscopy with 83.2% sensitivity and 85.8% specificity and spectrophotometric intracutaneous analysis (SIA) with 82.7% sensitivity and 80.1% specificity in melanoma detection (29,30). In addition, RCM has potential diagnostic advantage of cellular level evaluation compared with dermoscopy or SIAscope. Moreover, diagnostic criteria was created and successfully applied for diagnostic differentiation (22–24).However, confocal diagnosis of malignant melanoma is occasionally challenging, as the findings for melanoma are also observed occasionally in melanocytic nevi. For example, pagetoid infiltration constituted by rounded cells was reported in melanocytic nevi (17). Sometimes, non-edged papillae are also observable in dysplastic nevi (18). Atypical cerebriform nests are rarely observed in benign lesions (19).
Non-melanocytic pigmented skin tumors
Reflectance confocal microscopy examination of seborrheic keratosis showed several features which correlated with histological findings (29). RCM mosaic of the epidermis reveals a well-demarcated lesion with striking cerebriform architecture. Horn cysts seen on histology are easily identified using RCM as well-defined, round, black areas filled with whorled refractile material. Busam et al. (11) presented RCM findings of five pigmented seborrheic keratosis by revealing enlarged papillary rings lined by brightly refractile cells, corresponding to pigmented keratinocytes. Plump bright cells which are the melanophages inside dermal papillae are frequently observed.
Reflectance confocal microscopy of pigmented basal cell carcinoma demonstrates well-circumscribed tumor nests consisting of packed elongated cells with palisading, forming cord-like structures or nodules at the DEJ and into the papillary dermis. Bright dendritic cells were consistently observed both within the epidermis overlying the tumor and within tumor nests (Fig. 4). Immunohistochemical studies revealed that the cells within tumor nest correlated with melanocytes, whereas dendritic cells in the epidermis corresponded to Langerhans cells (30–32). In pigmented eccrine poroma, the dendritic melanocytes intermingled with small uniform cells of tumor nest were also observed (33). The importance had to be stressed of being aware of the presence of dendritic cells in RCM of pigmented non-melanocytic skin tumors to avoid the incorrect classification of a melanocytic tumor including melanoma.
Reflectance confocal microscopy examination of pigmented mammary Paget’s disease revealed the limitation of making a correct diagnosis with RCM (34). The RCM findings suggested the diagnosis of superficial spreading malignant melanoma as the presence of large atypical cells resembling a pagetoid melanocytosis usually seen in malignant melanocytic lesions. Immunohistochemical studies were required to diagnose Paget’s disease.
RCM for hyper- or hypopigmentary disorders
Currently, RCM has a wide range of applications to describe benign pigmentary skin conditions. Hypopigmented diseases like vitiligo as well as hyperpigmentary disorders such as lentigines and melasma have been evaluated to define their characteristic confocal features.
Reflectance confocal microscopy may allow the identification of characteristic features of vitiligo and repigmented skin during treatment. In the vitiligo lesions, normal dermal papillary rings disappeared and no bright keratinocytes could be detected in the epidermis (Fig. 5). RCM images of repigmented skin after excimer laser treatment showed dendritic melanocytes at the basal layer. Ardigo et al. (35) investigated the RCM findings in 16 diffuse or localized vitiligo lesions and compared with 10 healthy controls. The authors described some changes in non-lesional skin of the patients. The characteristic ring structures of healthy controls were hardly recognizable and half-rings or scalloped border-like features of the rings were observed. The authors suggested that this change could derive from an incomplete distribution of pigment along the basal layer because of an initial and progressive disappearing of melanocytes or to a congenital defective melanocyte distribution. In repigmented areas 2 months after UVB narrow band therapy, the presence of activated, dendritic melanocytes in the basal layer was seen in 6 of 16 patients. These findings suggested that RCM could be used in the therapeutic monitoring and evaluation of the evolution of vitiligo.
Lentigines are broadly divided into two subtypes; solar lentigo and lentigo simplex. All lentigo subtypes share basic histologic findings of proliferation of basal melanocytes usually accompanied by elongated and pigmented rete-ridges. In the case of solar lentigo, the rete-ridges range from short and club shaped to more complex anastomosing projection.
Reflectance confocal microscopy examination of lentigo showed several features which correlate well with these histologic findings: hyperrefractile cobblestone pattern of basal layer and hyperrefractile dermal papillary ring (36,37). The most outstanding RCM feature in all lentigines was observed at the DEJ with the distinct patterns created by alteration in the dermal papillae and rete-ridge morphology (Fig. 6). The bright refractile cells around the dermal papillae correspond to rete-ridges in cross-section. Dermal papillae were increased in number, enlarged, and had ovoid, polycyclic geometric shapes. Solar lentigo may display cerebriform appearance because of complex anastomosing of rete-ridges, which overlaps with seborrheic keratosis. The presence of increased numbers of melanocytes in the basal layer may not be appreciated on RCM and is difficult to distinguish from hyperpigmented basal keratinocytes on RCM. In four cases of six lentigo, melanophages were seen within dermal papillae (36).
On the face, lentigines may be confused occasionally with lentigo maligna and lentigo maligna melanoma resulting in biopsy. RCM provides a potential advantage to differentiate them as the whole lesion can be examined in vivo. If features of melanoma, such as bright atypical, polymorphous dendritic cells throughout the epidermis with tendency of atypical cells to group around the hair follicles, are identified, a diagnosis of melanoma should be considered (36).
Melasma is characterized by epidermal hyperpigmentation, possibly because of an increased melanin and to an increased number or activity of melanocytes (38). We investigated the relevant in vivo RCM features of melasma in our preliminary study. Twenty-five patients with melasma who gave informed consent, were enrolled in the study with approval of the ethic committee. Multiple mosaic and stack RCM images were acquired from the lesional skin and perilesional normal skin. Skin biopsy was obtained from seven patients. RCM images of melasma showed characteristic significantly increased hyperrefractile cobblestone pattern at the level of basal cell layer in lesional skin compared with perilesional normal skin (Kang, H. Y. et al., unpublished observations, Fig. 7). Dendritic cells and plump bright dermal cells corresponding to activated melanocytes and melanophages, respectively, were also observed in some patients. Most melasma skin showed an abrupt transition from stratum spinosum to papillary dermis and moderately refractile lacy structures (solar elastosis) in the dermis, suggesting the existence of chronic solar damage in the melasma. These findings suggested the possibility of future clinical applications of RCM for melasma diagnosis and treatment.
Areolar melanosis is a benign process characterized by basal hyperpigmentation in the presence of normal to slightly increased numbers of melanocytes (39). It may be confused with malignant melanoma clinically. RCM examination of areolar melanosis showed a normal honeycomb pattern and normal edged papilla at the DEJ (40). The vessels and pink areas seen under dermoscopy revealed numerous dendritic cells, some of which appeared to be located above the DEJ in RCM examination. CD1a immunohistochemical staining demonstrated that the intraepidermal dendritic cells identified on RCM were Langerhans cells, not pagetoid melanocytes. The authors suggested that histopathologic evaluation is still the gold standard and should be recommended in clinically equivocal cases.
As RCM has limitation of penetration depth, it seems that RCM does not allow us to explore the lesion characterized by dermal melanocytic proliferation. There is strong contrast attenuation because of absorption and scattering of light going through hyperpigmented or hyperkeratotic lesions (5). The presence of refractive structures in the dermis such as inflammatory cells and collagen bundles may also decrease contrast and difficult melanocyte visualization. It was shown that in common blue nevus, small bright dendritic or elongated cells corresponding to pigmented dendritic melanocyte could be identified between collagen bundles (41). However, the authors suggested as in deeper lesions RCM features may be completely lacking, confocal evaluation of a suspected blue nevus always be considered non-diagnostic.
RCM evaluation of therapeutics for pigmentary disorders
Reflectance confocal microscopy may be a tool in evaluating efficacy of therapeutics for pigmentary disorders. Yamashita et al. (37) showed the dynamic mechanism of pigment removal by intense pulsed light (IPL) therapy of solar lentigines using RCM. RCM images showed that the highly reflective plate (melanosomes) in the epidermal basal layer rapidly migrated to the skin surface on day 5 of treatment. IPL irradiated melanocytes in the lesions seemed to be intact and resumed their high activity after treatment, as numerous active melanocytes were apparent on day 9. In animal models, RCM has been used to analyse depigmenting agents (42). Overall brightness of keratinocytes in skin treated with depigmenting agent was clearly less than in vehicle-treated skin of irradiated guinea pig. These reports suggested the feasibility of non-invasive time course study with RCM in animal or human subjects to evaluate the cutaneous response to therapeutics for pigmentary disorders.
In vivo RCM is a relatively young field in dermatology. Nevertherless, application of RCM in pigmentary disorders has been expanded to describe hyper- and hypopigmentary disorders as well as pigmented skin tumors. The great advantages of non-invasive and repetitive examination of RCM provide its usefulness not only in the diagnosis and management of pigmentary disorders, but also in researching pigmentary disorders. It may be employed as a diagnostic tool to evaluate benign pigmentary skin lesions as well as pigmented skin tumors, especially on the face without performing skin biopsy. It would be further needed to have validation study and formal consensus process for the features of the different disease. RCM examination can be repeated and therefore can be used to monitor treatment of lesions through time. For example, appearance of melanocytes and resulting progressive repigmentation of vitiligo lesions during treatment could be observed under RCM examination. Finally, RCM may be a valuable tool in research for pathogenesis as well as for development of therapeutics of pigmentary disorders.
The present major limitation of RCM is the depth of imaging, and therefore there is limitation to evaluate deeper pigmentary lesions. Interestingly, in the case of lentigo maligna and lentigo maligna, melanoma depth is not a limitation, as most diagnostic features of these lesions are intraepidermal and therefore easily detectable by RCM (43). In addition, it should be noted that the horizontal images of RCM give different information from the vertical sections of histology. However, using image analysis software, it should soon be possible to obtain virtual vertical sections from horizontal RCM images. Finally, RCM imaging can still be time consuming especially for the systematic exploration of a lesion area at different depths, but future systems will probably be able to work faster and acquire larger fields.
Further advances in confocal instrumentation may offer enhanced diagnostic potential and application in pigmentary disorders. Several other models of contrast such as multiphoton laser tomography are being developed and studied for application for pigmentary disorders (44–46). Such modes may be synergistically combined for skin imaging to provide structure-specific contrast.
The authors have no conflicts of interest that are directly relevant to the content of this review. This work was supported by ‘GRRC’ Project of Gyeonggi Provincial Government, Republic of Korea and the Korean Science and Engineering Foundation (KOSEF) Grant funded by the Korean government (MOST) (R13-2003-019) to HY Kang. We thank Dr. S. Gonzalez for his kind supply of Figure 3.