Clinics in Dermatology (2014) 32, 94–100

Diagnosis and management of facial pigmented maculesAimilios Lallas, MDa, Giuseppe Argenziano, MDa, Elvira Moscarella, MDa,Caterina Longo, MDa, Vito Simonetti, MDa, Iris Zalaudek, MDa,b,⁎

aSkin Cancer Unit, Arcispedale Santa Maria Nuova, Istituto di Ricovero e Cura a Carattere Scientifico, Viale Risorgimento 80,42100 Reggio Emilia, ItalybDepartment of Dermatology, Medical University of Graz, Auenbruggerplatz 8, 8036 Graz, Austria

Abstract The differential diagnosis of pigmented macules on the mottled chronic sun-damaged skin ofthe face is challenging and includes lentigo maligna (LM), pigmented actinic (solar) keratosis, solarlentigo, and lichen-planus-like keratosis. Although dermatoscopy improves the diagnostic accuracy ofthe unaided eye, the accurate diagnosis and management of pigmented facial macules remains one of themost challenging scenarios in daily practice. This is related to the fact that pigmented actinic (solar)keratosis, lichen-planus-like keratosis, and LM may reveal overlapping criteria, making their differentialdiagnosis clinically difficult. For this reason, practical rules have been introduced, which should help tominimize the risk for inappropriate diagnosis and management of LM.© 2014 Published by Elsevier Inc.

Introduction

The differential diagnosis of pigmented macules on themottled chronic sun-damaged skin of the face is challeng-ing and includes lentigo maligna (LM), pigmented actinic(solar) keratosis (PAK), solar lentigo (SL), and lichen-planus-like keratosis (LPLK).1,2 Given that the naturalcourse and prognosis vary significantly among thesedifferent entities, accurate diagnosis is mandatory to ensureappropriate management.

Dermatoscopy improves the diagnostic accuracy com-pared with the unaided eye and accordingly, dermatoscopybecame an integral part of the clinical examination of skintumors3; however, the accurate diagnosis and managementof pigmented macules on the face remains one of the mostchallenging scenarios in daily practice, even if coupled withdermatoscopy.4

⁎ Corresponding author. Tel.: +43 676 33 282 69; fax: +39 069 762 5822.E-mail address: iris.zalaudek@gmail.com (I. Zalaudek).

0738-081X/$ – see front matter © 2014 Published by Elsevier Inc.http://dx.doi.org/10.1016/j.clindermatol.2013.05.030

Clinical features of facial pigmented macules

LM, SL, seborrheic keratosis (SK), LPLK, and PAK arecommon facial lesions that typically develop after thefourth decade of life.1,2,5,6 They share the clinicalappearance of a flat, pigmented macule of different sizeand color. Accordingly, neither age nor clinical criteriaappear useful for differentiating between these entities;however, there are some clinical differences between LMand the other nonmelanocytic skin lesions that should beconsidered in the differential diagnosis.

First, LM and PAK reveal significant sex-relateddifferences. Whereas PAK occurs at higher frequency inmen, LM has a certain predilection for women7,8; thus, sexshould be always considered in the differential diagnosis ofpigmented macules on the face.

Second, whereas PAK typically reveals a scaly and roughsurface, LM appears smooth on palpation.1,2 Accordingly,palpation of a given facial lesion represents an important partof the clinical examination.

Finally, LM develops more commonly as solitary ordifferent-looking lesion compared with SL and PAK, which

95Pigmented facial macules

typically present as multiple spots with a similar appear-ance.1,2,9 Accordingly, a solitary lesion appears to be moresuggestive of LM, whereas multiple similar lesions favorsomewhat the diagnosis of SL or PAK.

Dermatoscopic patterns of facial pigmentedmacules

The dermatoscopic aspects of facial melanocytic lesionsdiffer from the ones described for other locations (Figure 1). Apigment network representing the dermatoscopic hallmark ofmelanocytic tumors on the torso and extremities is only rarelydetected in facialmelanocytic skin tumors.9 This is reasonable,because pigment network results from epidermal melanin(either in melanocytes or keratinocytes) along elongated reteridges, whereby the tips of the rete ridges appear as networkholes and the lateral borders as network lines.10

Instead, the dermoepidermal junction of chronically sun-damaged facial skin appears flattened and may even lack reteridges. Pigmented keratinocytes or melanocytes along thisflattened dermoepidermal junction appear as structurelessdiffuse brown pigmentation on dermatoscopy; however, thisotherwise structureless diffuse brown pigmentation on theface is interrupted by numerous, variable broad andhypopigmented “holes,” which correspond to hair folliclesand sweat gland openings. Because the combination ofdiffuse brown pigmentation and nonpigmented adnexalopenings is reminiscent of a kind of network, the pattern of

Fig. 1 Dermatoscopic-histopathologic correlates of pigment network (leflower image). True pigment network correlates with pigmentation along elobrown pigmentation, which is intermingled by nonpigmented follicular openthe reader is referred to the web version of this contribution.)

facial melanocytic and nonmelanocytic pigmented maculesis also called a “pseudonetwork” pattern.9,11

It is important to emphasize that the pseudonetworkpattern occurs likewise in melanocytic and nonmelanocyticlesions.11,12 Accordingly, the diagnosis of a given pigmentedfacial macule relies on the detection of additional specificcriteria. The specific dermatoscopic criteria of melanocyticand nonmelanocytic skin lesions are summarized in Table 1.

A dermatoscopic model in the progressionof LM

A 2000 study proposed based on dermatoscopic criteria afour-step model in the progression of LM.13 According tothis model, asymmetric pigmented follicular openings (alsocalled gray circles), gray dots within the follicular openingor so-called circles within circles, represent the very initialcriteria in the development of LM. These structuressubsequently develop into a so-called annular-granularpattern, which consists of aggregated fine gray dots, grayglobules, and streaks around the follicle. With furtherprogression, these streaks become longer and intersect withneighboring streaks, forming finally rhomboidal structures(ie, angulated lines between hair follicles). Finally, begin-ning invasion is characterized by the development ofobliterated hair follicles (filled with black or blue-grayblotches) and structureless blue areas, white scarlike areas,and milky-red areas (Figure 2).

t upper and lower images) and facial pseudonetwork (right upper andngated rete ridges, whereas pseudonetwork results from a structurelessings. (For interpretation of the references to color in this figure legend,

Table 1 Dermatoscopic criteria of flat facial lesions

Lentigo maligna Pigmented actinic keratosis Solar lentigo/earlyseborrheic keratosis

Lichen-planus-like keratosis

Pseudonetwork Pseudonetwork Pseudonetwork Brownish gray granuleslocalized or diffuse

Asymmetric pigmentedfollicular openings

Slate-gray dots Light brown fingerprintareas

Bluish gray granules localizedor diffuse

Slate gray dots Annular-granular pattern Yellow opaque areas Whitish gray granules localizedor diffuse

Slate gray globules Rhomboidal structures Horny pseudocysts Any other criterion of SK or SLRhomboidal structures Black globules Milialike cystsAnnular-granular pattern Slate-gray globules Cerebriform structuresGray brown streaks Black dots Moth-eaten borderDark/blue homogenous areas Asymmetric pigmented

follicular openingsSharp demarcation

White scarlike areas Circle within a circle Jelly signMilky red areas Slate-gray areas Hairpin vesselsIncreased density of thevascular network

Gray-brown streaks Asymmetric pigmentedfollicular openings

Red rhomboidal structures Broken pseudonetworkTargetlike patterns Keratin plugsDarkening at dermatoscopicexamination

Circle within a circleZigzag pattern

96 A. Lallas et al.

It appears remarkable that according to this model, theearliest signs of LM development are related to predominantfollicular involvement and not to the interfollicular epider-mis. This observation has led some to question whether LMarises from cancer stem cells of the hair bulge rather thanfrom transformed epidermal melanocytes.14

Fig. 2 Dermatoscopic patterns of lentigo maligna. Panel A, Gray dots wfollicles and gray circles within a circle (arrows); panel C rhomboidal stpanel D annular-granular structures (arrows) and obliterated hair folinterpretation of the references to color in this figure legend, the reader

Dermatoscopic differential diagnosis of facialpigmented macules

Reasonably, if a lesion exhibits fully developed clinicaland/or specific dermatoscopic criteria, a straightforwarddiagnosis is feasible in most cases.

ithin the hair follicle (arrows); panel B asymmetric pigmented hairructures appearing as gray lines (arrows) forming a rhomboid; andlicles appearing as gray-blue structureless blotches (circle). (Foris referred to the web version of this contribution.)

97Pigmented facial macules

LM, PAK, and LPLK may show at times overlappingfeatures (Table 1); for these equivocal cases, the clinicallyrelevant question is whether dermatoscopy aids thecorrect management.

With regard to this question, it must be admitted thatappropriately designed diagnostic accuracy studies arelacking, although preliminary data suggest a profitable roleof dermatoscopy in differentiating LM from other non-melanocytic macules.

Criteria of LM versus SL and SK

Comparison of criteria between LM, SK, and SLallowed the identification of a set of four criteria thatallowed prediction of diagnosis of LM with a sensitivity of89% and a specificity of 96%.13 These criteria areasymmetric pigmented follicular openings, dark rhomboi-dal structures, slate-gray globules, and slate-gray dots.Although each single criterion may be seen also in SK orSL, the presence of all four features is strongly suggestiveof LM. In contrast, light brown fingerprint areas, yellowopaque areas, milia cysts, moth-eaten border, and sharpdemarcation (jelly sign) have been significantly associatedwith the diagnosis of SK or SL.13,15–17 More recently,elongated brown circles (fat fingers) have been added ascriteria of early SK18 (Figure 3).

Fig. 3 Dermatoscopic patterns of solar lentigo (SL) and early seborrhwavy parallel brown lines. Panel B, Sharp demarcation with convex and coAlthough the central fully developed SK reveals classic criteria of SK suchof SL reveals fingerprintlike structures. Panel D, Light brown elongated cireferences to color in this figure legend, the reader is referred to the web

Criteria of LM versus LPLK

LPLK is a term referring to an SL or SK undergoingregression.6 The discrimination between LPLK and LM isproblematic and can be basically made only if areas of thepreexisting benign lesion (SK or SL) are still preserved.19

Instead, fully or nearly fully regressed LPLK is characterizedby diffuse brownish gray granules, which may coalescenceto form globules, streaks, or even structures similar torhomboids.19 Because LM may exhibit the same features, abiopsy of a lesion dermatoscopically characterized by signsof evident regression (ie, localized or diffuse gray granules)should always be performed (Figure 4).

Criteria of PAK versus LM

A recent comparative study confirmed previous observa-tions suggesting that LM and PAK exhibit strikingly similarpatterns on dermatoscopy.20 Effectively, any of the estab-lished criteria of LM can be also seen in PAK, although blackblotches within the follicular opening as seen in the late stageof melanoma progression occur at higher frequency in LMthan PAK.20 Conversely, keratin plugs and a broken-up,superficial brown pseudonetwork is more suggestive ofPAK.4,21 In doubtful cases, histopathology is required todifferentiate between LM and PAK; however, the

eic keratosis (SK). Panel A, Fingerprintlike structures appearing asncave ends (moth-eaten borders). Panel C, SK developing from SL:as multiple milia cysts and comedo openings, the peripheral flat partrcles reminiscent of “fat” fingers (arrows). (For interpretation of theversion of this contribution.)

Fig. 4 Side-by-side comparison between lentigo maligna (LM) (panel A) and lichen-planus-like keratosis (panel B) showingoverlapping features of gray dots. Side-by-side comparison between LM (panel C) and pigmented actinic (solar) keratosis (panel D)showing rhomboidal structures. (For interpretation of the references to color in this figure legend, the reader is referred to the webversion of this contribution.)

98 A. Lallas et al.

discrimination between the two entities may be evenhistopathologically difficult, when it is not clear whetherthe pigmented atypical cells in the basal layer arekeratinocytes or melanocytes1,2 (Figure 4).

Practical rules and clues for the diagnosis andmanagement of facial pigmented macules

Although dermatoscopy provides valuable additionalmorphologic information, the differential diagnosis of flatpigmented macules on the face remains a challenge. Three

Fig. 5 Gray color is the single best criterion in the differential diagnosisinterpretation of the references to color in this figure legend, the reader i

practical rules may help to minimize the risk for inappro-priate diagnosis and management.

Importance of gray color

The single most important criterion in the differentialdiagnosis of LM from SL and SK is related to the presenceversus absence of gray color, respectively (Figure 5). Thiscan be explained by the different histopathologic correlatesof colors in dermatoscopy. Melanin located in the stratumcorneum, at the dermoepidermal junction, upper dermis, and

of solar lentigo and seborrheic keratosis from lentigo maligna. (Fors referred to the web version of this contribution.)

99Pigmented facial macules

deep dermis gives rise to black, brown, gray, and blue,respectively.22,23

That SL and SK contain melanin exclusively at theepidermal level (ie, pigmented keratinocytes) explains whythese lesions display only brown color on dermatoscopy.24

Instead, free or intracellular melanin in the upper dermis orintrafollicular melanin give rise to gray structures such asgray dots, globules, lines, or circles as typically seen inLM.13 Dermal melanin and its correlated gray dermato-scopic color are also seen in PAK and LPLK.19,20 Graycolor is therefore insufficient to differentiate among LM,LPLK, and PAK.

Importance of correlating clinical,dermatoscopic, and histopathologic findings

Dermatoscopic and histopathologic findings shouldalways be combined with clinical information and inter-preted within the clinical context of the patient. For instance,a different management approach should be applied inpatients with a solitary lesion compared with patients withmultiple lesions. More specifically, the presence of multiplelesions favors the diagnoses of PAK, SK, or SL, whereasdevelopment of multiple LMs is exceedingly rare.1,2,5 Thecomparative approach (ie, examination of all lesions) can beof further help in the management of patients with multiplefacial spots because it allows the identification of similarpatterns among all lesions (ie, signature pattern of PAK orSL). On the contrary, the threshold for biopsy should belowered if examining a solitary lesion that cannot beconfidently diagnosed as PAK, SK, or SL.

Histopathology is regarded as gold standard in thediagnosis of clinically equivocal melanocytic and nonmela-nocytic skin tumors, but it must be emphasized that it is not

Fig. 6 Dermatoscopy helps the selection of the most represen-tative area to undergo biopsy. In the case shown, the biopsy wasmade from the area showing gray asymmetric pigmented hairfollicles (circle) and reveals melanoma in situ. (For interpretationof the references to color in this figure legend, the reader is referredto the web version of this contribution.)

free of limitations. The diagnosis of facial lesions inparticular may be challenging because LM can escape thehistopathologic diagnosis, if the biopsy has been performedin a nonrepresentative area.25 Dermatoscopy can improve theselection of representative areas to undergo biopsy; in detail,biopsy should be performed in areas showing the mostsuggestive features of LM (Figure 6).13

Early LM may display unremarkable features on histopa-thology.26 This carries a certain risk for underdiagnosis ofmelanoma as a junctional nevus. Clinicians should always payparticular attention if confronted with a given histopathologicdiagnosis of a “junctional nevus” on the face of an elderlypatient and critically review such diagnosis in the context ofthe patient (considering the history, age, sex, location, andclinical or dermatoscopic features of the lesion).

Never perform ablative treatment ofequivocal lesions

Although LM, PAK, SL, SK, and LPLK share severalsimilar clinical, dermatoscopic, and histopathologic charac-teristics, they require a quite different management approach.Surgical excision is undoubtedly the treatment of choice forLM.27 Instead, nonablative modalities such as cryotherapy orlaser therapy are reasonable options in the treatment of PAKor cosmetically disturbing SK, SL, and LPLK.28

Although radiotherapy, cryotherapy, and laser therapyrepresent alternative treatment options for LM in selectedcases when surgical approach is not feasible, these optionsare problematic for two reasons: First, these options havebeen associated with a high risk for recurrence.29 Second,there are cases reporting a rapid progression of slow-growingLM to invasive, biologically aggressive nodular melanomaafter cryotherapy. This suggests a possible influence ofminimally invasive surgical procedures on the biologicalbehavior of melanoma.

Consequently, nonsurgical treatment modalities should bepersevered for unequivocal cases that can be diagnosed assurely benign. Any doubtful lesion should undergo biopsy orat least regularly followed up using sequential digitaldermatoscopic monitoring; destructive modalities should beavoided in the management of equivocal facial lesions.30–32

References

1. Cohen LM. Lentigo maligna and lentigo maligna melanoma. J AmAcad Dermatol. 1995;33:923-936.

2. Uhlenhake EE, Sangueza OP, Lee AD, et al. Spreading pigmentedactinic keratosis: A review. J Am Acad Dermatol. 2010;63:499-506.

3. Argenziano G, Soyer HP, Chimenti S, et al. Dermoscopy of pigmentedskin lesions: Results of a consensus meeting via the internet. J Am AcadDermatol. 2003;48:679-693.

4. Zalaudek I, Ferrara G, Leinweber B, et al. Pitfalls in the clinical anddermatoscopic diagnosis of pigmented actinic keratosis. J Am AcadDermatol. 2005;53:1071-1074.

100 A. Lallas et al.

5. Mehregan AH. Lentigo senilis and its evolution. J Invest Dermatol.1975;65:429-433.

6. Jang KA, Kim SH, Choi JH, et al. Lichenoid keratosis: Aclinicopathologic study of 17 patients. J Am Acad Dermatol.2000;43:511-516.

7. Memon AA, Tomenson JA, Bothwell J, et al. Prevalence of solardamage and actinic keratosis in a Merseyside population. Br JDermatol. 2000;142:1154-1159.

8. Cohen LM, McCall MW, Hodge SJ, et al. Successful treatment oflentigo maligna and lentigo maligna melanoma with Mohs micro-graphic surgery aided by rush permanent sections. Cancer. 1994;73:2964-2970.

9. Stolz W, Schiffner R, Burgdorf WH. Dermatoscopy for facialpigmented skin lesions. Clin Dermatol. 2002;20:276-278.

10. Braun RP, Rabinovitz HS, Oliviero M, et al. Dermoscopy of pigmentedskin lesions. J Am Acad Dermatol. 2005;52:109-121.

11. Argenziano G, Soyer HP, De Giorgi V, et al. Dermoscopy: AnInteractive Atlas. Milan, Italy: EDRA. 2000.

12. Johr RH, Stolz W. Lentigo maligna and lentigo maligna melanoma.J Am Acad Dermatol. 1997;37:512.

13. Schiffner R, Schiffner-Rohe J, Vogt T, et al. Improvement of earlyrecognition of lentigo maligna using dermatoscopy. J Am AcadDermatol. 2000;42:25-32.

14. Zalaudek I, Marghoob AA, Scope A, et al. Three roots of melanoma.Arch Dermatol. 2008;144:1375-1379.

15. Stante M, De Giorgi V, Stanganelli I, et al. Dermoscopy forearly detection of facial lentigo maligna. Br J Dermatol. 2005;152:361-364.

16. Sahin MT, Ozturkcan S, Ermertcan AT, et al. A comparison ofdermatoscopic features among lentigo senilis/initial seborrheic kerato-sis, seborrheic keratosis, lentigo maligna and lentigo maligna melanomaon the face. J Dermatol. 2004;31:884-889.

17. Tanaka M, Sawada M, Kobayashi K. Key points in dermatoscopicdifferentiation between lentigo maligna and solar lentigo. J Dermatol.2011;38:53-58.

18. Kopf AW, Rabinovitz H, Marghoob A, et al. “Fat fingers”: A clue in thedermatoscopic diagnosis of seborrheic keratoses. J Am Acad Dermatol.2006;55:1089-1091.

19. Zaballos P, Marti E, Cuellar F, et al. Dermoscopy of lichenoidregressing seborrheic keratosis. Arch Dermatol. 2006;142:410.

20. Akay BN, Kocyigit P, Heper AO, et al. Dermatoscopy of flat pigmentedfacial lesions: Diagnostic challenge between pigmented actinickeratosis and lentigo maligna. Br J Dermatol. 2010;163:1212-1217.

21. Peris K, Micantonio T, Piccolo D, et al. Dermatoscopic features ofactinic keratosis. J Dtsch Dermatol Ges. 2007;5:970-976.

22. Weismann K, Lorentzen HF. Dermatoscopic color perspective. ArchDermatol. 2006;142:1250.

23. Rosendahl C, Cameron A, Tschandl P, et al. The cause ofdermatoscopic grey circles in non-invasive melanomas: An hypothesissupported by histopathological correlation. Poster presented at: 2ndCongress of the International Dermoscopy Society; November 12–14,2009; Barcelona, Spain. Available from: http://www.greycircles.blogspot.com. Accessed September 28, 2010.

24. Braun RP, Rabinowitz H, Oliviero M, et al. Dermatoscopy ofpigmented lesions. Ann Dermatol Venereol. 2002;129:187-202.

25. Larsen TE, Grude TH. A retrospective histological study of 669 cases ofprimary cutaneous malignant melanoma in clinical stage I. Acta PatholMicrobiol Scand. 1979;87:255-260.

26. Larsen TE, Little JH, Grell SR, et al. International pathologistscongruence survey on quantitation of malignant melanoma. Pathology.1980;12:245-253.

27. Bichakjian CK, Halpern AC, Johnson TM, et al. Guidelines of care forthe management of primary cutaneous melanoma. J Am Acad Dermatol.2011;65:1032-1047.

28. Stockfleth E, Ferrandiz C, Grob JJ. Development of a treatmentalgorithm for actinic keratoses: A European Consensus. Eur JDermatol. 2008;18:651-659.

29. Zalaudek I, Horn M, Richtig E, et al. Local recurrence in melanoma insitu: Influence of sex, age, site of involvement and therapeuticmodalities. Br J Dermatol. 2001;148:703-708.

30. Dummer R. About moles, melanomas, and lasers: The dermatologist'sschizophrenic attitude toward pigmented lesions. Arch Dermatol.2003;139:1405-1406.

31. Gottschaller C, Hohenleutner U, Landthaler M. Metastasis of amalignant melanoma 2 years after carbon dioxide laser treatment of apigmented lesion: Case report and review of the literature. Acta DermVenereol. 2006;86:44-47.

32. Rocamora V, Puig L, Romani J, et al. Amelanotic lentigo malignamelanoma: Report of a case and review of the literature. Cutis. 1999;64:53-56.