CHAPTER 15 Dermoscopy

Ashfaq A. Marghoob, Azadeh Esmaeili, Alon Scope

Skin cancer, the most common malignancy in the United States, is associated with significant morbidity and mortality. Fortunately, early detection of skin cancer, through visual examination of the entire cutaneous surface, can have a positive impact on patient outcomes. Because dermoscopy can assist clinicians in correctly identifying cutaneous malignancies, it has been well received by physicians engaged in skin cancer screening efforts.

Dermoscopy is a technique that requires the use of hand-held magnification devices known as dermoscopes (or dermatoscopes). These instruments illuminate the skin and, by exploiting the optical properties of the skin, allow the physician to visualize subsurface colors and structures. Dermoscopes are designed to reduce the amount of light reflected off the skin surface, thereby allowing clinicians to appreciate the appearance of the subsurface anatomic structures of the epidermis and papillary dermis that are otherwise not discernible to the unaided eye. Two types of dermoscopes are available, one using standard light-emitting diode (LED) illumination (i.e., nonpolarized dermoscopy) and the other using cross-polarized light (polarized dermoscopy; Fig. 15-1). To reduce skin surface light reflection, nonpolarized dermoscopy requires direct skin contact in the presence of a liquid interface (e.g., mineral oil or alcohol) between the dermoscope and the skin. Polarized dermoscopy, however, does not require an immersion liquid because one of the inherent properties of cross-polarized light is to filter out light reflected from the skin surface, allowing only light reflected from deeper layers of the skin to reach the observer’s retina. Polarized and nonpolarized dermoscopy provide complementary information. For example, polarized dermoscopy is the preferred method for visualizing blood vessels because it does not require direct skin contact. The ability to see dermoscopic structures without direct skin contact eliminates the effect of contact pressure–induced blanching of the blood vessel. Nonpolarized dermoscopy, on the other hand, is better for visualizing structures within the superficial layers of the epidermis, such as milia cysts, which are important structures that enable observers to correctly identify seborreic keratoses. With these scopes, clinicians can now appreciate morphologic alterations in skin lesions as dermoscopic structures of different shapes and colors (Table 15-1). Because most dermoscopic colors and structures have been correlated with histopathologic findings, dermoscopy can be considered a form of bedside in vivo gross tissue inspection that can help to predict tissue pathology.

Figure 15-1 Physician examining a pigmented skin lesion under contact nonpolarized dermoscopy (A) and noncontact polarized dermoscopy (B).

TABLE 15-1 Dermoscopic Structures and Their Histopathologic Correlations

Dermoscopic Structures	Definition	Histopathologic Correlation
Pigment network (reticulation)	Gridlike network consisting of pigmented “lines” and hypopigmented “holes.”	Melanin in keratinocytes or melanocytes along the epidermal rete ridges.
Pseudonetwork	In facial lesions, diffuse pigmentation interrupted by nonpigmented follicular openings, appearing similar to a network.	Pigment in the epidermis or dermis interrupted by follicular and adnexal openings of the face.
Structureless (homogeneous) areas	Areas devoid of dermoscopic structures and without regression. These areas can be pigmented or nonpigmented. If the area is uniformly dark, it is referred to as a “blotch” (see below).	Lack of melanin or presence of melanin in all layers of the skin.
Dots	Small, round structures less than 0.1 mm in diameter that may be black, brown, gray, or bluish.	Aggregates of melanocytes or melanin granules. Black dots represent pigment in the upper epidermis or stratum corneum. Brown dots represent pigment at the dermoepidermal junction. Gray-blue dots represent pigment in the papillary dermis.
Peppering	Tiny, blue-gray granules.	Melanin deposited as intracellular (mostly within melanophages) or extracellular particles in the upper dermis
Globules	Round to oval structures that may be brown, black, or red with diameters greater than 0.1 mm.	Nests of melanocytes in the dermis and dermal–epidermal junction.
Streaks (pseudopods, radial streaming)	Radially arranged projections of dark pigment (brown to black) at the periphery of the lesion.	Confluent junctional nests of melanocytes.
Blotches	Dark brown to black, usually homogeneous areas of pigment that obscure underlying structures.	Aggregates of melanin in the stratum corneum, epidermis, and upper dermis.
Regression areas	White, scarlike depigmentation (lighter than the surrounding skin, shiny white under polarized dermoscopy) often combined with or adjacent to blue-gray areas or peppering.	Scarlike changes: thickened fibrotic papillary dermis, dilated blood vessels, sparse lymphocytic infiltrates, and variable numbers of melanophages.
Blue-white veil	Irregular, confluent blue pigmentation with an overlying white “ground-glass” haze.	Aggregation of heavily pigmented cells (usually melanoma cells or melanophages) with compact orthokeratosis of the stratum corneum and acanthosis (thickened epidermis).
Vascular pattern	See Table 15-5 for vascular terminology.	Tumor neoangiogenesis and dilated blood vessels in the papillary dermis (“vascular blush”).
Milia-like cysts	Round whitish or yellowish structures that shine brightly (like “stars in the sky”) under nonpolarized dermoscopy.	Horn pseudocysts.
Comedo-like openings	“Blackhead”-like plugs on the surface of the lesion.	Concave clefts in the surface of the epidermis, often filled with keratin.
Fingerprint-like structures	Thin, light brown, parallel running lines.	Probably represent thin, elongated, pigmented epidermal rete ridges.
Ridges and fissures	Cerebriform surface resulting in gyri (ridges) and sulci (fissures). Confluence of adjacent comedo-like openings will create a fissure.	Wedge-shaped clefts of the surface of the epidermis, often filled with keratin (fissures).
“Moth-eaten” border	Concave invaginations of the lesion border.	Not available.
Leaf-like areas	Brown to gray-blue, discrete bulbous structures resembling a leaf pattern.	Large, complex nodules of pigmented basal cell carcinoma in the upper dermis.
Spoke-wheel–like structures	Well-circumscribed brown to gray-blue-brown radial projections meeting at a darker brown central hub.	Nests of basal cell carcinoma radiating from the follicular epithelium.
Large blue-gray ovoid nests	Large, well-circumscribed areas, larger than globules.	Large nests of basal cell tumor in the dermis.
Multiple blue-gray globules	Round well-circumscribed structures that, in the absence of a pigment network, suggest basal-cell carcinoma.	Small nests of basal cell tumor in the dermis.
Lacunae	Red, maroon, or black lagoons.	Dilated vascular spaces.
Parallel patterns	On acral areas, parallel rows of pigmentation following the furrows (nevi) or ridges (melanoma) of the dermoglyphics.	Pigmented melanocytes in the furrows (crista limitans) or ridges (crista intermedia) of acral skin.

The overall clinical diagnostic accuracy for malignant melanoma (MM), without the added benefit of dermoscopy, for experienced dermatologists is only about 60%. Dermoscopy enhances the diagnostic accuracy for MM and helps triage those lesions requiring a biopsy. In a large meta-analysis of dermoscopy studies, Bafounta and colleagues revealed that dermoscopy significantly increased diagnostic accuracy (by 49%) compared with unaided examination, with mean sensitivity increasing by 19% and mean specificity by 6%. The increase in specificity with dermoscopy translates into a reduction in the excision of benign lesions. This is consistent with a retrospective analysis that showed a significant reduction in the benign–malignant ratio of excised melanocytic lesions from 18 : 1 in the predermoscopy era to 4 : 1 after dermoscopy was implemented by trained clinicians. The benefit of using dermoscopy greatly depends on experience, and reliance on dermoscopy by untrained or less experienced examiners was found to be no better than clinical inspection alone. However, studies indicate that participation in short dermoscopy training courses improves confidence and diagnostic performance of nonexperts when evaluating lesions by dermoscopy. The benefits provided by the use of dermoscopy are presented in Box 15-1.

Evaluation Technique: Two-Step Dermoscopy Algorithm

The two-step dermoscopy algorithm forms the foundation for the dermoscopic evaluation of skin lesions (Fig. 15-2).

Figure 15-2 The two-step algorithm for the evaluation of the pigmented lesions. In the first step, the observer must decide whether the lesion is of melanocytic or nonmelanocytic origin (Step 1 [A]), using a stepwise evaluation of dermoscopic features (Step 1 [B]). In the second step, the observer differentiates between benign melanocytic nevi and melanoma using pattern analysis or one of the score-based algorithms mentioned in the text (e.g., ABCD rule, Menzies method, seven-point checklist).

The first step in performing dermoscopy requires that the observer classify the lesion under investigation as either a growth of melanocytic or nonmelanocytic origin. On nonglabrous skin, the presence of a pigment network, aggregated globules, streaks, or homogeneous blue pigmentation identifies the lesion as melanocytic (Figs. 15-3 through 15-6). In addition, a pseudonetwork pattern can be seen in melanocytic lesions on facial skin (Fig. 15-7). On the other hand, melanocytic lesions on the palms and soles are recognized primarily by the presence of a parallel pigment pattern (Fig. 15-8; see Table 15-1). If, however, the lesion does not manifest any of the aforementioned melanocytic criteria, the observer seeks to identify specific criteria that can identify the lesion as a nonmelanocytic lesion (see Fig. 15-2) such as basal cell carcinoma (Figs. 15-9 through 15-11), squamous cell carcinoma (Fig. 15-12), he mangioma (Fig. 15-13), seborrheic keratoses (Figs. 15-14 and 15-15), or dermatofibroma (Fig. 15-16). In addition, a group of lesions exists that is termed “structureless” in that they do not manifest any melanocytic or nonmelanocytic lesion structures. Because it is not uncommon to encounter amelanotic and hypomelanotic MMs that are structureless, all such lesions should be viewed with extreme suspicion, especially if the lesion manifests linear irregular or dotted blood vessels, both commonly seen in MM (Fig. 15-17).