INTRODUCTION — Dermoscopy is a noninvasive in vivo technique used for the examination of lesions and diseases involving the glabrous and nonglabrous skin as well as abnormalities involving the hair and nails. Dermoscopy, dermatoscopy, epiluminescence microscopy, incident light microscopy, and skin-surface microscopy are synonyms. Dermoscopy is performed with a handheld instrument called a dermatoscope. The procedure allows for the visualization of subsurface skin structures in the epidermis, dermoepidermal junction, and superficial dermis, which are generally not visible to the naked eye [1-3].
The general principles of dermoscopy; dermoscopic structure terminology; dermoscopic evaluation of skin lesions; and dermoscopic evaluation of facial, mucosal, and acral volar skin lesions are discussed separately. Dermoscopy of nail pigmentation and dermoscopic algorithms for skin cancer triage are also discussed separately.
●(See "Overview of dermoscopy".)
●(See "Dermoscopy of facial lesions".)
●(See "Dermoscopy of mucosal lesions".)
●(See "Dermoscopy of pigmented lesions of the palms and soles".)
●(See "Dermoscopy of nail pigmentations".)
●(See "Dermoscopic algorithms for skin cancer triage".)
PRINCIPLES OF DERMOSCOPIC DIAGNOSIS — The dermoscopic diagnosis of skin lesions involves the recognition of the presence or absence of specific structures to confirm or rule out a given diagnosis. From a cognitive perspective, this task may be accomplished using a bottom-up or a top-down strategy. In the bottom-up strategy, the observer performs a visual search for salient details (individual features) to arrive at a diagnosis, whereas in the top-down strategy, the observer recognizes the general context, generates a hypothesis of the likely clinical diagnosis, and performs a targeted dermoscopic search for specific features to confirm or negate the presumed clinical diagnosis [4].
This topic will review several algorithms and scoring systems that predominantly use a top-down strategy to assist clinicians in the decision-making process as to whether a lesion needs to be biopsied.
PATTERN ANALYSIS: THE REVISED TWO-STEP ALGORITHM FOR SKIN LESION EVALUATION
Overview — The two-step pattern analysis algorithm forms the foundation for the dermoscopic evaluation of skin lesions. It was first introduced by the panel of the Consensus Internet Meeting on Dermoscopy in 2003 and has since undergone several modifications [5-7].
The original two-step algorithm required lesions to be classified as melanocytic (nevus versus melanoma) or as nonmelanocytic (algorithm 1). The requirement to classify lesions as melanocytic (based on the presence of network; streaks; angulated lines; negative network; aggregated globules; homogeneous, blue color; pseudonetwork on facial skin; or parallel pattern on volar lesions) came under criticism since some nonmelanocytic lesions may reveal features resembling the aforementioned structures, and on rare occasions, melanomas may lack these structures [8]. Although a subsequent study provided evidence validating the first step [9], the authors of this topic introduced a modification of the two-step process to address the criticisms and to amalgamate new insights regarding the morphology of benign and malignant lesions.
The updated top-down approach discussed here uses pattern analysis and builds upon the previous two-step algorithm while combining it with concepts derived from the triage amalgamated dermoscopy algorithm (see "Dermoscopic algorithms for skin cancer triage"). The revised version of the two-step algorithm (algorithm 2) eliminates the former requirement to differentiate melanocytic from nonmelanocytic lesions in the first step [6,10,11]:
●Step 1 – In the new approach, the observer is required to make a specific diagnosis by recognizing the patterns associated with common, benign neoplasms (eg, nevi, lentigines, seborrheic keratoses, angiomas) (algorithm 3). If an unequivocal diagnosis of one of the benign lesions cannot be made with confidence, then the lesion under evaluation will proceed to step 2.
●Step 2 – In step 2, the observer will determine if the overall pattern of the lesion (ie, distribution of colors and structures) is organized or disorganized (algorithm 4). Lesions displaying a disorganized pattern (ie, asymmetric or chaotic) are evaluated in step 2a, and lesions manifesting an organized pattern are evaluated in step 2b:
•Step 2a – Lesions displaying a disorganized pattern have a high likelihood of being malignant, and based on the presence of specific structures, these lesions can usually be subclassified as basal cell carcinoma (BCC), squamous cell carcinoma (SCC), or melanoma.
•Step 2b – Lesions that are organized are likely benign. However, organized lesions displaying any of the following should still raise concern for malignancy:
-Starburst pattern
-Negative network pattern
-Vessels
-Ulceration
-Blue-black, gray, or white color
Step 1: Benign lesions — In the first step of the revised two-step pattern analysis algorithm (algorithm 2), the observer is required to make a specific diagnosis by recognizing patterns associated with common, benign skin lesions (eg, nevi, dermatofibromas, intradermal nevi, seborrheic keratoses/solar lentigo, angiomas, sebaceous hyperplasias). This list of benign lesions can be expanded upon depending on the observer's experience in recognizing less common, benign lesions, such as clear cell acanthoma (picture 1), poroma, and xanthogranuloma.
Although benign lesions tend to display an overall organized pattern, individual lesions require a contextualized interpretation that considers size and location of the lesion, the patient's age, and lesion history. As an example, a lesion with features of a seborrheic keratosis located on the palm should raise concern since seborrheic keratoses do not normally develop on volar surfaces.
Acquired nevi and small congenital nevi — Acquired and small congenital nevi can be of multiple subtypes (eg, junctional; compound; Spitz, Reed, Clark nevi). Although most of these nevi display an organized pattern, each one requires interpretation considering contextual features, such as history, size, patient age, skin phenotype, anatomical location, and whether the lesion is an outlier (ugly duckling). Outlier nevi are those that do not adhere to the signature nevus pattern displayed by the other nevi in the patient.
Overall, there are three dermoscopic structures that form recognizable patterns (ie, network, globules, and homogeneous areas). Their combinations result in different benign patterns observed (figure 1):
●Diffuse network/reticular pattern and patchy network/reticular pattern – These patterns consist of an organized network composed of lines that have minimal variation in color and thickness. The holes of the network also appear relatively homogeneous in size. The network tends to have shades of brown and rarely displays gray or black color. The network tends to fade toward the periphery. The diffuse, reticular pattern is commonly seen in melanocytic nevi with a prominent, junctional component (ie, junctional nevi, superficial congenital nevi) (picture 2).
The patchy network pattern consists of focal network patches that are distributed in a symmetric and organized manner and separated by homogeneous, structureless areas that are of the same color or slightly darker than the background skin. The network patches all have the same type of network, with minimal variability in the thickness and color of the lines. This pattern is commonly seen in acquired melanocytic nevi and some congenital nevi (picture 3).
•Key contextual interpretation – Isolated, large, acquired, lentiginous lesions in older individuals located on sun-damaged skin with a network or patchy network pattern should raise suspicion since melanoma on sun-damaged skin (lentigo maligna type) can manifest this pattern [12].
●Peripheral network with central hypopigmentation or hyperpigmentation – The peripheral network is typical/regular in nevi exhibiting these patterns. In nevi with central hypopigmentation, the central portion consists of a homogeneous, hypopigmented, structureless area of the same color (or slightly darker) as the background skin. In nevi exhibiting central hyperpigmentation, the central area (blotch) often results from the accumulation of melanin-laden corneocytes in the stratum corneum (ie, lamella), which can usually be tape-stripped off, revealing an underlying typical network.
•Key contextual interpretation – Skin phenotype. Central hypopigmentation (picture 4) is commonly seen in acquired melanocytic nevi in individuals with lightly pigmented skin (picture 5), whereas central hyperpigmentation is commonly seen in acquired melanocytic nevi in individuals with darkly pigmented skin (picture 4). Thus, lesions displaying central hypopigmentation in individuals with darkly pigmented skin and lesions with central hyperpigmentation in individuals with lightly pigmented skin should raise suspicion of melanoma [13-15].
●Globular pattern (including peripheral network with central, brown globules) and globular nevi (including cobblestone globular nevi) – Globules of similar shape, size, and color are distributed throughout the lesion or in the center of the lesion with a regular peripheral network. Globules may be large and angulated, creating a cobblestone pattern as seen in dermal nevi and some congenital nevi (picture 6). On rare occasions, the globules can be white (as seen on balloon cell nevi). While black and blue globules can occasionally be seen in congenital nevi, their presence should raise suspicion for melanoma.
•Key contextual interpretation – The globular pattern is commonly seen in congenital nevi (picture 7) most often located on the head/neck or torso. Thus, newly acquired lesions with a globular morphology occurring in an older individual, and especially if the lesion is located on an extremity, should raise concern for melanoma [14,16,17].
●Reticular pattern with peripheral globules – It consists of a typical network nevus with a peripheral rim of regular, brown globules (picture 8). This pattern is associated with the radial growth phase of Clark/dysplastic nevi and is commonly encountered in younger patients with atypical mole syndrome (see "Atypical (dysplastic) nevi"). On histopathology, nevi with one peripheral rim of globules usually correspond to dysplastic nevi with mild to moderate atypia, while nevi displaying a multitiered, peripheral, globular pattern usually correspond to dysplastic nevi with some degree of a spitzoid morphology [18].
•Key contextual interpretation – Consider age and location. These nevi are most common on the torso of younger to middle-aged patients. Newly acquired lesions displaying peripheral globules should be viewed with suspicion if seen in older individuals, especially if the lesion is located on an extremity or reveals a melanoma-specific structure [14,19-22].
●Starburst pattern – Consists of a darkly pigmented nevus with streaks at the perimeter of the lesion distributed in a symmetric fashion. This pattern is associated with Reed nevi that are in their radial growth phase (picture 8).
•Key contextual interpretation – Consider the patient's age. Although streaks can also be present in superficial spreading melanoma, melanomas are very rare in preadolescent individuals (see "Melanoma in children"). Thus, lesions with a starburst pattern seen in children have an extremely high pretest probability for being Reed nevi. In contrast, lesions with a starburst pattern in older individuals have a heightened pretest probability for melanoma [23].
●Homogeneous, blue pattern – This pattern is the hallmark of blue nevus. The blue color should not be confused with a blue-white veil. In contrast to the blue-white veil, the blue color with white, veil-like appearance in blue nevi encompasses the entire lesion's surface with minimal variation in the blue hues (picture 9). If there are multiple hues (ie, heterogeneous, blue color or nonhomogeneous veil), the diagnosis of melanoma should be considered.
•Key contextual interpretation – In patients with a previous history of invasive melanoma, epidermotropic metastasis, which can mimic blue nevi, should be ruled out [24].
●Homogeneous, brown pattern – The nevus shows brown color with minimal to no variation in its hues. While it is usually devoid of any other structures, on rare occasions one can see a few regular dots/globules and fragments of network. This pattern is associated with small to medium congenital nevi (picture 9).
•Key contextual interpretation – Consider the patient's age. Since these nevi are usually present at birth or develop within the first few years of life and tend to be stable lesions, a newly acquired lesion in an adult demonstrating a homogeneous, brown pattern should be viewed with suspicion. Such lesions, if not palpable, can be digitally monitored and biopsied if they display change [25].
●Homogeneous, tan/pink pattern – Nevi with homogeneous, tan/pink color are featureless or may display some vessels (picture 9).
•Key contextual interpretation – Consider skin phenotype. Since these nevi are seen predominantly in individuals with skin types I to II, a newly acquired nevus with a homogeneous, tan to pink color in individuals with darkly pigmented skin should raise concern for amelanotic or hypomelanotic melanoma. In addition, since amelanotic melanomas are also common in skin types I to II, it remains imperative that these lesions be evaluated in context with the other nevi present on the patient's skin. Outlier lesions should always raise suspicion for melanoma. If melanoma is in the differential diagnosis and the lesion is nonpalpable, then it can be subjected to digital monitoring. Lesions that are raised or manifest change on follow-up should be biopsied [25,26].
Intradermal nevi and dermatofibroma (both palpable/raised lesions)
●Intradermal nevi – Intradermal nevi display a characteristic clinical morphology consisting of a raised, dome-shaped, soft papule located on the face (Miescher nevi) (picture 10) or a sessile, mamillated papule located on the torso (Unna nevi) (picture 11). Intradermal nevi usually demonstrate preserved skin markings and may have hairs emanating from their surface.
•Clinical clues:
-Typically, intradermal nevi have been present on the skin for many years, without any history of change or spontaneous bleeding.
-The skin markings in an intradermal nevus tend to be preserved, whereas they are lost in BCC (resulting in BCCs having a shiny surface).
-The presence of hair within the lesion is highly suggestive of an intradermal nevus.
•Dermoscopy clues (at least one) [27-29]:
-Predominantly comma-shaped vessels or a polymorphous, vascular pattern with at least some comma-shaped vessels (picture 12)
-Brown halo
-Globules (including cobblestone-type globules)
-Small foci of tan to brown, structureless pigment and/or hypopigmented areas
•Dermoscopic clues to differentiate intradermal nevi from basal cell carcinoma – Differentiating intradermal nevi, especially Miescher type, from BCC can be challenging [9]. One of the main reasons for this is that, occasionally, intradermal nevi reveal arborizing-like vessels, which raise concern for BCC. However, clues that can assist in differentiating intradermal nevus from BCC include [30]:
-Wobble test – Intradermal nevi wobble on dermoscopy, while BCC tend not to wobble [31].
-Arborizing vessels – The extent of arborization tends to be much lesser in intradermal nevi than in BCC. In fact, in intradermal nevi, there may be only one arborizing point created by a "Y-shaped" vessel [28]. Furthermore, the vessels in BCC tend to be much sharper; in focus; and with a brighter, red color.
-Shiny, white blotches and strands – On polarized dermoscopy, these structures are seen in BCC but not in intradermal nevi.
●Dermatofibromas – Clinically, dermatofibromas present as firm papules (picture 13) that dimple on later pressure (ie, dimple sign (picture 14)). (See "Overview of benign lesions of the skin", section on 'Dermatofibroma'.)
•Dermoscopy clues:
-A symmetric lesion with a peripheral network and a central, white, scar-like area (picture 15). When viewed with polarized light, the central, scar-like area will usually manifest a pink hue and will also often reveal shiny, white lines.
-Ring-like globules and vessels, which can be seen in the region between the network and the central, scar-like area.
Seborrheic keratoses and solar lentigo — Most of the dermoscopic features of seborrheic keratoses are related to the papillomatous growth of the epidermis and the abundance of keratin in these tumors, and include (figure 2 and picture 16 and algorithm 3) [32]:
●Dermoscopy clues:
•Multiple milia-like cysts (three or more).
•Comedo-like openings.
•Moth-eaten borders.
•Gyri and sulci (also known as fissures and ridges) creating a cerebriform pattern. At times, these gyri and sulci can create a pattern resembling a network.
•Fingerprint-like structures. On rare occasions, these lines can be short and slightly curved ("U shaped"). It is interesting that when present, all of the "U-shaped" structures tend to be oriented in the same vectorial direction, resembling jelly that has been spread with a knife across a slice of bread (ie, jelly sign).
•Hairpin vessels surrounded by a white halo, which correspond to the capillary loops present in the dermal papilla.
•Sharp and well-demarcated borders
•Negative "wobble sign" [31]. The wobble sign allows differentiation of an epidermal, keratinocytic lesion from a lesion with a dermal component (eg, a compound or intradermal nevus).
Solar lentigo usually reveals fingerprint-like structures, sharp and well-demarcated borders, and moth-eaten borders.
●Clinical-dermoscopic variants:
•Heavily pigmented seborrheic keratoses (melanoacanthoma) – Heavily pigmented seborrheic keratoses can display a blue-white veil. Some clues to help differentiate the blue-white veil in a seborrheic keratosis from a melanoma include:
-A seborrheic keratosis will not display any other melanoma-specific structures (eg, streaks; atypical network; angulated lines; negative network; shiny, white lines; atypical dots; and globules) [33].
-Within the blue-white veil of a seborrheic keratosis, one will almost always see multiple comedo-like openings.
-The wobble test, in which the seborrheic keratosis lesion will slide en bloc, whereas a melanoma will not and will wobble instead [31].
•Lichen planus-like keratosis – Usually represents an involuting solar lentigo or seborrheic keratosis. Under dermoscopy, lichen planus-like keratoses can reveal regression structures (granularity/peppering). The presence of these structures may make it difficult to differentiate these lesions from melanoma. Clues that may help in the differential diagnosis include:
-Involution of lentigo/seborrheic keratosis usually occurs centripetally (ie, from the edge). Thus, a presumed lentigo or seborrheic keratosis revealing regression structures in the center, with the appearance of the peppering expanding centrifugally, should raise concern for melanoma.
-In lesions that have not yet fully involuted, one can visualize classic features of lentigo/seborrheic keratosis at the edge opposite or lateral to the area of peppering.
-The granularity in involuting lentigo/seborrheic keratosis tends to be coarser, and often, the granules appear to be clumped.
Other benign lesions
●Criteria for hemangioma/angioma, angiokeratoma, and lymphangioma – Red, maroon, purple, blue-black, or clear lagoons are the diagnostic criteria for hemangioma/angioma, angiokeratoma, and lymphangioma. These features are small, well-demarcated areas (often separated by septa), corresponding to dilated blood vessels in the dermis:
•Angioma – The lacunae tend to have a red to maroon color. Thrombosed lacunae will have a black color (picture 17).
•Angiokeratoma – Lacunae with varying shades of red, maroon, and blue (picture 17). There may be thrombosed lacunae displaying a black color. The central portion of the lesion often has a blue-whitish veil, and the periphery often has an erythematous halo.
•Lymphangioma – Lacunae with a clear to slight yellowish color (picture 18). On occasion, there may be some blood present within a lymphangioma. The blood tends to settle to the bottom of the lacunae due to gravity, creating lacunae displaying two discrete tones (also called two-toned lacunae with the hyphema sign) (picture 19) [34].
●Sebaceous hyperplasia – Whitish/yellowish, globular or polyglobular structures with a resemblance to popcorn (picture 20). Crown vessels at the periphery of the lesion, which radiate towards the center but tend not to cross its midline. Molluscum contagiosum will manifest a similar pattern.
●Clear cell acanthoma – Dotted or glomerular vessels, arranged in a "string of pearls" pattern (picture 1).
Experienced clinicians can recognize other benign lesions (eg, poroma and xanthogranuloma) in this step.
Step 2: Equivocal lesions — Lesions that cannot be diagnosed with certainty as one of the benign lesions listed in step 1 will proceed to step 2 for further evaluation (algorithm 4). In step 2, the observer will determine if the overall pattern of the lesion is organized or disorganized. Organization refers to the distribution of colors and structures within the lesion. Lesions displaying a disorganized pattern (ie, asymmetric, chaotic) are evaluated in step 2a, while lesions manifesting an organized pattern are evaluated in step 2b.
Step 2a: Disorganized lesions
General considerations — In general, disorganized lesions will prove to be cancer or cancer mimickers (algorithm 4).
●The most common diagnosable cancers are BCC, SCC, and melanoma.
●More rare skin malignancies (eg, Merkel cell carcinoma, sebaceous carcinoma, porocarcinoma) will also often appear disorganized. Although no structure or combination of structures can definitively diagnose these rare cancers via dermoscopy, some clues such as color, presence of certain dermoscopic structures (eg, polylobular structures in sebaceous carcinoma or looped vessels with rounded endings in porocarcinoma), and location of the tumor may suggest a specific rare cancer diagnosis.
●Some benign lesions (eg, a subset of dysplastic nevi, spitzoid nevi, BAP1-inactivated melanocytomas) can manifest a disorganized pattern that mimics melanoma. Because dermoscopy cannot discriminate them from melanoma, these lesions need to be biopsied for a correct diagnosis. With all the caveats mentioned above, step 2a focuses on correctly identifying BCC, SCC, and melanoma while not missing other less common skin malignancies.
Criteria for basal cell carcinoma — The diagnostic criteria for BCC include the lack of a pigment network and the presence of at least one positive feature for BCC (figure 3 and picture 21 and algorithm 4) [35]:
●Nonpigmented structures:
•Arborizing vessels
•Ulceration and small erosions
•Shiny, white blotches and strands (seen with polarized dermoscopy) [36]
•Fine, short, superficial telangiectasias
•Multiple aggregated, yellowish-whitish globules
●Pigmented structures:
•Leaf-like areas
•Large, blue-gray, ovoid nests
•Multiple blue-gray, nonaggregated globules
•Spoke-wheel structures, including concentric globules
•Multiple in-focus, fine brown to gray dots
•Blue-gray dots in a bird shot scatter distribution
Criteria for cutaneous squamous cell carcinoma — The diagnostic criteria for cutaneous squamous cell carcinoma (cSCC) include (figure 4 and picture 22A-B and algorithm 4) [37-39]:
●Glomerular or coiled vessels, usually focally distributed towards the periphery of the lesion but can also be diffusely present throughout the lesion or in a linear fashion within the lesion
●Rosettes (seen with polarized light)
●White circles or keratin pearls
●Yellow scale/rough texture
●Brown dots/globules aligned in straight, radially oriented lines, usually located towards the periphery
●Brown circles
Variants
●Keratoacanthoma – Keratoacanthoma is a form of a well-differentiated cSCC with the potential to spontaneously involute. Clinically, keratoacanthoma is characterized by a dome-shaped appearance with central scale/keratin. Dermoscopically, keratoacanthoma often reveals white circles; shiny, white blotches; and looped to twisted-looped (hairpin) vessels with a whitish halo at the periphery (picture 23).
●Actinic keratosis – Actinic keratosis contains cells similar to those seen in cSCC, but the cells tend not to encompass the entire thickness of the epidermis (as they do in unequivocal cSCC). Actinic keratoses reveal a rough texture on palpation. Some consider these lesions to be early cSCC but with the caveats that most do not progress to full thickness atypia and many can spontaneously involute. (See "Actinic keratosis: Epidemiology, clinical features, and diagnosis".)
•Nonpigmented actinic keratosis – Dermoscopy features include:
-Strawberry pattern – Consists of prominent, white, follicular openings with or without rosettes and perifollicular erythema [40].
-Occasionally, some actinic keratoses display large areas with adherent scale-crust (hypertrophic actinic keratosis) and dotted or glomerular vessels. In such lesions, it is almost impossible to differentiate them from full thickness cSCC cancers.
•Pigmented actinic keratosis – Pigmented actinic keratosis usually results from the collision of an actinic keratosis with a solar lentigo. These lesions can be difficult to differentiate from lentigo maligna, but some clues include:
-Actinic keratosis will have a rough texture, whereas lentigo maligna will have a smooth surface.
-The follicular (adnexal/ostial) openings in actinic keratosis tend to be large, whitish, and uniformly distributed. In lentigo maligna, the follicular openings will be of variable size, with asymmetric distribution of pigment (often with grayish hues), follicles surrounded by gray dots, and adnexal openings not uniformly distributed.
-Actinic keratoses will often have many rosettes, whereas lentigo maligna will have few to no rosettes.
-Both pigmented actinic keratosis and lentigo maligna can reveal structures that look like angulated lines. However, unlike lentigo maligna, the angulated lines in pigmented actinic keratosis are due to keratin and, thus, will have a brownish color and a rough texture. In lentigo maligna, the angulated lines will not have a rough texture but will often have grayish tones.
The aforementioned clues usually help differentiate a pigmented actinic keratosis from lentigo maligna. However, at times it can be impossible, and a biopsy or evaluation with reflectance confocal microscopy will be required. Also, it is important to remember that it is not uncommon to find collision lesions, resulting from pigmented actinic keratosis and lentigo maligna, on areas of sun-damaged skin.
Criteria for melanoma — The dermoscopic melanoma-specific structures consist of (picture 24A-D and table 1 and algorithm 4):
●Atypical network.
●Angulated lines.
●Streaks (ie, pseudopods and radial streaming).
●Negative network.
●Off-centered blotch.
●Atypical dots and/or globules.
●Regression structures, including granularity (also known as peppering) and scar-like areas. The presence of granularity and scar-like areas within the same lesion results in the appearance of a blue-white coloration, usually overlying macular areas.
●Blue-white veil overlying raised areas.
●Atypical, vascular structures.
●Shiny, white lines (formerly known as crystalline structures).
●Tan, peripheral, structureless areas.
Step 2b: Organized lesions — Step 2b is reserved for lesions that are not unequivocally benign (as listed in step 1) and manifest an organized architectural morphology on dermoscopy (algorithm 4). In such lesions, it is imperative not to miss a malignancy displaying an organized pattern, such as rare and uncommon tumors (eg, Merkel cell carcinoma, leiomyosarcoma, nodular melanoma, nevoid melanoma, desmoplastic melanoma, spitzoid amelanotic melanoma, epidermotropic metastasis, featureless/structureless/feature-poor melanoma).
●Any organized lesion manifesting any of two patterns, two structures, or four colors should be biopsied [10,41]:
•Patterns – Starburst pattern, negative network
•Structures – Vessels, ulceration
•Colors – Blue, black, gray, white
●Monitoring is reserved only for lesions that are flat (not palpable). Digital monitoring entails taking a baseline dermoscopy image of the lesion, which is then compared with a new image taken three to four months later. Images should be of good quality to be compared side by side. If there is any change (except in the number of milia cysts or pigmentation due to ultraviolet [UV] exposure), a skin biopsy should be considered.
●Other technologies, such as confocal microscopy, electrical impedance spectroscopy, and gene expression profiling via tape stripping of cells off the stratum corneum, may assist in managing these lesions.
●Palpable lesions with the above-mentioned features should never be subjected to monitoring and should always be biopsied to rule out a nodular melanoma.
The rules presented so far in the two-step algorithm can be applied to any lesion on the skin. However, additional dermoscopic criteria exist for lesions located on special sites, including facial skin and the glabrous areas (including the palms, soles, nails, and mucosae). (See "Dermoscopy of facial lesions" and "Dermoscopy of pigmented lesions of the palms and soles" and "Dermoscopy of nail pigmentations" and "Dermoscopy of mucosal lesions" and "Dermoscopy of nonpigmented nail lesions".)
OTHER ALGORITHMS AND METHODS — It should be underscored that the revised two-step process (algorithm 2) does not completely negate the usefulness of the original two-step algorithm (algorithm 1) [9]. Some clinicians still prefer to first segregate lesions into melanocytic and nonmelanocytic lesions. Lesions displaying network; streaks; angulated lines; negative network; aggregated globules; homogeneous, blue color; pseudonetwork; or parallel pattern on palms/soles would be classified as melanocytic lesions.
Additional algorithms and methods that aid the observer in differentiating nevi from melanoma include [2,5,42-46]:
●ABCD (asymmetry, border sharpness, colors, differential dermoscopic structures) rule of dermoscopy (table 2)
●Menzies method (table 3)
●The seven-point checklist (table 4)
●CASH (color, architectural disorder, symmetry, homogeneity/heterogeneity of dermoscopic structures) algorithm (table 5)
Although all methods have a similar sensitivity (table 6), pattern analysis has a superior specificity compared with the other quantitative scoring systems and is the method preferred by most experienced clinicians [5,47]. Novices in dermoscopy will benefit from quantitative methods such as the dermoscopy ABCD rule, Menzies method, and the seven-point checklist, which are relatively simple, accurate, and reproducible [44,48,49].
Online tutorials on dermoscopy can be found at www.dermoscopedia.org, www.dermnetnz.org/doctors/dermoscopy-course/introduction.html, and www.genomel.org/dermoscopy.
ABCD rule of dermoscopy — The ABCD rule of dermoscopy is a semiquantitative scoring system that employs four dermoscopic criteria to evaluate a pigmented lesion: asymmetry, border sharpness, colors, and differential dermoscopic structures (table 2) [42].
●Asymmetry – Asymmetry refers to the contour and distribution of colors and structures within the lesion in none, one, or two perpendicular axes (figure 5). The score for asymmetry ranges between zero and two points. The ABCD rule of dermoscopy and the CASH algorithm are the only algorithms that take into account both contour and distribution of colors and structures.
●Border sharpness – The border is evaluated for the presence of abrupt cutoffs between the lesion and the normal skin. The lesion is divided into a virtual pie with eight sections. For each segment presenting with an abrupt cutoff of pigment, a score of 1 is assigned. Hence, the border scoring ranges between zero and eight points.
●Colors – Presence of any of six colors within the lesion: white, red, light brown, dark brown, blue-gray, and black. The score for color ranges from one to six points.
●Differential dermoscopic structures – Presence of any of five structures including pigment network, homogeneous/structureless areas greater than 10 percent of the lesion (ie, hypopigmented or hyperpigmented blotches and depigmented or scar-like areas), branched streaks, dots, and globules. The score for dermoscopic structures ranges between one and five points.
The scores assigned to each feature are multiplied by a weighted factor and summed up in a final dermoscopy score. Sensitivity ranges from 78 to 90 percent and specificity ranges from 45 to 90 percent among experts and nonexperts [5,7,42,44,45,49-54].
Menzies method — The Menzies method was originally developed to differentiate invasive melanomas from other pigmented lesions (table 3) [55]. The method is based upon the evaluation of two negative features with a low sensitivity for melanoma (0 percent) and nine positive features with a high specificity for melanoma (>85 percent).
●Negative features:
•Symmetry of the pigmentation pattern
•Presence of only one color (black, gray, blue, red, dark brown, or tan)
●Positive features:
•Blue-white veil
•Multiple brown dots
•Pseudopods
•Radial streaming
•Scar-like depigmentation
•Peripheral, black dots/globules
•Broadened network
•Multiple blue/gray dots
•Multiple (five to six) colors
The presence of both negative features virtually excludes the diagnosis of melanoma. For all other lesions, the presence of any one of the positive features raises the suspicion for melanoma. The Menzies method has a sensitivity of 85 to 92 percent and a specificity of 38 to 78 percent among examiners with various degrees of experience [5,7,45,50,53,55,56].
The seven-point checklist — The seven-point checklist is based upon seven dermoscopic features frequently associated with melanoma (table 4) [44].
●Major criteria:
•Atypical pigment network
•Blue-whitish veil
•Atypical, vascular pattern
●Minor criteria:
•Irregular streaks
•Irregular dots/globules
•Irregular blotches
•Regression structures
A score is calculated by summing points, allotted as two points for each of the three major criteria and one point for each of the four minor criteria. A final score of 3 or more suggests melanoma [44]. The seven-point checklist has a sensitivity of 62 to 95 percent and a specificity of 35 to 97 percent among experts and nonexperts [5,7,44,50-52,54,55,57].
The presence of any one of the criteria has been proposed as sufficient to warrant a biopsy in a revised version of this checklist [57]. This revised seven-point checklist lowered the threshold for biopsy, using a total score of 1 (instead of 3) as sufficient to warrant a biopsy. Although this revision increased the sensitivity of the criteria, the authors acknowledge that the most sensitive and specific method to diagnose melanoma requires supportive evidence based on clinical characteristics, follow-up, and the comparative approach [58].
CASH algorithm — CASH is an acronym for color, architectural disorder, symmetry, and homogeneity/heterogeneity of dermoscopic structures (table 5) [45].
This method is based upon evaluating a pigmented lesion for the following:
●Presence of few versus many colors
●Architectural order versus disorder
●Symmetry of shape and pattern versus asymmetry
●Homogeneity versus heterogeneity of dermoscopic structures
The scoring system for the CASH algorithm is shown in a table (table 5). A total CASH score of 8 or more is suspicious of melanoma (range 2 to 17) [45]. A score of 8 was chosen as a threshold that optimizes sensitivity and specificity for individuals with all levels of experience. However, a lower threshold for lesion excision may be appropriate for novices. The CASH algorithm has a sensitivity of 87 to 98 percent and a specificity of 67 to 68 percent [45,59].
DERMOSCOPY FROM A MANAGEMENT PERSPECTIVE
Determining whether a lesion should be biopsied — The primary purpose of examining a skin lesion with a dermatoscope is to determine whether the lesion should be biopsied or not [60]. The decision to biopsy a suspicious lesion should be based upon the combination of the clinical and dermoscopic examination of the lesion in question as well as surrounding lesions (comparative approach) and other relevant information (ie, context), including history of change, associated symptoms, skin type, location, and personal and family history of melanoma and other skin cancers.
In patients with multiple nevi, it is useful to identify the "signature nevus" pattern (the predominant type of nevus) as well as lesions that deviate from the predominant pattern ("ugly duckling" lesions) both clinically and dermoscopically [61,62]. A comparative dermoscopic approach to the patient with multiple nevi reduces the number of excisions of benign nevi [26]. (See "Melanoma: Clinical features and diagnosis", section on 'The 'ugly duckling' sign'.)
After a complete clinical and dermoscopic examination utilizing the revised two-step dermoscopy algorithm, a management decision can be rendered:
●If the lesion is considered to be unequivocally benign (algorithm 2), the patient can be reassured, educated on the importance of skin self-examination, and instructed to return if changes are detected [63,64].
●If the lesion is considered to be a melanoma or other skin malignancy (algorithm 4), it should undergo excisional biopsy and histopathologic examination [65-68].
●If the lesion is considered suspicious or undetermined, there are two options: Perform a biopsy or refer the patient to an expert clinician for further evaluation.
The management decision will depend on several factors, including the pretest probability of the diagnosis of the lesion. For example, a lesion with a spitzoid morphology in a child is less likely to be a melanoma than a similar lesion in an adult. (See "Spitz nevus/tumor in children: Diagnosis and management".)
Based on the pretest probability, the clinician may be more likely to biopsy spitzoid lesions in adults than in children. Other factors that may influence the management decision include whether or not the lesion is an isolated lesion or one in a sea of many nevi and whether or not the lesion is a clinical or dermoscopic outlier lesion.
Lesions referred to an expert for further evaluation may be deemed benign, biopsied, or subjected to short-term monitoring. The rationale behind short-term monitoring is that stable lesions are biologically indolent and represent nevi, whereas changing lesions are biologically dynamic and may be atypical nevi or melanoma [69,70].
Short-term dermoscopic monitoring — Short-term dermoscopic monitoring, which consists of comparing digital dermoscopic images of the same lesion taken approximately three to four months apart, should ideally be performed in specialized centers by experienced clinicians [69-71]. This type of monitoring is only suitable for macular (nonpalpable) lesions; suspicious or atypical, nodular (palpable) lesions should be biopsied.
Short-term dermoscopic monitoring is a safe and accepted approach to monitor these flat (nonpalpable), atypical lesions. In one study, 19 percent of 318 nevi showed a change during this time period (2.5 to 4.5 months), and 11 percent of those changing lesions were found to be early melanomas [69].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Dermoscopy".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Seborrheic keratosis (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Principles of dermoscopic diagnosis – Dermoscopic diagnosis of skin lesions involves the recognition of the presence or absence of specific structures to confirm or rule out a given diagnosis. From a cognitive perspective, this task may be accomplished using a bottom-up or a top-down strategy. (See 'Principles of dermoscopic diagnosis' above and "Overview of dermoscopy".)
●The revised two-step pattern analysis algorithm – The revised version of the two-step pattern analysis algorithm eliminates the former requirement to differentiate melanocytic from nonmelanocytic lesions in the first step (algorithm 2). (See 'Pattern analysis: The revised two-step algorithm for skin lesion evaluation' above.)
•Step 1 – In the first step, the observer is required to make a specific diagnosis by recognizing the patterns associated with common, benign neoplasms (eg, nevi, lentigines, seborrheic keratoses, angiomas) (algorithm 3). If an unequivocal diagnosis of one of the benign lesions cannot be made with confidence during step 1, then the lesion under evaluation will proceed to step 2. (See 'Step 1: Benign lesions' above.)
•Step 2 – In the second step, the observer will determine if the overall pattern of the lesion (ie, distribution of colors and structures) is organized or disorganized (algorithm 4). Lesions displaying a disorganized pattern (ie, asymmetric or chaotic) are evaluated in step 2a, and lesions manifesting an organized pattern are evaluated in step 2b. (See 'Step 2: Equivocal lesions' above.)
-Lesions displaying a disorganized pattern have a high likelihood of being malignant, and based on the presence of specific structures, these lesions are subclassified as basal cell carcinoma (BCC), squamous cell carcinoma (SCC), or melanoma in most cases.
-Lesions displaying an organized pattern are likely benign. However, organized lesions should still raise concern for malignancy if they display any of the following: starburst pattern; negative network pattern; vessels; ulceration; or blue-black, gray, or white color.
●Other algorithms – Other algorithms can be used to differentiate nevi from melanoma. Clinicians with limited experience in dermoscopy may benefit from quantitative methods, such as the ABCD (asymmetry, border sharpness, colors, differential dermoscopic structures) rule, the Menzies method, and the seven-point checklist. These methods are relatively simple, accurate, and reproducible. (See 'ABCD rule of dermoscopy' above and 'Menzies method' above and 'The seven-point checklist' above and 'CASH algorithm' above.)
●Management implications – If a lesion is considered benign after a thorough clinical and dermoscopic examination, the patient can be reassured and educated on the importance of skin self-examination and instructed to return if changes occur. A lesion that shows dermoscopic features of melanoma or other skin cancers should be excised and sent for histopathologic examination. If a lesion is considered suspicious, it should be biopsied, or the patient should be referred to an expert clinician for further evaluation and monitoring. (See 'Dermoscopy from a management perspective' above.)
Short-term dermoscopic monitoring (ie, comparison of digital dermoscopic images of the same lesion taken approximately three to four months apart) should ideally be performed in specialized centers by experienced clinicians. (See 'Short-term dermoscopic monitoring' above.)
Do you want to add Medilib to your home screen?