Dermatos
Dermatos
Dermatoscopy
Pattern analysis of pigmented and non-pigmented lesions
2nd edition
Harald Kittler, Cliff Rosendahl, Alan Cameron, Philipp Tschandl · Dermatoscopy
Powered by TCPDF (www.tcpdf.org)
Harald Kittler
Cliff Rosendahl, Alan Cameron, Philipp Tschandl
Dermatoscopy
Pattern analysis of pigmented and non-pigmented lesions
2nd edition
All rights are reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted, in
any form or by any means, electronic, mechanical, photocopying, recording or otherwise without prior permission
of the copyright owner.
Dermatoscopy is a simple method that anyone can the Department of Dermatology in Vienna. I thank Univ.
learn. However, many doctors still doubt that they can Prof. Dr. Michael Binder; he was my first teacher of
gain sufficient skill in dermatoscopy for it to be useful dermatoscopy and generously provided his camera for
in their everyday clinical practice. One of the main taking many of the dermatoscopic pictures in this book.
reasons for this attitude is the discrepancy between the However, the photographs in this book are not derived
simplicity of the method and the difficulty of the jargon from Vienna alone. Colleagues from all corners of the
used by experts, which may be quite incomprehensible globe generously and selflessly provided photographs,
to the uninitiated. Such jargon rarely adds significant include Giuseppe Argenziano, Ralph Braun, Ian McColl,
information to one’s observation but it does create an Jean-Yves Gourhant, Maggie Oliviero, Harold Rabinovitz,
irksome barrier which makes dermatoscopy unnecessarily Isil Kilinc Karaarslan, Iris Zalaudek, and of course the
difficult to learn. One of the main concerns of this book two co-authors Cliff Rosendahl and Alan Cameron. The
is to remove this barrier from the path of learning. English edition of this book could not have been produced
The reason for writing this book was the clear need for without their help. Philipp Tschandl helped us to inspect
a single source of concise, easily comprehensible, and and prepare the photographs, select references, and
consistent learning materials to teach the method of especially write the chapter on Inflammatory Skin Dis-
pattern analysis. Pattern analysis is a comprehensive and eases. The excellent English draft of the German edition
powerful diagnostic tool and has proved to be the method was produced by Sujata Wagner. I am indebted to her
with the greatest diagnostic accuracy in the majority of for her patience and accuracy. Finally, I wish to thank
studies. The method presented here is not a new invention, the staff of Facultas Verlag (publisher) for their support
or even another new algorithm, but plain and simple and cooperation. Specifically, Hani Aghakhani and
pattern analysis. Its novelty is merely that inaccuracies Norbert Novak worked extensively in designing the book.
and ambiguities have been consistently avoided, and Finally I must thank Dr. Sigrid Neulinger for her extreme
prime importance has been given to clarity, logic, and patience in dealing with innumerable missed deadlines.
consistency in both the language used to describe lesions
and the method of using these descriptions to reach Vienna, October 2011 Harald Kittler
a diagnosis. I would not have undertaken this project
without the encouragement and support of Dr. Elisabeth
Riedl and Dr. A. Bernard Ackerman, who contributed Preface to 2nd edition
to the development of the method by providing a large
number of valuable ideas and concepts. Regrettably, Dr. The success of the first edition took us by surprise. It has
Ackerman passed away suddenly in December 2008. been translated into seven languages. The second edition
He has left a void that cannot be filled. He was a mag- is not merely a reprint of the first. All chapters have been
nificent teacher, full of spirit and verve, and possessed updated and new material added, incorporating sugges-
unique originality. tions of readers and correcting the (inevitable!) mistakes
I also wish to thank those who generously provided that were brought to our attention. Some chapters have
the visual material and thus made it possible to write been completely rewritten. We thank Jean-Yves Gourhant,
this book at all. As the majority of clinical photographs Pedro Zaballos, Iris Zalaudek, Giuseppe Argenziano,
have been derived from the archives of the Department Bengü Nisa Akay, and Bianca Carlos for providing us
of Dermatology in Vienna, I am most grateful to the their images. We have worked hard on the 2nd edition
current Head of the Department, Univ. Prof. Dr. Hubert and we hope that we exceeded your expectations.
Pehamberger who, with no hesitation and with the great-
est willingness, gave permission to use these images. I Harald Kittler, Alan Cameron,
also thank his retired predecessors Univ. Prof. Dr. Klaus Cliff Rosendahl, Philipp Tschandl
Wolff and Univ. Prof. Dr. Herbert Hönigsmann. Most of
the clinical pictures were taken by Andreas Ebner, who Vienna, Austria and Brisbane, Australia,
is an extremely talented and patient photographer at September 2016
1 General Principles....................................................................................................................................... 9
1.2 Indication and Benefits of Dermatoscopy....................................................................................... 10
1.3 Diagnostic Accuracy................................................................................................................... 14
1.4 Training.................................................................................................................................... 15
1.5 Development of the Method........................................................................................................ 16
1.5.1 Pattern Analysis................................................................................................................ 16
1.5.2 Evolution of a diagnostic algorithm .................................................................................... 17
1.5.3 Scoring Systems for Melanocytic Lesions............................................................................. 17
1.5.4 What happened to pattern analysis?.................................................................................. 18
1.5.5 Standardization and Consensus......................................................................................... 19
1.5.6 Critique of diagnostic methods and metaphoric terminology.................................................. 20
2 Principal pigmented skin lesions relevant to dermatoscopy......................................................................... 27
2.1 Melanocytic lesions.................................................................................................................... 27
2.1.1 Melanocytic nevi.............................................................................................................. 27
2.1.2 Melanoma....................................................................................................................... 42
2.2 Non-melanocytic pigmented lesions............................................................................................. 42
2.2.1 Vascular proliferations, vascular malformations and hemorrhage............................................ 42
2.2.2 Melanotic macules............................................................................................................45
2.2.3 Benign epithelial neoplasms............................................................................................... 49
2.2.4 Malignant epithelial neoplasms (“keratinocyte cancer”)......................................................... 50
2.2.5 Adnexal neoplasms.......................................................................................................... 50
2.2.6 Dermatofibroma............................................................................................................... 51
2.2.7 Other pigmented lesions relevant to dermatoscopy............................................................... 51
3 Pattern Analysis – Basic Principles.............................................................................................................. 53
3.1 Basic elements........................................................................................................................... 53
3.2 Basic patterns............................................................................................................................ 53
3.2.1 Pattern of lines................................................................................................................. 53
3.2.2 Pattern of dots.................................................................................................................. 55
3.2.3 Pattern of clods................................................................................................................ 55
3.2.4 Pattern of circles............................................................................................................... 55
3.2.5 Pattern of pseudopods...................................................................................................... 55
3.2.6 Structureless pattern.......................................................................................................... 55
3.2.7 Combinations of patterns................................................................................................... 62
3.3 Colors...................................................................................................................................... 62
3.3.1 Melanin.......................................................................................................................... 62
3.3.2 Other pigments................................................................................................................ 64
3.3.3 Color combinations........................................................................................................... 65
3.4 Descriptions of pigmented lesions on the basis of patterns and colors............................................... 65
3.5 Clues........................................................................................................................................ 66
3.6 Characteristic features of pigmented non‑melanocytic lesions.......................................................... 74
3.6.1 Proliferation of vessels....................................................................................................... 74
3.6.2 Intracorneal hemorrhage...................................................................................................77
3.6.3 Solar lentigo, seborrheic keratosis and lichen planus-like keratosis..........................................77
3.6.4 Dermatofibroma............................................................................................................... 81
3.6.5 Melanotic macules............................................................................................................ 81
3.6.6 Pigmented basal cell carcinoma......................................................................................... 89
3.6.7 Squamous cell carcinoma.................................................................................................. 89
3.7 Characteristic features of melanocytic lesions................................................................................ 97
3.7.1 Melanocytic nevi.............................................................................................................. 97
3.7.2 Melanoma......................................................................................................................113
3.7.3 Metastases of melanoma................................................................................................. 121
4 Metaphoric dermatoscopic terms and what they mean............................................................................. 125
5 An algorithmic method for the diagnosis of pigmented lesions...................................................................147
5.1 One pattern............................................................................................................................ 147
5.1.1 Lines............................................................................................................................. 147
5.1.2 Pseudopods................................................................................................................... 156
5.1.3 Circles.......................................................................................................................... 156
5.1.4 Clods............................................................................................................................ 158
5.1.5 Dots.............................................................................................................................. 167
5.1.6 Structureless................................................................................................................... 169
5.2 More than one pattern.............................................................................................................. 171
5.2.1 Lines............................................................................................................................. 173
5.2.2 Pseudopods................................................................................................................... 184
5.2.3 Circles.......................................................................................................................... 185
5.2.4 Clods............................................................................................................................ 186
5.2.5 Dots.............................................................................................................................. 187
5.3 Applying pattern analysis to clinical practice............................................................................... 190
5.4 Chaos and Clues .................................................................................................................... 190
6 Non-pigmented (amelanotic) lesions......................................................................................................... 203
6.1 Clues used in the diagnosis of non-pigmented (amelanotic) lesions................................................. 203
6.2 Vascular patterns .....................................................................................................................211
6.3 Differential diagnosis of non-pigmented lesions............................................................................ 212
7 Clues and Clichés.................................................................................................................................... 235
7.1 Clues...................................................................................................................................... 235
7.2 Common Clichés ..................................................................................................................... 243
8 Special situations.................................................................................................................................... 253
8.1 Nails...................................................................................................................................... 253
8.2 Acral lesions............................................................................................................................ 260
8.3 The face................................................................................................................................. 269
8.4 Mucosal lesions....................................................................................................................... 280
8.5 Recurrent melanocytic lesions.................................................................................................... 280
8.6 Difficult lesions ........................................................................................................................ 280
8.7 Inflammatory skin diseases........................................................................................................ 286
9 Digital Dermatoscopic Monitoring............................................................................................................ 297
9.1 Choice of lesions to monitor...................................................................................................... 298
9.2 Interpretation of changes.......................................................................................................... 302
9.3 Growing nevus or melanoma?................................................................................................... 302
9.4 Benefits and Risks..................................................................................................................... 306
10 Cases..................................................................................................................................................... 309
11 Dermatoscopic-dermatopathologic correlation.......................................................................................... 373
Supplement............................................................................................................................................ 387
Index...................................................................................................................................................... 389
1 General Principles
Figure 1.2: Commonly used handheld dermatoscope of Heine Figure 1.4: Dermatoscope with polarized light, which dispenses
Company. When using this simple hand-held device one needs a with the need for a contact fluid or direct contact with the skin.
contact fluid such as paraffin oil or ultrasound gel.
lesion pigmentation and vascularity, as these features visible with polarized dermatoscopy (1.6, bottom row),
are located in deeper layers of the epidermis, and the while the white dots and clods of seborrheic keratosis
dermis. Light remitted from deeper structures is irregularly are best viewed with non-polarized dermatoscopy (1.6,
refracted by the unevenness of the superficial keratin top row) Polarized and non-polarized dermatoscopy
layer, further degrading the perceived image. These are therefore best considered complementary. Most
phenomena are largely eliminated by using a contact new handheld dermatoscopes can switch between
fluid (such as alcohol, paraffin oil or ultrasound gel) to polarized and non-polarized mode.
couple the baseplate of the dermatoscope to the skin.
(1.3 and 1.5). Replacing the air between the skin and
faceplate glass with a fluid smoothens the skin surface 1.2 Indication and Benefits of Dermatoscopy
and creates a far better match of refractive indices, In short, dermatoscopy is indicated when better resolu-
which greatly reduces reflection from the skin surface. tion of pigment or vascular structures in the epidermis or
More recently, dermatoscopes have been developed upper dermis will help resolve a differential diagnosis.
which eliminate surface reflection by the use of polar- Immediately after the introduction of handheld instru-
izing filters (1.4). These instruments do not require ments, dermatoscopy was promoted as being particu-
a contact fluid, or even direct contact with the skin. larly useful in differentiating nevi from melanomas by
Although the images seen using polarizing instruments assessment of pigment patterns. While this is important,
are very similar to those seen using contact dermatosco- the vast majority of cutaneous malignancies are not
py, a few significant differences exist (2). For example, pigmented. Furthermore, even skin neoplasms which
perpendicular white lines (“shiny white lines”) are only are most commonly pigmented have lightly pigmented
General Principles 11
A B
C D
Figure 1.5: Two pigment lesions: A and B represent a melanocytic nevus while C and D show a seborrheic keratosis. The pictures in the left
column (A, C) show what is seen with the naked eye while the right column (B, D) shows the image seen through the dermatoscope. In the
dermatoscopic image one finds additional structural details that escape detection by the naked eye. This enhancement of detail is partly
attributable to magnification, but more to the reduction of reflection on the surface of the skin.
or entirely non-pigmented variants. This includes a can confidently be diagnosed clinically, but this is a
significant minority of melanomas. misunderstanding as to the role of dermatoscopy. The
While patterns formed by blood vessels and keratin foremost role of dermatoscopy is not confirmation of
are less diagnostically specific than patterns formed by a diagnosis established clearly with the naked eye but
melanin pigment, they still provide significant additional the unveiling of morphological criteria that revise the
diagnostic information when pigment is absent, over diagnosis established with the naked eye. Dermatosco-
and above naked eye clinical examination (3, 4). py can shift the point of diagnosis closer to the initial
emergence of the neoplasm, but only if lesions with
Diagnosis of Melanoma no naked eye evidence of malignancy are routinely
Despite strong evidence to the contrary, the belief that examined.
dermatoscopy adds nothing to the diagnosis of mel- We consider it self-evident that every melanoma goes
anoma compared to naked eye examination persists through a stage in its evolution when it lacks the crite-
into the 21st century. ria required to allow diagnosis. This is the reason the
In a trivial sense, this is true in that dermatoscopy does clinical ABCD rule (1.7) contains a size criterion — not
not add anything to the diagnosis of melanomas which because melanomas are never less than 6 mm diameter,
12 General Principles
A B
C D
Figure 1.7: The concept of the clinical ABCD rule is illustrated by four melanomas. The ABCD criteria are applied when the melanoma has
achieved a certain size and has been present for a longer period of time (usually a few years). All of these melanomas are already invasive.
In other words, they are not confined to the epidermis (in situ), but have invaded the underlying dermis. The chances of cure are reduced in
proportion to the increasing depth of invasion.
but because the accuracy of clinical diagnosis is only epidermis. After excision of this melanoma, the patient
acceptable for larger lesions. Melanomas less than 6 mm may be deemed to be cured of the disease.
diameter are routinely diagnosable by dermatoscopy. Like every morphological method, dermatoscopy has
Indeed, dermatoscopic monitoring over time allows limitations. Dermatoscopy cannot entirely replace his-
diagnosis of melanomas even before the emergence topathology; in some cases histopathology is the only
of specific dermatoscopic features. way to establish an unequivocal diagnosis. Rarely,
dermatoscopy may be misleading; the naked eye criteria
Figure 1.8 shows a melanoma just a few millimeters point in the right direction and dermatoscopic criteria
in size, which shows no melanoma-specific criteria on erroneously point to a different diagnosis. However,
naked-eye inspection. It is neither asymmetrical nor these exceptions are only that, exceptions, and a large
has irregular margins, is not multicolored, and is not body of evidence demonstrates that the addition of
larger than 6 mm in size. However, dermatoscopic dermatoscopy improves overall diagnostic accuracy.
investigation shows that the criteria of a melanoma Histopathology is also a purely morphological method
are clearly fulfilled. The diagnosis was confirmed by with its own limitations. Correlation of histopathologic
histology showing an in situ melanoma (1.9). In other with dermatoscopic findings may allow a diagnosis
words, neoplastic melanocytes are confined to the even when histopathology alone is not diagnostic.
14 General Principles
Figure 1.8: A melanoma on the forearm, just a few millimeters in size. The condition may be clearly diagnosed as a melanoma on the basis
of dermatoscopy because of the presence of so-called pseudopods, whereas the application of the ABCD rule and naked-eye assessment
are both unreliable. The histological image clearly shows an in situ melanoma (Figure 1.9).
Table 1.1
First author and year of
Sample size (n) Sensitivity Specificity
publication
Unaided eye Dermatoscopy Unaided eye Dermatoscopy
Benelli 1999 401 67 % 80 % 79 % 89 %
Binder 1995 240 58 % 68 % 91 % 91 %
Binder 1997 100 73 % 73 % 70 % 78 %
Carli 1998 15 42 % 75 % 78 % 89 %
Cristofolini 1994 220 85 % 88 % 75 % 79 %
Dummer 1993 824 65 % 96 % 93 % 98 %
Krähn 1998 80 79 % 90 % 78 % 93 %
Lorentzen 1999 232 77 % 82 % 89 % 94 %
Nachbar 1994 172 84 % 93 % 84 % 91 %
Soyer 1995 159 94 % 94 % 82 % 82 %
Stanganelli 1998 20 55 % 73 % 79 % 73 %
Stanganelli 2000 3.329 67 % 93 % 99 % 100 %
Westerhoff 2000 100 63 % 76 % 54 % 58 %
the manner of presenting dermatoscopic images, and been speculated that the unfamiliar structures revealed
the subjects’ level of training, to name a few. Still, the by dermatoscopy only served to confuse clinicians who
majority of studies show the diagnostic accuracy of are trained in naked eye assessment. The trivial but
dermatoscopy to be higher than that of the naked-eye important conclusion drawn from this study was that
investigation. In 2002 and in 2008 the results of the dermatoscopy serves only those who know how to use
studies were confirmed by two meta-analyses (5, 6). In the procedure. A similarly structured study showed that a
2011 Rosendahl et al. confirmed that dermatoscopy also short and intensive phase of training – of just a few days’
improves the diagnostic accuracy for non-melanocytic duration – is sufficient to learn the basic principles of the
lesions (7). method and markedly improve diagnostic accuracy (10).
1.5 Development of the Method Table 1.2: List of dermatoscopic criteria established at the
It is instructive to follow the evolution of dermatoscopy consensus conference in Hamburg in 1989
as a tool for the assessment of pigmented skin lesions Pigment network
– on the one hand to understand the origins of common discrete
methods, and on the other hand to comprehend and
prominent
classify the diverse terms in use (the ad hoc prolifera-
regular
tion of terms is a major source of confusion). Pioneers
in the field of dermatoscopy such as Saphier largely irregular
confined themselves to the description of inflammatory wide
skin lesions like lichen planus, lupus erythematosus, or narrow
scabies. At this time dermatoscopy was apparently broad
of no importance for the diagnosis of pigmented skin delicate
lesions or melanoma. The first serious report about the
Irregular extensions, pseudopods
value of dermatoscopy for the diagnosis of melanoma
Radial streaming
was published by Rona MacKie in 1971 (12). Ten years
later the Austrians Fritsch and Pechlaner published Brown globules
“Differentiation of benign from malignant melanocytic Black dots
lesions using incident light microscopy”. In addition to Whitish veil, milky way
other criteria, the authors describe in detail the basic White scar-like depigmented areas
anatomical features of the pigment network, which is Grayish-blue areas
one of the principal structures in dermatoscopy (13). This
Hypopigmentation
report mentions dermatoscopic differences between nevi
Reticular depigmentation
and melanomas, but a general method for the diagnosis
of pigmented skin lesions is just briefly outlined. Milia-like cysts
Comedo-like openings
1.5.1 Pattern Analysis Telangiectasia
In 1987 Pehamberger, Steiner, and Wolff described Reddish-blue areas
pattern analysis, the first analytical method to distinguish Maple leaf-like areas
between the primary types of pigmented skin lesions (at
the time, the rather cumbersome term epiluminescence
microscopy was used instead of dermatoscopy) (14, 15).
Pattern analysis is based on recognition of a number of regular/irregular, or delicate/prominent, and narrow/
dermatoscopic structures which constitute reproducible broad. Unfortunately (though inevitably) these poorly
patterns characteristic of the more common pigmented defined qualitative properties were subject to a wide
lesions. As the first studies on pattern analysis were range of inter-individual differences in interpretation,
published in English-language journals, the Austrians and were poorly reproducible. Despite justified criti-
Pehamberger, Steiner, and Wolff used only English cisms, however, the studies of Pehamberger, Steiner,
terms for the structures they described, such as radial and Wolff were the first systematic approaches in this
streaming, blue-whitish veil or the milky way. These field and the starting point for further developments
neologisms were poorly defined or not defined at all. that followed in subsequent years.
This artificial metaphoric language created a barrier Shortly afterwards, other research groups in Europe
even to those willing to learn. Furthermore, the diag- also showed interest in dermatoscopy, which soon led
nosis was based not only on the presence or absence to a variety of approaches. The consequences were an
of a dermatoscopic structure, but also on qualitative uncontrolled growth of terms on the one hand, and the
aspects. For instance, the German term “Schollen” absence of consensus about fundamental aspects on
(clods) which was given the English designation of the other. The first attempt to counteract this evolution
“globules” was assessed according to whether they and standardize dermatoscopy was made as early as
were distributed regularly or irregularly, and whether in 1989 at a consensus conference in Hamburg (16).
they were of the same size or different sizes. Qualitative The results of this consensus conference were published
aspects of the pigment network described a few years in 1990. The participants established a list of diagnostic
earlier by Fritsch and Pechlaner included, according to criteria that is shown in table 1.2. One outcome of
Pehamberger, Steiner, and Wolff, paired terms such as the consensus conference was speculation about the
General Principles 17
histopathological correlates of dermatoscopic criteria, In the dermatoscopic ABCD rule, scores are assigned
but there was no attempt to define the listed criteria. to the four criteria of asymmetry, border, color and
Today this list is mainly of historical value. dermatoscopic structures, each of which are multiplied
by a fixed factor (the latter is determined by the use of
1.5.2 Evolution of a diagnostic algorithm statistical methods and a large random sample). Der-
The pattern analysis published by Pehamberger, Steiner matoscopic structures scored in the method of Stolz are
and Wolff in 1987 was mainly confined to a descrip- pigment network, dots, clods (“globules”), “branched
tion of the frequencies of dermatoscopic structures for streaks”, and structureless areas. These 4 scores are
the most important pigmented skin lesions. A formal summed to determine a total dermatoscopy score.
method that can be used for melanocytic as well as This score categorizes the lesion as either benign,
non-melanocytic skin lesions and which guides the suspicious or malignant. The ABCD rule only applies
investigator in a structured manner to a specific diag- to melanocytic lesions.
nosis was not provided. This was developed in the
following years. In this regard, the studies of Jürgen Argenziano’s 7-point check-list
Kreusch (17) and Wilhelm Stolz (18) are worthy of When using Argenziano’s 7-point check-list lesions are
mention. They proposed a 2-step algorithm. The first step assessed for the presence of seven criteria; 3 major
classified pigmented skin lesions as either melanocytic which score 2 each, and 4 minor which score 1 each.
or non-melanocytic, and specifically diagnosed several A total score of three or more is indicative of melanoma.
common non-melanocytic tumors. The second step was The major criteria are an atypical pigment network,
applied to melanocytic lesions only, with the goal of a blue-whitish veil, and an atypical vascular pattern.
distinguishing melanoma from melanocytic nevi. This The minor criteria are irregular streaks, irregular dots/
method of investigation gained acceptance, although globules, irregular blotches, and regression structures.
in slightly modified form and despite a few weaknesses Like the ABCD rule, the 7-point check-list is only suitable
(which will be addressed later). for melanocytic lesions.
Figure 1.10: This pigmented lesion can be clearly diagnosed as a melanoma with any dermatoscopic method.
Stolz’s dermatoscopic ABCD rule: A (asymmetrical in both axes; 2.6 points), B (sharp interruption of pigment in four segments; 0.4 points),
C (4 different colors: light brown, dark brown, blue-gray, black; 2 points), D (4 different dermatoscopic structures: reticular lines, dots,
clods, and a structureless area; 2 points) – yield 7 points in all and thus confirm the diagnosis of melanoma (if the total sum is > 4.75 points,
the diagnosis is melanoma).
Argenziano’s 7-point check-list: Two major criteria (asymmetry and blue-whitish veil) and a minor criterion (irregular dots/globules) yield
5 points and thus confirm the presence of a melanoma (a melanoma is presumed to exist from a score of 3 points onward).
Menzies’ method: Asymmetry and more than one color and the simultaneous presence of positive criteria, such as peripheral black dots/
globules or a blue-whitish veil lead to the diagnosis of melanoma.
Chaos and clues: A chaotic lesion with multiple clues to malignancy (gray/blue structures, eccentric structureless zone, peripheral black
dots) should be excised to exclude malignancy.
Pattern analysis: A clearly asymmetrical pattern (reticular lines, dots, structureless area), more than one color with melanin being predomi-
nant (brown, blue, black), also arranged asymmetrically, and several specific criteria confirming the presence of a melanoma (black dots in
the periphery and a structureless eccentric blue area) clearly indicate the presence of melanoma.
of progression patterns of facial melanoma by Stolz of Dermatoscopy in 2002 in Rome. The results of the
(27), the description of the patterns of Clark nevi by consensus conference were summarized in a consensus
Hofmann-Wellenhof (28), the evaluation of dermatoscop- paper which was presented a year later in the Journal
ic criteria of pigmented seborrheic keratosis by Braun of the American Academy of Dermatology (JAAD) (36).
(29), the categorization of the protean dermatoscopic The consensus included the first step of the diagnostic
patterns of dermatofibroma by Zaballos (30), the clas- algorithm, namely the distinction between melanocytic
sification of patterns of acral melanocytic lesions by and non-melanocytic lesions, as well as the previously
Saida and Tanaka (31, 32), the analysis of pigmented mentioned scoring systems for melanocytic lesions and
mucosal lesions and recurrent nevi and melanoma the definitions of the most commonly used terms in
by Blum (33), the discovery of dermatoscopic clues dermatoscopy (1.11 to 1.13). Regrettably, this unique
for pigmented Bowen’s disease by Cameron (34), opportunity to simplify the language of dermatoscopy
and finally the dermatoscopic classification of nevi in was missed. Instead, metaphoric terms were adhered
different age groups by Zalaudek (35). All the above to and incomplete or contradictory definitions were
mentioned achievements have been accomplished by formulated.
the application of pattern analysis. After the results of the second consensus were published
In fact, it has been shown that pattern analysis is superior in 2003 the vocabulary of dermatoscopy expanded
to other investigation techniques in many respects. Begin- significantly. Even experts struggled with the multitude
ners find it easier to cope with the simple algorithms, of terms. The main driving forces for the creation of
but they are soon confronted with barriers which can new terms were the expansion of dermatoscopy to
be resolved only by a comprehensive method such as new realms such as inflammatory skin diseases and
pattern analysis. the introduction and dissemination of polarized der-
What is one talking about when one refers to pat- matoscopes that allowed observations of structures
tern analysis? In actual fact, it is still not clear what previously invisible with classic contact dermatoscopy.
a person does when he/she uses pattern analysis. Many new terms, especially those that were published
Due to the large number of criteria to be considered, in case reports, were ill-defined metaphors with dubious
pattern analysis is more difficult and demanding than diagnostic significance.
simple scoring systems, but it is also more powerful In 2007 Harald Kittler introduced a simple descriptive
and flexible. How the investigator combines these terminology that avoids metaphoric terms and is based
criteria to reach a diagnosis remained a mystery for on five geometrically defined basic elements, namely
a long time because the rules of this skill were never lines, pseudopods, circles, clods and dots (37). The
clearly formulated. Thus, pattern analysis appeared advantages of this terminology are its simplicity, its
to be mysteriously dependent on the user’s ingenuity. logical structure, and the lack of need for definitions
Teaching this technique was a somewhat mystifying beyond those of basic elements. In the following years
subject. The greatest challenge of this book is to render the descriptive terminology became increasingly popu-
pattern analysis – this powerful methodological tool – lar. The growing controversy between descriptive and
communicable and comprehensible. metaphoric terminology and the growing number of
new terms demanded the need for a new consensus.
1.5.5 Standardization and Consensus In 2013 Alan Halpern initiated the International Skin
After the previously mentioned first consensus confer- Imaging collaboration (ISIC) and appointed Harald
ence held in 1989 in Hamburg, nothing happened for Kittler to lead a selected group of experts charged
a long time. Dermatoscopy remained split into various with creating a standardized dictionary of dermatos-
schools. A uniform method, homogeneous criteria, copy. This process led to the 3rd consensus conference
and congruent definitions were absent. It was not until which was finalized during the 4th World Congress of
the founding of the International Dermoscopy Society Dermatoscopy in Vienna in April 2015. After two years
(IDS) in 2001, under co-founder and first president of extensive discussions the expert group succeeded in
Peter Soyer, a dermatologist from Graz, that a forum creating a dictionary of standardized terms that takes
was formed. This forum declared that it was responsible into account descriptive and metaphoric terminology.
for answering questions relating to the consensus. As it This dictionary, which was published along with the
combined all important research groups, it appeared consensus paper in 2016 (38), is now the standard
to be legitimized to perform the task. This development reference for all issues related to terminology. We
culminated in a consensus conference held via the Inter- will deal with the dictionary in detail in chapter 4. For
net and the organization of the First World Congress reasons that we will explain below the authors of this
20 General Principles
Metaphoric terms
Rather like the names of the constellations of the night
sky, many dermatoscopic terms require considerable
imagination before they can be related to the mor-
phological structures they are supposed to describe.
These include spoke-wheel-areas, blue-whitish veil,
radial streaming, fat fingers, or moth-eaten border, to
name just a few (also see figures 1.11 and 1.12). In
the collective memory of the dermatoscopic communi-
ty, most of these terms are linked to the inventor and
are certified as such. This, possibly, is the reason why
new terms are being constantly created. The strength
of this vivid and gripping terminology undoubtedly
lies in its ability to stimulate associative thinking and
therefore memory. However, this advantage is offset
by the fact that most of the terms are the outcome of
individual associations by their inventors. Only in ideal
cases does any real similarity exist between the terms
and actual structures they are intended to represent.
Figures 1.11 and 1.12: Modified original tables with the criteria
and definitions worked out at the consensus conference. From:
Argenziano G, Soyer HP, Chimenti S, et al. Dermoscopy of pig-
mented skin lesions: results of a consensus meeting via the Inter-
net. J Am Acad Dermatol 2003; 48: 679–93. The inconsistent
definitions of aggregated globules, blue-gray globules and dots
and globules are highlighted.
General Principles 21
A B
C D
E F
Figure 1.14: Small brown clods of the same size and shape (A), blue clods of different sizes and dissimilar form (B), red clods (C), large
polygonal skin-colored and light-brown clods (D), yellow and orange clods (E), violet and black clods (F). Translated into the language of
classical dermatoscopy this means: globules (A), blue ovoid nests (B), red lacunae (C), cobblestone pattern (D), orange and yellow clods are
not criteria in classical dermatoscopy and therefore bear no names although they are very specific for seborrheic keratoses (E), purple and
black clods are also termed lacunae (F). Thus, the same structures, namely clods, are named differently, depending on their color.
General Principles 23
A B
Figure 1.15: Non-melanocytic lesions with a pigmented network. (A) Clinical appearance; (B) Appearance on dermatoscopy: At the
periphery one finds obvious reticular lines (pigment network). Histopathological diagnosis: seborrheic keratosis.
on consistent definitions for all suitable metaphoric terms crucial life decisions (“Slow and steady wins the race”).
and reached a consensus for definitions of vascular These proverbs may be justified as rules of thumb but
patterns. We will deal with the consensus definitions are too rigid, simplified and limited to guide all (or
of metaphoric terms of pigmented lesions in chapter even most) of our actions.
4. The vascular patterns and their definitions will be With the exception of the chaos and clues algorithm
addressed in chapter 6. all scoring systems are restricted to distinguishing nevi
from melanomas, and so can only be applied after
Scoring systems one has decided whether the lesion is melanocytic or
The basic intention behind the so-called scoring systems non-melanocytic (“the first step”). Furthermore, scoring
was to make dermatoscopy more accessible to begin- systems are of limited use at some specific locations
ners. Like a raffle, one threw all dermatoscopy criteria such as the face, mucosa, or the palms and soles, where
ever described into a large drum, mixed them thoroughly, different criteria apply.
and selected, on the basis of statistical procedures, a Having said this, and despite any criticism one may level
few that were considered relevant to the diagnosis. All against scoring systems, one must acknowledge that they
others were ignored. Then one invented complicated do perform reasonably well under many circumstances.
procedures (every author invented a different one) as to Beginners will be able to achieve a moderate degree
how one distinguishes between melanomas and nevi on of diagnostic accuracy fairly rapidly by using a scoring
the basis of the chosen criteria. For all scoring systems, system. In this sense, the different scoring systems have
the final diagnosis is always “benign” or “malignant”, been thoroughly tested, and all found to be useful. We
as if the world of melanocytic lesions could be reduced also acknowledge that not everybody has the time to
to these two terms. The scoring systems trivialized learn a comprehensive method like pattern analysis. For
dermatoscopy without simplifying it. those who want to achieve quick and fairly accurate
Scoring systems have a strong framework (usually one results we recommend the chaos and clues algorithm
sums up a few parameters and when the total sum (see chapter 5) because it is fast and simple and cir-
exceeds a certain value it is a melanoma). The user is cumvents the cumbersome and problematic first step.
degraded to a robot.
A method of this type does not encourage reflection. The first step
Without reflection, one does not achieve a deeper Established dogma is that one must distinguish between
understanding of a subject. If one wishes to achieve melanocytic and non-melanocytic lesions as the first
a more profound comprehension of a morphological step in assessing a pigmented lesion (39). In principle
method like dermatoscopy, it will not be sufficient to there is nothing wrong with this approach as it provides
add one and one, or say B after one has said A. That structure to the procedure of investigation. However, it
would be tantamount to only using proverbs to guide is not self-evident that this is the best possible first step,
24 General Principles
A B
Figure 1.16: Non-melanocytic lesions with a pigment network. (A) Clinical appearance; (B) Appearance on dermatoscopy: dark brown and
black reticular lines (pigment network). Histopathological diagnosis: “Ink-spot” Lentigo.
A B
14 Pehamberger H, Steiner A, Wolff K. In vivo epilumines- 28 Hofmann-Wellenhof R, Blum A, Wolf IH, Piccolo D, Kerl
cence microscopy of pigmented skin lesions. I. Pattern H, Garbe C et al. Dermoscopic classification of atypical
analysis of pigmented skin lesions. J Am Acad Dermatol melanocytic nevi (Clark nevi). Arch Dermatol 2001;
1987; 17: 571–83. 137: 1575–80.
15 Steiner A, Pehamberger H, Wolff K. In vivo epilumines- 29 Braun RP, Rabinovitz HS, Krischer J, Kreusch J, Oliviero
cence microscopy of pigmented skin lesions. II. Diagnosis M, Naldi L et al. Dermoscopy of pigmented seborrheic
of small pigmented skin lesions and early detection keratosis: a morphological study. Arch Dermatol 2002;
of malignant melanoma. J Am Acad Dermatol 1987; 138: 1556–60.
17: 584–91. 30 Zaballos P, Puig S, Llambrich A, Malvehy J. Dermoscopy
16 Bahmer FA, Fritsch P, Kreusch J, Pehamberger H, Rohrer of dermatofibromas: a prospective morphological study
C, Schindera I et al. [Diagnostic criteria in epilumines- of 412 cases. Arch Dermatol 2008; 144: 75–83.
cence microscopy. Consensus meeting of the professional 31 Saida T, Oguchi S, Ishihara Y. In vivo observation of
committee of analytic morphology of the Society of Der- magnified features of pigmented lesions on volar skin
matologic Research, 17 November 1989 in Hamburg]. using video macroscope. Usefulness of epiluminescence
Hautarzt 1990; 41: 513–4. techniques in clinical diagnosis. Arch Dermatol 1995;
17 Kreusch J, Rassner G. Auflichtmikroskopie pigmentierter 131: 298–304.
Hauttumoren. Stuttgart: Thieme; 1991. 32 Saida T, Koga H, Yamazaki Y, Tanaka M. Acral Mel-
18 Stolz W, Braun-Falco O, Bilek P. Color Atlas of Derma- anoma. In: H. P. Soyer editor. Color Atlas of Melano-
toscopy: Blackwell Wissenschafts-Verlag; 2002. cytic Lesions of the Skin. New York: Springer; 2007.
19 Nachbar F, Stolz W, Merkle T, Cognetta AB, Vogt T, p. 196–203.
Landthaler M et al. The ABCD rule of dermatoscopy. High 33 Blum A, Simionescu O, Argenziano G, Braun R, Cabo
prospective value in the diagnosis of doubtful melanocytic H, Eichhorn A et al. Dermoscopy of pigmented lesions
skin lesions. J Am Acad Dermatol 1994; 30: 551–9. of the mucosa and the mucocutaneous junction: results
20 Argenziano G, Fabbrocini G, Carli P, De Giorgi V, of a multicenter study by the International Dermoscopy
Sammarco E, Delfino M. Epiluminescence microscopy Society (IDS). Arch Dermatol 2011; 147: 1181–7.
for the diagnosis of doubtful melanocytic skin lesions. 34 Cameron A, Rosendahl C, Tschandl P, Riedl E, Kittler
Comparison of the ABCD rule of dermatoscopy and a H. Dermatoscopy of pigmented Bowen’s disease. J Am
new 7-point checklist based on pattern analysis. Arch Acad Dermatol 2010; 62: 597–604.
Dermatol 1998; 134: 1563–70. 35 Zalaudek I, Schmid K, Marghoob AA, Scope A, Manzo
21 Menzies SW, Crotty KA, Ingvar C, McCarthy W. Der- M, Moscarella E et al. Frequency of dermoscopic nevus
moscopy: An Atlas: Mcgraw-Hill Education Ltd; 2009. subtypes by age and body site: a cross-sectional study.
22 Soyer HP, Argenziano G, Zalaudek I, Corona R, Sera F, Arch Dermatol 2011; 147: 663–70.
Talamini R et al. Three-point checklist of dermoscopy. A 36 Argenziano G, Soyer HP, Chimenti S, Talamini R, Corona
new screening method for early detection of melanoma. R, Sera F et al. Dermoscopy of pigmented skin lesions:
Dermatology 2004; 208: 27–31. results of a consensus meeting via the Internet. J Am
23 Henning JS, Dusza SW, Wang SQ, Marghoob AA, Rab- Acad Dermatol 2003; 48: 679–93.
inovitz HS, Polsky D et al. The CASH (color, architecture, 37 Kittler H. Introduction of a new algorithmic method
symmetry, and homogeneity) algorithm for dermoscopy. based on pattern analysis for diagnosis of pigmented
J Am Acad Dermatol 2007; 56: 45–52. skin lesions. Dermatopathol: Pract & Conc 2007; 13: 3.
24 Rosendahl C, Cameron A, McColl I, Wilkinson D. Der- 38 Kittler H, Marghoob AA, Argenziano G, Carrera G,
matoscopy in routine practice – ‘chaos and clues’. Aust Curiel-Lewandrowski C, Hofmann-Wellenhof R et al.
Fam Physician 2012; 41: 482–7. Standardization of Terminology In Dermoscopy/Der-
25 Marghoob AA, Braun R, Kopf AW. An Atlas of Dermos- matoscopy: Results of the 3rd Consensus Conference
copy: Informa Healthcare; 2004. of the International Society of Dermoscopy. J Am Acad
26 Menzies SW, Westerhoff K, Rabinovitz H, Kopf AW, Dermatol 2016; 74: 1093–106.
McCarthy WH, Katz B. Surface microscopy of pigmented 39 Marghoob AA, Braun R. Proposal for a revised 2-step
basal cell carcinoma. Arch Dermatol 2000; 136: 1012–6. algorithm for the classification of lesions of the skin using
27 Stolz W, Schiffner R, Burgdorf WH. Dermatoscopy for dermoscopy. Arch Dermatol 2010;146:426–8.
facial pigmented skin lesions. Clin Dermatol 2002; 40 Tschandl P, Rosendahl C, Kittler H. Accuracy of the first
20: 276–8. step of the dermatoscopic 2-step algorithm for pigmented
skin lesions. Dermatol Pract Concept 2012; 2: 203a08.
In this chapter we present, in the form of a glossary, a lesions are often lumped together and classified as
list of pigmented skin lesions for which dermatoscopic “non-melanocytic” but this is problematic because it
assessment is likely to be helpful. We do this to clarify includes diverse conditions which are entirely unrelated,
terminology and to prevent misunderstanding. The for example dermatofibroma and seborrheic keratosis.
usefulness of dermatoscopy is not limited to assessing We prefer to classify conditions by what they are and
the lesions mentioned in this chapter, which is restricted not by what they are not and so whenever possible
to conditions that are usually or commonly pigmented will avoid the term “non-melanocytic” by using more
(never forgetting that all pigmented neoplasms also specific terms.
have non-pigmented variants). The universe of lesions
which are (nearly) always non-pigmented, which also 2.1.1 Melanocytic nevi
includes inflammatory conditions, is more complex and A melanocytic nevus is a benign proliferation of mela-
very large. A lexicon that includes all conditions that nocytes, either in the form of a congenital malformation
can usefully be examined by dermatoscopy is beyond (hamartoma) or an acquired neoplasia. Melanocytes
the scope of this chapter. We strongly recommend that are not merely increased in number but are also (at least
you read this chapter even though it does not directly partly) arranged in nests, chords or strands. An increa-
address the subject of dermatoscopy. The photographs sed number of melanocytes (melanocytic hyperplasia)
shown here are all clinical images of what would be without the formation of nests, chords or strands is not
seen when viewing these lesions with the naked eye. sufficient to justify calling a lesion a nevus. One may also
One purpose of doing this is to show how difficult it observe an increase in the number of melanocytes in
can be to make a diagnosis without dermatoscopy. “non-melanocytic” lesions, for example solar lentigines,
Dermatoscopic views are shown in chapter 3 after but nests never develop in these cases.
presentation of the method. In this book we use a modified version of Ackerman’s
Clinicians, “dermatoscopists” and pathologists speak (1) classification for melanocytic nevi, which is based
languages which partly overlap, but have significant on histopathological criteria. The disadvantage of this
areas of difference. This is particularly confusing when classification is that only the pathologist can see the
the same terms are used, but with different meanings. features required to make a definitive diagnosis; clini-
Adding to the confusion, several competing schools cians and “dermatoscopists” merely try, on the basis of
of dermatopathology propagate different concepts their description, to predict the pathologist’s diagnosis.
and use different terminology. As a clinician one must Ideally, a classification system would incorporate clinical
therefore expect diverse histopathological reports. The and dermatoscopic as well as histopathological findings.
definitions that now follow may not resolve the widely Several grave historical errors such as the interpreta-
prevalent confusion, but they will at least not intensify tion of the Spitz nevus as a “juvenile melanoma” are
it. Needless to say, we use a consistent form of termi- attributable to the fact that pathologists did not look
nology in this book. beyond the objective of their microscope. Clinicians
and “dermatoscopists” who are not familiar with der-
matopathology are also at risk of this type of error.
2.1 Melanocytic lesions In Ackerman’s modified classification of nevi, a distinction
Dermatoscopy is commonly used to confirm or exclude is made between the following entities: Clark nevus,
the diagnosis of melanoma. Melanocytic nevi and Spitz nevus, Reed nevus, congenital nevus (with the
melanomas — the benign and malignant neoplasms sub-types “superficial” and “superficial and deep”),
arising from melanocytes — together form the group combined congenital nevus, blue nevus (with subtypes),
of lesions classified as “melanocytic”. All other skin Unna nevus and Miescher nevus. It should be noted that
28 Principal pigmented skin lesions relevant to dermatoscop
the term “congenital” does not necessarily mean that the In a rare example of unanimity, both clinicians and
nevus was visible at birth. Many so-called congenital dermatopathologists use “acral nevus” as a general
nevi appear after birth and some even appear as late term for nevi at acral sites; the location determines the
as early adulthood (2, 3). In addition, we use the term name. In fact, most types of nevi, such as Spitz nevi,
“acral nevus” acknowledging that the designation is Reed nevi and “superficial” or “superficial and deep”
not consistent; location is not a histopathologic feature congenital nevi, may occur on acral skin. Whenever
and is otherwise irrelevant to the classification of nevi. possible the specific diagnosis should be preferred to
Contrary to common practice, the terms “dysplastic the general term “acral nevus”. We name the special
nevus” and “atypical nevus” are not used here. type of nevus that has no equivalent at other locations
and occurs only on acral skin a “classical acral nevus”.
The terms used in this book and their definitions are This type of nevus is small and flat, and its pigmentation
given below in alphabetical order: is uniformly brown. In terms of histopathology it consists
of small nests of melanocytes exclusively at the dermo-
Acral nevus epidermal junction. In this book, while we have no term
Acral skin is that found on the palms and soles. Anatomi- to replace the ambiguous “acral nevus”, the specific
cally the acral skin is marked by its specific arrangement histopathological diagnosis is additionally given when
of rete ridges, which are represented on the surface of possible. Only when no histology is available and the
the skin by characteristic papillary ridges and furrows. clinical or dermatoscopic diagnosis is equivocal do
Principal pigmented skin lesions relevant to dermatoscop 29
we use the collective term “acral nevus” with no further subjected to histological investigation (which occurred in
specification (2.1). most cases), the dermatopathological sub-classification
is also given.
Blue nevus Not all “blue nevi” are blue on clinical examination
This is a collective term for several types of nevi whose or dermatoscopy. Some may be gray or skin-colored
principal common pathological criterion is the prolife- while others may be partly brown. There are also non-
ration of spindle-shaped and dendritic melanocytes pigmented varieties which are white. Conversely, of
in the dermis. As a rule these dermal melanocytes course, not all blue melanocytic lesions are “blue nevi”.
contain abundant melanin. This gives rise to the charac- The term “malignant blue nevus” is not used in this book.
teristic eponymous blue color on clinical (2.2) as well The entity known as a malignant blue nevus is, in our
as dermatoscopic examination. Dermatopathologists opinion, a melanoma (4).
distinguish between several sub-types of blue nevi.
The most important of these are the so-called common Clark nevus
blue nevus and the cellular blue nevus. Because these The Clark nevus is the most common type of acqui-
dermatopathological sub-groups cannot be identified red melanocytic neoplasm. They are usually macular,
clinically or by dermatoscopy, when we only show the ranging in size from a few millimeters up to about
clinical or dermatoscopic appearance, we use the coll- 1 cm. Colour is usually brown and may be variegate.
ective term “blue nevus”. If the nevus was excised and Occasionally, but more commonly in persons with a
30 Principal pigmented skin lesions relevant to dermatoscop
very light skin, one finds lightly pigmented or non- Depending on their size, these nevi are sub-divided into
pigmented varieties. Clark nevi usually occur on the large (> 20 cm), medium-sized (1.5–19.9 cm) and small
trunk and the proximal portion of the extremities, but (< 1.5 cm) congenital nevi (2.4). At the end of their
not on facial or acral skin. Clark nevi are very common. period of growth, congenital nevi are usually raised
On average, people with lighter skin phototypes have above the skin and may be heavily or lightly pigmented.
ten to twenty, but it is not unusual to see people with Sometimes, but not always, they have terminal hair.
hundreds of Clark nevi. The majority of “small congenital nevi” appear after
In terms of dermatopathology Clark nevi have a charac- birth, i.e. usually during childhood and puberty. Their
teristic appearance: the silhouette is symmetrical and flat, occurrence appears to be independent of exposure to
and the melanocytes are located in small, regular nests ultraviolet rays.
at the dermo-epidermal junction (junctional Clark nevus). The size of a congenital nevus is proportional to the
Occasionally one finds small nests of melanocytes in likelihood of the nevus being visible at birth. When
the papillary dermis as well (compound type of Clark dermatopathologists refer to a congenital nevus they
nevus). In contrast to the “superficial” congenital nevus, are usually talking about a nevus with a specific type
these nests do not entirely fill the papillary dermis. A of fine tissue architecture (arrangement and distribution
purely dermal Clark nevus does not exist. of melanocytes), which obviously remains invisible to
This nevus is named after American pathologist Wallace the clinician. Whether the nevus was present at birth
H. Clark (5). He considered this nevus an intermediate or not, as well as its size, are of little importance to
step in the development of melanoma and therefore the dermatopathologist. The two types of congenital
called them “dysplastic nevi”. This concept is no longer nevus are the “superficial” congenital nevus (Ackerman
tenable. Unfortunately Clark incorporated several dif- nevus) in which the accumulation of melanocytes is no
ferent types of nevi in the term, with many of the nevi deeper than the papillary dermis, and the “superficial
termed “dysplastic” by Clark actually being “superficial” and deep” congenital nevus (Zitelli nevus) in which
or “superficial and deep” congenital nevi (2.3). Many melanocytes extend at least into the reticular dermis.
histopathologists continue to use the term “dysplastic” Both types of congenital nevi are common – possibly
and continue to include various small congenital nevi as common as Clark nevi.
under this name. Clark nevi may mimic melanoma and In dermatoscopy one is mainly interested in small con-
therefore occasionally require excision for diagnostic genital nevi. Many of the nevi termed “dysplastic” or
reasons. Prophylactic excision is not indicated because “atypical” in patients with so-called “dysplastic nevus
the risk of malignant transformation of a single Clark syndrome” are actually small congenital nevi (2.5).
nevus is extremely low. The majority of melanomas arise These patients are subject to a higher risk of melanoma
de novo; they do not arise from a pre-existing nevus. because the number of congenital nevi is very likely an
expression of a genetic predisposition. Many individuals
Congenital nevus have both small congenital nevi and Clark nevi (2.6).
In the absence of further specification this is an ambi- As mentioned earlier, many dermatologists and derma-
guous collective term used to mean different things by topathologists adhere to a different concept and refer
clinicians and pathologists. Clinicians refer to mela- to Clark nevi as well as “superficial” and “superficial
nocytic nevi as congenital only when they are visible and deep” congenital nevi as “dysplastic nevi”. This
during or shortly after birth or when the size of the creates the wrong impression that one is referring to
nevus does not permit any other differential diagnosis. the same type of nevus.
32 Principal pigmented skin lesions relevant to dermatoscop
Figure 2.7 demonstrates the difficulties in distinguishing dermis (therefore deep). The arrow in Figure G points
between Clark nevi and small congenital nevi when to a melanocyte nest in the reticular dermis. The arrows
one relies on the clinical appearance alone. In the left in Figure H point to nests in the papillary dermis. In a
column a Clark nevus is shown from a distance (A) and Clark nevus as we define it, nests of melanocytes are
in detail (B). Adjacent to it (E, F) are corresponding found no deeper than the papillary dermis.
photographs of a “superficial and deep” congenital
nevus (Zitelli nevus). A distinction based on clinical Combined congenital nevus
criteria appears impossible. However, the difference When clinicians and dermatopathologists refer to a
is easily identified on dermatopathology. In Clark nevi combined nevus they mean different things. For the
(C, D) there are small nests of melanocytes at the der- clinician a combined nevus is a nevus that is pigmented
mo-epidermal junction (arrows in image D) while the in the junctional as well as the dermal portion. The
papillary dermis is largely unaffected (the cells in the junctional portion is brown and the dermal portion blue.
papillary dermis are inflammatory cells and melano- (Melanocytes in the reticular dermis, filled with melanin,
phages). The “superficial and deep” congenital nevus appear blue on the surface of the skin). In this case the
(G, H) has a different histopathological appearance. term “combined” refers to the simultaneous occurrence
The melanocyte nests are large and are present in the of brown and blue. When a dermatopathologist (who,
papillary (therefore superficial) as well as the reticular as a rule, is unaware of the clinical appearance) uses
Principal pigmented skin lesions relevant to dermatoscop 33
Figure 2.6: Patients with multiple nevi, who in colloquial language are referred to as patients with “dysplastic nevus syndrome”. These indi-
viduals not only have multiple nevi but also multiple types of nevi, namely Clark nevi, and “superficial” – as well as “superficial and deep”
congenital nevi.
the term “combined nevus” they are usually referring cytes with melanin on the one hand; and small, round,
to a nevus composed of two or more cell populations. junctional melanocytes on the other. Combined nevi,
The term “combined” refers in this case to the cytology regardless of whether they are viewed from the clinical
which, in turn, is not known to the clinician. In some or the pathological perspective, are usually congenital.
cases the two viewpoints overlap, for instance when a However, this does not always mean that they were
nevus is composed of dermal spindle-shaped melano- visible at birth (2.8).
34 Principal pigmented skin lesions relevant to dermatoscop
A E
B F
C G
D H
“High-grade dysplastic nevi” are at no higher risk of compound nevus. When melanocytes are exclusively
developing into melanoma than any other nevi. Rather, located in the dermis, the nevus is known as a dermal
a lesion diagnosed as a “high-grade dysplastic” nevus nevus. One may refer to a junctional Clark nevus or a
is more likely to actually be a melanoma. Occasionally Clark nevus of the compound type, or make a distinction
the term “dysplastic” is used in the inflationary sense, between a junctional Spitz nevus and a dermal Spitz
i.e. all or nearly all excised flat nevi are termed “dys- nevus, but the terms junctional compound and dermal
plastic”. Used in this sense, the term loses any meaning without specifying the type of nevus is as unspecific
it may have ever had. as the term “lentigo” and therefore they are not used
in this book.
Junctional, Compound and Dermal nevus
These terms only refer to the location of the melanocytes. Other designations of melanocytic nevi not used in
They say nothing about the type of nevus. When mela- this book
nocyte nests are exclusively located in the epidermis this Textbooks of dermatology and dermatopathology are
is termed a junctional nevus. With melanocytes in the full of different terms to describe nevi. An exhaustive
epidermis and the dermis, the nevus is referred to as a list of all terms that have ever been used would make a
40 Principal pigmented skin lesions relevant to dermatoscop
book of its own, but space is not the only reason why we
do not list them all here. Some entities are controversial
and ill-defined. Some are characterized pathologically
but do not have a specific dermatoscopic appearance.
For example, pathologists speak of “deep penetrating
nevus” and “benign melanocytoma” but clinicians
practically never do, because only pathologists are
able to observe the specific features required to speci-
fically diagnose these types of nevi. When examined
with the unaided eye or with dermatoscopy both nevi
usually look like blue nevi. On the other hand there
are quite a number of nevi that have been defined by
clinicians and dermatoscopists but do not show a spe-
cific pathology. Examples would be “eclipse nevus”, a
term that has been used for a type of congenital nevus, Figure 2.17: Meyerson nevus.
and “hypermelanotic nevus”, a small Clark nevus with A Meyerson nevus is not a specific type of nevus. It is a nevus with
a spongiotic (“eczematous”) reaction. In this case, the inflamed
a hyperpigmented center (11). Other names are used
nevus is most likely a “superficial and deep” congenital nevus.
to describe very specific circumstances, for example Ectatic blood vessels, redness and scales at the periphery are
a Meyerson nevus (12) is a congenital nevus or, less signs of inflammation.
frequently, a Clark nevus, with a spongiotic (“eczema- invasive melanomas state whether the depth of the tumor
tous”) reaction (2.17). (Breslow’s invasion thickness) is < 1 mm or > 1 mm.
The molecular revolution has seen an increasing avai-
lability of molecular data, allowing some nevi to be
characterized genetically. For example, nevi that har- 2.2 Non-melanocytic pigmented lesions
bor mutations in the BAP1 gene (“BAPoma”, “Wiesner These lesions are included here because they may
nevus”) have a peculiar cytomorphology with large enter into the differential diagnosis of melanoma. Non-
melanocytes and abundant cytoplasm (13, 14). Cli- melanocytic does not mean non-pigmented or even
nically and dermatoscopically they resemble “Unna not pigmented by melanin; non-melanocytic lesions
nevi” and are frequently referred to as “dermal nevi” may be pigmented by melanin or by hemoglobin or
by clinicians. When they appear in large number, they hemosiderin. Non-melanocytic lesions that are always
may indicate a germline mutation in the BAP1 gene, non-pigmented are presented in chapter 6. As already
which predisposes to melanoma (especially ocular mentioned the term “non-melanocytic” is problematic
melanoma) and other malignant neoplasms (2.18). and artificial because it combines conditions that are
entirely unrelated under one umbrella.
2.1.2 Melanoma
Melanoma is the only malignant neoplasia of melano- 2.2.1 Vascular proliferations, vascular malformations
cytes. In order to avoid ambiguities, in this book we and hemorrhage
use only this term. We do not use terms such as lentigo The pigmentation of vascular proliferations or malfor-
maligna, lentigo maligna melanoma, superficial sprea- mations and hemorrhage is not caused by melanin, but
ding melanoma, nodular melanoma or acral lentiginous by the blood pigment hemoglobin or its degradation
melanoma because these traditional classifications are product hemosiderin. In the following, a distinction is
inconsistent anatomically and histomorphologically, nor made between hemangiomas, which represent a proli-
are they of any prognostic significance (2.19). Nor do feration of blood vessels; and malformations of vessels
we use special designations based on pathological such as nevus flammeus, which include dilatations of
features, such as desmoplastic melanoma or nevoid pre-existing vessels such as nevus araneus (“spider
melanoma. We acknowledge, however, that there are nevus”) or angiokeratoma (15).
different manifestations of melanoma and that it is impor-
tant to be familiar with these different manifestations. Hemangiomas
Melanomas can be non-pigmented. They may mimic Hemangiomas are proliferations of blood vessels. Given
nevi and viral warts. No two melanomas look alike. the large number of different entities and the existing
While it may be useful to group melanomas into diffe- confusion of nomenclature, hemangiomas cannot be
rent categories, for example for therapeutic reasons, dealt with exhaustively in this book. So-called “senile”
we do not think it is necessary to give a name to every or tardive angiomas (“cherry angioma”) are worthy of
manifestation of melanoma, like nevoid melanoma, mention because they are common. These are cherry-
verrucous melanoma, or animal-type melanoma, to red small papules which appear in large numbers. Like
name just a few of the terms that have been used. the much larger infantile hemangiomas (“strawberry
Every current classification of melanoma is arbitrary hemangioma”), tardive angiomas have such a cha-
and currently all classifications have drawbacks. New racteristic clinical appearance – merely because of
classifications based on molecular data are emerging. their color – that they rarely pose difficulties in terms of
It is not yet known whether this will lead to a rational differential diagnosis (e.g. traumatized or thrombosed
basis for using one or more of the current classification angiomas) (2.20).
systems, or to a new, purely molecular classification of Pyogenic granuloma is a benign, reactive (it often occurs
melanoma. Most likely we will see a more meaningful after trauma or as a consequence of bacterial infection)
and generally accepted classification that integra- vascular proliferation that presents clinically as a rapidly
tes morphologic, biologic and molecular data in the growing, usually eroded, reddish nodule. Melanoma
future. Until such a classification exists, we will call all mimics pyogenic granuloma far more frequently than
melanomas just melanoma. any other proliferation of vessels. To avoid this grave
We do use the term “melanoma in situ”, which refers error, tissue should always be submitted for histology
to a non-invasive melanoma confined to the epider- when treating a “pyogenic granuloma” (2.21).
mis. A lentigo maligna is in fact one type of in situ Hemangiomas are more diverse from the pathologist’s
melanoma. Throughout the book, legends to figures of point of view than from the clinician’s or dermatoscopist’s
Principal pigmented skin lesions relevant to dermatoscop 43
point of view. Many benign vascular proliferations such papule. The latter is a small and extended arteriole.
as the targetoid hemosiderotic hemangioma, glomeru- Nevus araneus may occur singly or multiply. In some
loid hemangioma and microvenular hemangioma can cases this condition is believed to be associated with
only be distinguished by the pathologist and will not liver disease, pregnancy or hormone therapy.
be addressed here. Angiokeratomas are ectasias of the vessels of the upper
vascular plexus with reactive hyperplasia of the epider-
Vascular malformations mis. Various types of angiokeratomas exist. From the
The nevus flammeus (“port-wine stain”) is a common dermatoscopic point of view, the solitary type is most
malformation of vessels with a characteristic clinical relevant. Clinically it is seen as a solitary, red, occasi-
appearance. There is a pale, and subsequently darker, onally black nodule (or papule) with a hyperkeratotic
erythema caused by superficial telangectasias. Predilec- surface (2.22).
tion sites are the scalp, the neck, and the sacral region.
As the clinical diagnosis is usually very clear, the nevus Kaposi’s disease (“Kaposi sarcoma”)
flammeus is rarely investigated by dermatoscopy. Kaposi sarcoma is actually not a sarcoma, i.e. a mali-
The nevus araneus (“spider nevus”) is composed of small gnant neoplasm, but a reactive proliferation of vessels
superficial telangectasias that arise from a central red due to infection with the human herpes virus type 8
Principal pigmented skin lesions relevant to dermatoscop 45
(HHV-8). The skin lesions in Kaposi sarcoma usually 2.2.2 Melanotic macules
occur on the distal extremities, but when associated Before we consider melanotic macules we must clarify
with HIV, Kaposi sarcoma may occur at any location. the term “lentigo”. Lentigo is an extremely vague term,
Clinically one finds multiple rust-brown or livid spots that derived from the word “lentil”, which signifies no specific
may develop into plaques or nodules over time (2.23). diagnosis on its own. Solar lentigo, mucosal lentigo, ink-
spot lentigo, lentigo simplex and lentigo maligna have
Hemorrhage nothing in common except the word “lentigo”. Lentigo
Hemoglobin and its degradation products produce the simplex is the name given to a small junctional Clark
color of a hemorrhage. A fresh superficial hemorrhage nevus while lentigo maligna is an in situ melanoma.
usually appears red while older ones are brown or black. Both of these are melanocytic and are described in the
Hemorrhage becomes relevant to the dermatoscopist section on melanocytic lesions. All other lesions termed
when it mimics melanoma. Hemorrhage in the nail-bed lentigo are not melanocytic lesions.
or bleeding in the stratum corneum of the epidermis (most The term “melanotic macule” includes all non-neoplastic
frequent on acral skin) are the two conditions which lesions that are caused by hyperpigmentation of basal
commonly raise concern. The hemorrhage sometimes keratinocytes, but without significant increase in the
termed “black heel” is seen in figure (2.24). number of melanocytes (16). This includes all types of
46 Principal pigmented skin lesions relevant to dermatoscop
genital lentiginosis and lentiginosis of the lip and the finds several light-brown or dark-brown spots. Genital
oral mucosa, PUVA lentigines, and ink-spot lentigo. Len- lentiginosis is much more diverse in terms of morpho-
tiginous lesions may also occur in the course of diseases logy. As a reliable distinction between this entity and
such as Peutz-Jeghers syndrome, the LEOPARD syndrome, a mucosal melanoma cannot always be made on the
or the Laugier-Hunziker syndrome. Excluded from this basis of clinical features alone, these lesions are fre-
list is solar lentigo, which is associated with epidermal quently biopsied (2.25).
hyperplasia and is therefore regarded, together with
seborrheic keratosis, as a benign epithelial neoplasia. PUVA lentigines and Ink-spot lentigo
The small dark pigmented macules that occur after
Genital lentiginosis and lentiginosis of the lip and the PUVA radiation and ink-spot lentigo are similar lesions,
oral mucosa probably induced by UV light. Because of their striking
The relatively common genital lentiginosis and lentigi- dark pigmentation, they are occasionally biopsied to
nosis of the lip and the oral mucosa are also grouped histopathologically rule out melanoma (17). PUVA len-
under the term mucosal lentiginosis. Clinically one tigines are usually multiple, whereas ink-spot lentigo is
48 Principal pigmented skin lesions relevant to dermatoscop
commonly solitary. On histology the basal keratinocytes spots, age spots, liver spots and “freckles” (although
at the base of the rete ridges are strongly pigmented “freckle” is more correctly used for the ephilis). They
(18) (2.26). usually occur on chronically UV-exposed skin and
become more common with advancing age. Their
Lentigines in the course of diseases brown color is due to concentration of melanin in basal
Multiple lentigines may occur as part of syndromes such keratinocytes (melanin produced by melanocytes is
as Peutz-Jeghers syndrome, the LEOPARD syndrome, transferred to keratinocytes via melanocytic dendrites).
the Laugier-Hunziker syndrome, the Bannayan-Riley- These lesions are classified as non-melanocytic because
Ruvalcaba syndrome, and the diseases grouped under number of melanocytes is only slightly increased, if at
the term “Carney complex” (19–23). Whether the len- all. The epidermis is hyperplastic and has elongated rete
tigines of the various syndromes can be distinguished ridges (except often on the face). Solar lentigines may
from each other on clinical examination, dermatoscopy develop into seborrheic keratoses and are regarded
or histopathology has not yet been investigated. as a precursor of these by many authors (1) (2.27).
Another explanation for the morphological diversity on hair-bearing skin, they are not found on the palms
may be various mutations and environmental factors. or the soles (except in a rare genodermatosis called
The main clinical significance of these harmless lesions the basal cell nevus or Gorlin-Goltz syndrome), the lips
is that occasionally they can mimic melanoma, and or the mucosa. A distinction is made between various
vice versa (2.28). types, but this classification differs according to the
viewpoint. As is true for melanocytic nevi, clinicians
Lichen planus-like keratosis and dermatopathologists speak different languages
The term “lichen planus-like keratosis” was coined by in this regard.
Shapiro and Ackermann in 1966 (24). The term does A basal cell carcinoma may be pigmented (2.30) but
not refer to a specific diagnosis but to a solar lentigo, or most are non-pigmented. The term “pigmented basal
less often a seborrheic keratosis, in a stage of regression. cell carcinoma” is used by clinicians, but not always
Lichen planus-like keratosis usually occurs on chronic by pathologists. A common pathological classification
UV-exposed locations, such as the face or the dorsum includes the following subtypes: nodular, superficial,
of the hand. As a rule one finds a solitary lesion or morpheaform, fibroepithelial and infundibulocystic (1).
more rarely an accumulation of several lesions. They
are usually flat with sharply defined margins, colored Squamous cell carcinoma
brown and/or gray. Several solar lentigines are nearly Superficial types of squamous cell carcinoma of the skin
always found in the vicinity of the lesion. Many lichen include actinic keratosis and Bowen’s disease (1). Someti-
planus-like keratoses are biopsied because the clinical mes the term “intraepidermal carcinoma” is used instead
appearance raises the suspicion of a melanoma in of Bowen’s disease. The causal role of UV exposure in
regression, or of a basal cell carcinoma (2.29). all forms of squamous cell carcinoma is undisputed. The
view that actinic keratoses are a variety of superficial
2.2.4 Malignant epithelial neoplasms (“keratinocyte squamous cell carcinoma is not universally accepted,
cancer”) with some preferring to call them “precancerous lesions”.
Basal cell carcinoma and squamous cell carcinoma are Actinic keratoses are usually numerous and mainly found
commonly referred to together as “non-melanoma skin in chronic UV-exposed areas. Clinically they appear as
cancer”, but for reasons already mentioned this is a rough white hyperkeratoses on an erythematous base.
problematic term. Alternative terms proposed to refer Bowen’s disease, another type of superficial squamous
to basal cell carcinoma and squamous cell carcinoma cell carcinoma, is usually a red scaly plaque which can
collectively include “cutaneous malignant epithelial easily be mistaken clinically for a psoriatic lesion or
neoplasms” or “keratinocyte skin cancer”, but these eczema. Both actinic keratosis and Bowen’s disease are
terms are also problematic. There are other cutaneous usually non-pigmented, but pigmented varieties also
malignant epithelial neoplasms, not just basal cell car- exist, and constitute an important differential diagnosis
cinoma and squamous cell carcinoma, and traditionally for melanoma. In contrast to superficial types, invasive
neoplasms are classified according to their differentiation cutaneous squamous cell carcinomas are only very
and not according to the cell of origin, which is often rarely pigmented (2.31 and 2.32).
unknown. Strictly speaking, a basal cell carcinoma is
an adnexal neoplasm with follicular differentiation and 2.2.5 Adnexal neoplasms
not “keratinocyte cancer”. The general term adnexal neoplasm includes those with
It is best to avoid all these collective terms, whenever follicular, sebaceous, apocrine and eccrine differenti-
possible. ation. Although basal cell carcinoma is often lumped
together with squamous cell carcinoma under umbrella
Basal cell carcinoma terms like “non-melanoma skin cancer” or “keratinocyte
The basal cell carcinoma is a malignant epithelial neo- cancer” it is an adnexal neoplasm with follicular diffe-
plasm whose differentiation is similar to that of follicular rentiation. Most adnexal neoplasms are not pigmented
epithelium. As already mentioned it is an adnexal neo- and are therefore not mentioned in this glossary. Apart
plasm with follicular differentiation. Another name for from basal cell carcinoma, trichoblastoma is the only
basal cell carcinoma is “trichoblastic carcinoma” but adnexal neoplasm that is commonly pigmented (2.33).
this is rarely used. Basal cell carcinoma is occasionally Trichoblastoma is a benign neoplasm with follicular dif-
referred to as a semi-malignant lesion on the basis that ferentiation mostly occurring in conjunction with a nevus
while they grow in a locally destructive manner, they sebaceous. All other adnexal neoplasms are rarely or
rarely metastasize. As basal cell carcinomas only occur never pigmented. Images of pigmented spiradenoma
Principal pigmented skin lesions relevant to dermatoscop 51
2.2.6 Dermatofibroma
Dermatofibroma is not a neoplasm in the strict sense
of the word, but a post-inflammatory tissue reaction
associated with fibrosis of the dermis (1). For this rea-
son, dermatofibromas are nodular and typically sink
below skin level when squeezed between two fingers.
Dermatofibromas occur most commonly on the calf,
but they may occur at any location. With melanin
hyperpigmentation of basal keratinocytes above the
zone of dermal fibrosis, dermatofibromas are usually Figure 2.38: Tinea nigra.
Light-brown macule on the toe (Tinea nigra).
light-brown in color (2.36).
References
1 Ackerman AB, Kerl H, Sanchez J. A Clinical Atlas of 101 18 Bolognia JL. Reticulated black solar lentigo (‘ink spot’
Common Skin Diseases: Ardor Scribendi; 2000. lentigo). Arch Dermatol 1992; 128: 934–40.
2 Darlington S, Siskind V, Green L, Green A. Longitudinal 19 Wilkes D, McDermott DA, Basson CT. Clinical phenotypes
study of melanocytic nevi in adolescents. J Am Acad and molecular genetic mechanisms of Carney complex.
Dermatol 2002; 46: 715–22. Lancet Oncol 2005; 6: 501–8.
3 Siskind V, Darlington S, Green L, Green A. Evolution of 20 Fargnoli MC, Orlow SJ, Semel-Concepcion J, Bolognia
melanocytic nevi on the faces and necks of adolescents: JL. Clinicopathologic findings in the Bannayan-Riley-
a 4 y longitudinal study. J Invest Dermatol 2002; 118: Ruvalcaba syndrome. Arch Dermatol 1996; 132: 1214–8.
500–4. 21 Coppin BD, Temple IK. Multiple lentigines syndrome
4 Kachare SD, Agle SC, Englert ZP, Zervos EE, Vohra NA, (LEOPARD syndrome or progressive cardiomyopathic
Wong JH et al. Malignant blue nevus: clinicopathologi- lentiginosis). J Med Genet 1997; 34: 582–6.
cally similar to melanoma. Am Surg 2013; 79: 651–6. 22 McGarrity TJ, Amos C. Peutz-Jeghers syndrome: clinico-
5 Clark WH, Jr., Reimer RR, Greene M, Ainsworth AM, pathology and molecular alterations. Cell Mol Life Sci
Mastrangelo MJ. Origin of familial malignant melano- 2006; 63: 2135–44.
mas from heritable melanocytic lesions. ‘The B-K mole 23 Kanwar AJ, Kaur S, Kaur C, Thami GP. Laugier-Hunziker
syndrome’. Arch Dermatol 1978; 114: 732–8. syndrome. J Dermatol 2001; 28: 54–7.
6 Ackerman AB, Magana-Garcia M. Naming acquired 24 Shapiro L, Ackerman AB. Solitary lichen planus-like
melanocytic nevi. Unna’s, Miescher’s, Spitz’s Clark’s. keratosis. Dermatologica 1966; 132: 386–92.
Am J Dermatopathol 1990; 12: 193–209. 25 Schwartz RA. Superficial fungal infections. Lancet 2004;
7 Reed RJ, Ichinose H, Clark WH, Jr., Mihm MC, Jr. Com- 364: 1173–82.
mon and uncommon melanocytic nevi and borderline
melanomas. Semin Oncol 1975; 2: 119–47.
8 Bar M, Tschandl P, Kittler H. Differentiation of pigmented
Spitz nevi and Reed nevi by integration of dermatopa-
thologic and dermatoscopic findings. Dermatol Pract
Concept 2012; 2: 13–24.
9 Spitz S. Melanomas of childhood. Am J Pathol 1948;
24: 591–609.
10 Sutton RL. An unusual variety of vitiligo (leukoderma
acquisitum centrifugum). J Cutan Dis 1916; 34: 797–801.
11 Schaffer JV, Glusac EJ, Bolognia JL. The eclipse naevus:
tan centre with stellate brown rim. Br J Dermatol 2001;
145: 1023–6.
12 Meyerson LB. A peculiar papulosquamous eruption
involving pigmented nevi. Arch Dermatol 1971; 103:
510–2.
13 Llamas-Velasco M, Perez-Gonzalez YC, Requena L,
Kutzner H. Histopathologic clues for the diagnosis of
Wiesner nevus. J Am Acad Dermatol 2014; 70: 549–54.
14 Wiesner T, Obenauf AC, Murali R, Fried I, Griewank
KG, Ulz P et al. Germline mutations in BAP1 predispose
to melanocytic tumors. Nat Genet 2011; 43: 1018–21.
15 Fritsch P. Dermatologie und Venerologie: Grundlagen.
Klinik. Atlas. Berlin: Springer; 2003.
16 Sanchez J. Unifying Concept of Melanotic Macule:
Synonymy for Melanotic Macule on Different Anatomic
Sites. Dermatopathol: Pract & Conc 1998; 4: 2.
17 Bleehen SS. Freckles induced by PUVA treatment [pro-
ceedings]. Br J Dermatol 1978; 99: 20.
A B C D E F
A B C
Figure 3.6: Schematic diagram of various types of parallel lines on acral skin. A, parallel lines in furrows; B, parallel lines on ridges, and
C, parallel lines that cross ridges.
Pattern Analysis – Basic Principles 57
A B C D E
A B C
Black
Melanin in stratum corneum, congealed blood
Dark brown
Melanin in the epidermis, dense
Gray
Melanin in the papillary dermis
Blue
Melanin in the reticular dermis
Orange
Combination of melanin and keratin, serum crust
Keratin
Yellow
Keratin
White
Fibrosis, sclerosis, keratin, pus, sebum, sebaceous glands
Red
Blood
Hemoglobin
Purple
Blood (poorly oxygenated)
A B
C D
Furthermore, the observed color depends on the den- extravasation of red blood cells occurs (i.e. hemorrhage),
sity of melanin and the thickness of the epidermis. A the entire coagulated blood is dark red, or may be
dense accumulation of melanin in the basal layer of the black when it is in the stratum corneum (corneal bleed-
epidermis may appear dark-brown or nearly black, a ing). In the dermis, fresh blood may again be red to
less dense accumulation would be light-brown. When blue, but degradation of hemoglobin may give rise to
the epidermis is thickened due to acanthosis (e.g. some various shades of color ranging from green to brown.
seborrheic keratoses) melanin in the epidermis may In cases of ulcerated lesions, serum exudes from the
appear blue. surface, dries, and forms a crust. The serum is often
It is often useful to divide pigmented lesions into those mixed with red blood cells and is therefore orange.
where melanin is the dominant pigment, and those Keratin is white or yellow (when the stratum corneum
where other pigments dominate. is not pigmented it appears yellow). Mixture of the
white or yellow of keratin with the brown of melanin
3.3.2 Other pigments gives rise to colors in the range orange to yellow which
Hemoglobin is the next most important pigment. Depend- are characteristic of pigmented seborrheic keratoses.
ing on the level of oxygen saturation, hemoglobin in Fibrosis or sclerosis of the dermis also appears white.
vessels ranges from bright red to blue. When a massive Pus, sebum and sebaceous glands also appear white
Pattern Analysis – Basic Principles 65
in dermatoscopy. In contrast to keratin and fibrosis, Distribution of color in lesions with a purely reticular
which are shiny white, pus, sebum and sebaceous pattern
glands appear dull white (3.19). It is usually sufficient to distinguish between symmetrical
and asymmetrical arrangements of more than one color.
3.3.3 Color combinations Pigmented lesions that consist exclusively of a reticular
A pigmented lesion may be composed of one or several pattern are, however, a special case, as three further
colors. As is true for patterns, colors may be arranged specific arrangements are diagnostically significant.
symmetrically or asymmetrically. With the exception of When a darker shade is seen in the center and a lighter
brown, variations in shade should not be interpreted one at the periphery, so that symmetry is retained, the
as a separate color. It is diagnostically meaningful lesion is considered to be centrally hyperpigmented.
to distinguish between light-brown and dark-brown, If the darker shade is seen at the periphery the lesion
but only in clear-cut cases and particularly when the is called eccentrically hyperpigmented. If the colors
transition between the two shades of brown is abrupt. within the lesions are distributed in such a way that
It is very common to see lighter pigmentation at the areas of dark pigmentation alternate with areas of
periphery of a lesion, or around follicular openings; light pigmentation, the lesion is termed speckled or
such lesions should still be classified as one color. The variegated (3.21).
number of colors and the presence of specific colors This can involve two shades of the single color brown
are of immense importance in dermatoscopy. When (a single color because the transition is gradual), or
evaluating colors (as patterns), the investigator should two shades of brown plus black.
know when the observation should be very accurate
and when it may be less exact. The color of normal skin
varies from person to person, and even according to 3.4 Descriptions of pigmented lesions on the
location on the body. While this normal skin color is not basis of patterns and colors
counted as a separate color, it is used as a reference to Formulating a description of pattern(s) and color(s)
define “white”; white structures must be clearly lighter is always the first step towards diagnosis. One first
in color than surrounding normal skin (3.20). looks at the lesion from a distance. A pattern should
66 Pattern Analysis – Basic Principles
A B C
Figure 3.21: Possible distribution of color in lesions that consist exclusively of reticular lines. Centrally hyperpigmented (A), eccentrically
hyperpigmented (B), speckled or variegated (C).
occupy a significant portion of the pigmented lesion; contain basic elements, but too few to constitute a
everything else may be initially ignored. Beginners tend pattern. In other words, not all pigment is structure,
to become immediately absorbed in details. However, some is noise.
at least at the initial step, single dots or single clods
are of no importance. If necessary, these details can Examples of pigmented lesions with more than one
be incorporated later in the analysis. pattern
Figure 3.24 shows pigmented lesions with more than
Examples of pigmented lesions with a single pattern one pattern. When assessing a lesion consisting of
Figures 3.22 and 3.23 show pigmented lesions with more than one pattern, the first question one should ask
just one pattern.1 Usually it is simple to decide which is whether the patterns are combined symmetrically or
pattern is present. Occasionally it may be difficult to asymmetrically. When assessing symmetry one should
distinguish between reticular lines and branched lines, exercise latitude, as symmetry in biology never reaches
in which case one should prefer the more common geometrical perfection. A beginner tends to over-inter-
reticular pattern. pret in favor of asymmetry. The examples in figure 3.24
In almost all cases, it is pigment that defines structure. will help the reader to develop a feeling for biological
A few examples will show how potential confusion symmetry and asymmetry. The vast majority of lesions
can be avoided by keeping this principle in mind. The can be unequivocally classified as either symmetrical or
less pigmented areas between reticular lines are not a asymmetrical. When (very rarely!) this cannot be done
separate pattern, i.e. they are not clods. Hypopigmented with certainty, one “investigates” in both directions, as
areas between dots, clods, circles and all other lines we will see later.
should not be viewed as structureless areas. A circle of
hypopigmentation around a hair follicle is not a structure,
only an interruption to the pattern of the lesion. The few 3.5 Clues
exceptions to pigment defining structure – mainly white Sometimes pigmented lesions can be unequivocally
structures – will be addressed in detail later. diagnosed on the basis of pattern(s) and color(s) alone.
For a zone to constitute the structureless pattern it must More often, assessment of pattern and color leads to
– as for patterns composed of basic elements – occupy a small differential diagnosis. In this case, one looks
a significant portion of the lesion. On the other hand, for clues. A clue is simply a feature which favors one
a sufficiently large zone does not have to be entirely possible diagnosis over another possible diagnosis.
devoid of pigmented structures to be considered struc- Sometimes a pattern may also constitute a clue, for
tureless. Areas with visible structures which cannot be example a structureless eccentric zone is both a pat-
definitively classified as one of the basic elements are tern, and (in some contexts) a clue to the diagnosis of
still correctly termed structureless, as are areas which melanoma. Usually however, clues are features too
localized to constitute a pattern, but nonetheless favoring
1 As this chapter is mainly focused on a description of patterns and colors one diagnosis over another. Clues include a special
rather than diagnosis, only the dermatoscopic images are shown here.
Clinical and dermatoscopic appearances are shown simultaneously in arrangement of basic elements, a typical color, a special
most of the remaining chapters. combination of pattern and color, a characteristic pattern
Pattern Analysis – Basic Principles 67
A B C
D E F
G H I
J K L
M N O
A B C
D E F
G H I
J K L
A B C
D E F
G H I
J K L
M N O
of vessels, or even absence of a feature. Clues may, but terns of vessels are most conspicuous in the absence of
need not necessarily, be present. The more numerous pigmentation. In cases of dense melanin pigmentation,
the clues that support a diagnosis (and the fewer that vessels are difficult or impossible to visualize.
support an alternative diagnosis), the more likely that Patterns formed by vessels are usually much less specific
specific diagnosis is correct. Pattern and color limit the than patterns formed by melanin. As a general princi-
differential diagnoses, but clues confirm a diagnosis ple, whenever pigment is present one should attempt
or rule it out. Often however, clues are weak, or even to reach a diagnosis on the basis of these pigmented
contradictory. For these cases, judgement is required in structures, and relegate blood vessel analysis to the
deciding how much weight to assign to each clue, and status of a clue to diagnosis. However, when diagnos-
hence which final diagnosis should be favored. This is ing non-pigmented lesions, one has to rely on analysis
discussed in greater detail in chapter 7. As a general of vessels and keratin structures to reach a diagnosis.
principle, more weight should be given to the pattern Blood vessels are described using the same geometri-
overall than to any single clue. The most difficult part cally defined basic elements used to describe pigment-
of dermatoscopy is to correctly assign weight to clues, ed structures. Like pigmented structures, vessels may
and particularly to avoid overvaluing an unreliable appear as dots, clods or lines. Additional line types
or misleading clue. This is the role of experience in (looped, curved, serpentine, helical and coiled) are
dermatoscopy. defined for vessels, as these are not seen in pigmented
structures. Analogous to patterns formed by pigment,
Pattern of vessels a collection of vessels of the same type gives rise to a
Blood vessels are visible dermatoscopically due to the pattern of vessels.
hemoglobin they contain. Analysis of vessel pattern(s) Linear vessels are classified based on the number and
often serves as an additional clue to diagnosis. Pat- type of curves (3.25). Those with no curves are termed
Pattern Analysis – Basic Principles 71
“straight”. Those with one curve are termed “looped” A pattern of vessels is composed of multiple vessels of
when the bend is so sharp that two sections are formed the same type. When one vessel type predominates
which are sensibly parallel. A vessel with a single over the others, we term the pattern of vessels “mono-
obtuse bend is termed “curved”. Vessels with more morphous”. When more than one pattern of vessels is
than one bend are termed “serpentine”. A serpentine present we use the term “polymorphous”.
vessel is termed “helical” when the curves are focused Evaluation of the vascular pattern is not always simple.
on a central axis. Vessels are called “coiled” when Sometimes one is unable to conclusively assign individual
convoluted compactly. vessels to a specific type. Rather than becoming absorbed
“Thick” or “thin” and “short” or “long” are additional in details one should observe the general pattern; the
attributes one may use to describe linear vessels. Log- assessment of individual vessels is rarely useful (3.27,
ically, the use of these terms is limited to linear vessels. 3.28). For instance, in some melanomas one finds – to
Usually these additional terms contribute nothing to the the extent that pigment allows the vascular pattern to be
diagnostic process. However, in exceptional cases these inspected at all – a mixture of short and straight, short
distinctions might provide useful information. Generally, and curved, short and serpentine, and coiled serpentine
vessels are “thick” only when they are much thicker than vessels. The exact classification of individual vessels is
normal nail fold capillaries. Vessels are “long” only not important in this instance; the overall impression is
when they cross a significant part of the lesion and so important and that is of polymorphous vessels.
applies only to straight, serpentine, or helical vessels. It may be difficult or even impossible to distinguish
Conversely, vessels are “short” only when their length vessels as dots from vessels as small coils (3.27 top
does not greatly exceed their breadth. This applies – if left). At higher magnifications nearly all vessels will
at all – only to straight, curved and serpentine linear have discernable shape; vessels that are dots at mag-
vessels. nifications equivalent to the handheld dermatoscope
In addition to the morphology of individual vessels, their are classified as dots, regardless of their appearance
arrangement relative to one another and to the lesion at higher magnifications. Vessels on large, skin colored
as a whole is also important (3.26). In the majority clods are usually of the linear type; either curved,
of cases, vessels appear to be distributed randomly, serpentine or looped (3.29). If each clod contains
i.e. not arranged in any specific manner throughout multiple linear vessels the arrangement is clustered
the lesion. However, a few important exceptions exist. (3.29 top row). If each clod contains a single linear
Vessels as dots or coils may be arranged as lines. When vessel (usually of the curved type) the arrangement is
vessels as dots or coils are arranged in straight lines, centered (3.29 bottom row).
this arrangement is termed “linear” (it is important not As regards the occasionally difficult distinction between
to confuse linear vessels with linear arrangement of curved vessels and short serpentine ones, one should
vessels). When these vessels are arranged in serpentine remember that curved vessels are usually much thicker
lines the pattern is known as “serpiginous”. When vessels than short serpentine ones. Sometimes it is also difficult
are not uniformly distributed but are much more dense to decide if there is a specific arrangement of vessels
at some sites than others, this arrangement is termed or not. In these doubtful cases it is better to assume
“clustered”. Linear vessels of any type at the periphery that the vessels are arranged randomly. One should
that are oriented towards but do not cross the center are also make a distinction between vascular patterns and
termed “radial”. The arrangement of linear vessels (most erythema. Erythema is not a pattern of vessels but a
commonly curved, sometimes serpentine or looped) in reddening caused by vasodilatation (usually due to
the center of skin colored or light brown clods is termed inflammation) which, in contrast to a red structureless
“centered”. As for pigmented lesions, we term straight area, does not cover pigmented structures but gives
linear vessels that intersect each other nearly at right the background skin a red hue.
angles “reticular”. Finally, serpentine vessels may be Correct technique is critical for dermatoscopic imaging
arranged such that multiple vessels originate from one of vessels. Strictly speaking, we do not see the vessels
common vessel; the derivative vessels typically origi- themselves, we see hemoglobin in blood. Too much
nate from a thicker vessel. This arrangement is termed pressure on the glass plate compresses the vessels
“branched”. Equipped with these definitions, one may and, with blood thus excluded, renders the vessels
describe the morphology of, and classify, all vascular invisible. For optimum evaluation of vessels one should
patterns. Patterns of vessels are described using the use a contact medium of higher viscosity, such as ultra-
same general principles as those used to describe the sound gel, which is retained between the lesion and
patterns formed by pigment (3.25, 3.26). the glass plate of the dermatoscope without the use of
Pattern Analysis – Basic Principles 73
pressure. Alternatively, contact with the skin surface 3.6 Characteristic features of pigmented
(and thus vessel compression) can be avoided entirely non‑melanocytic lesions
by using a polarizing dermatoscope with the contact
plate removed. In some cases vessels may be obscured 3.6.1 Proliferation of vessels
by polarizing-specific white lines and clods, and in Hemangiomas and vascular malformations
these cases they may be visualized more clearly with Hemangiomas and vascular malformations arguably
non-polarizing (contact) dermatoscopy. have the most distinctive dermatoscopic appearance of
all lesions. There is one pattern, clods, with color ranging
Other clues between red and purple, depending on the degree of
Vascular patterns are only one of many clues that we use oxygenation of the blood in the vessels (3.30, 3.31).
to establish the diagnosis when assessment of pattern and Black clods are caused by thrombosis of vessels or are
color alone are insufficient. These clues will be addressed indicative of older blood crusts due to exogenous trauma.
in the following chapters describing the principal pigment- Other basic elements are completely absent, i.e. there are
ed lesions. It should be mentioned here that a stepwise no lines, pseudopods, circles or dots. In some instances
description – first of pattern, then color, and finally clues, one may find a structureless area adjacent to the clods.
is the best way of arriving at the diagnosis. Hemangioma should not be diagnosed when any vessels
as lines or dots are found within red or purple clods,
Pattern + Color + Clue = Diagnosis as this pattern may be seen in amelanotic melanoma.
Pattern Analysis – Basic Principles 75
Hemangioma
Pattern Colors Clues
Typical: Typical: None
Only clods Red and/or purple
Occasional: Occasional:
Clods and structureless Black
1 2
3 4
Pyogenic granuloma
Pattern Colors Clues
Typical: Typical: Typical:
Clods and structureless Pink, red, white The clods are separated by thick white or
skin-colored lines, and the tumor is sur-
rounded by a white or light-brown margin.
Occasional:
Erosions and ulcerations, which are seen as
orange, dark-red or black clods or struc-
tureless areas.
Correlation between dermatoscopy and tive tissue (thick, white or skin-colored lines). Pyogenic
dermatopathology granuloma is frequently eroded and therefore coated
Red or purple clods correspond to dilated and blood- with a crust of blood or serum. These erosions may be
filled vessels in the dermis. The color depends on the seen as orange, red or black clods or structureless areas.
oxygenation of blood and the location of the prolifera-
tion of vessels. Vessels located higher in the dermis are Solitary angiokeratomas
red whereas deeper ones tend to be purple. Solitary angiokeratomas reveal a similar pattern of
vessels as hemangiomas or senile angiomas. However,
Pyogenic granuloma in contrast to hemangiomas which are usually marked
The pyogenic granuloma is a reactive proliferation of by bright red clods, solitary angiokeratomas have
vessels. Its pattern is similar to that of hemangiomas, dark-red, purple or black clods (3.33). Structureless
but the clods are usually pink or bright-red and typically areas are also more common in angiokeratomas than
separated from each other by thick, white or skin-colored in hemangiomas. Sometimes there is marked hyper-
lines (3.32). Occasionally pyogenic granuloma has a keratosis (3.34). Occasionally it may be difficult to
white or light-brown periphery. As pyogenic granuloma differentiate the purple clods of angiokeratoma from
is frequently eroded, one may find orange, dark-red or the blue clods pigmented by melanin (like for exam-
black clods or structureless areas. ple in basal cell carcinoma). As a rule of thumb, one
should assume that the pigment is hemoglobin if most
Correlation between dermatoscopy and of the other clods are red, and assume the pigment
dermatopathology is melanin when found associated with brown. Of
The pink clods represent dense proliferations of vessels course it is prudent to assume the pigment is melanin
in the dermis, which are separated by septa of connec- in equivocal cases.
Pattern Analysis – Basic Principles 77
Intracorneal hemorrhage
Pattern Colors Clues
Typical: Typical: Typical:
Structureless, clods or parallel lines on the Red, reddish-brown (recent), black (old) Sharp contours Small satellite clods
ridges detached from the main lesion
Solar lentigo
Pattern Colors Clues
Typical: Typical: Typical:
Trunk: Reticular and/or curved lines Light-brown Sharply demarcated, scalloped (with multi-
Face: Structureless, reticular or curved lines ple concavities) border
Forearm and dorsum of the hand:
Structureless and/or dots
Correlation between dermatoscopy and White dots or clods are seen in all types of seborrheic
dermatopathology keratosis, but become more common in more raised
The brown reticular and curved lines are due to hyper- lesions, i.e. with more advanced acanthosis. As with
pigmentation of basal keratinocytes when rete ridges all aggregations of basic elements, white clods or
are present. The structureless brown pattern corresponds dots must be multiple to form a pattern (though lesser
to hyperpigmentation of basal keratinocytes when the numbers may still constitute a clue).
epidermis is flat (rete ridges are absent). This is usually In addition to white dots or clods, a sharply demarcated
the case on chronic sun-damaged skin. border, a scalloped border and looped and/or coiled
vessels are important clues to seborrheic keratosis.
Seborrheic keratosis
No other benign lesion shows the diversity of derma- Correlation between dermatoscopy and
toscopic appearances seen in seborrheic keratosis dermatopathology
(3.39–3.43). Except for the pseudopod pattern, any As in solar lentigo, the brown lines and circles of flat
pattern or color may be found. seborrheic keratoses result from hyperpigmentation of
Flat seborrheic keratoses (and the flat portions of basal keratinocytes. Reticular lines may become thick
raised types) on the trunk show similar patterns to with acanthosis of the epidermis. The hypopigmented
solar lentigo, i.e. light-brown reticular or curved lines. areas between lines are dermal papillae and infundibula
With early acanthosis (thickening of the epidermis) of the hair follicles. Thick curved lines, clods and circles
and hence thickening of the lesion, thin curved lines of raised or verrucous seborrheic keratoses represent
(frequently arranged as parallel pairs) and circles invaginations of the epidermis filled with keratin (thick
become more prominent. With advanced acanthosis, lines and clods) or infundibula (clods and circles) filled
the predominant structures become thick curved lines with keratin. As white or yellow keratin may be mixed
and clods. with melanin, the spectrum of colors of lines and clods
In early acanthosis, brown and orange (or yellow) are ranges from yellow (no melanin) to orange (moderate
the predominant colors seen. Verrucous types when quantity of melanin), brown (large quantity of melanin),
heavily pigmented are marked by thick curved lines in and in exceptional cases even black.
combination with brown and/or orange clods and/or White dots or clods correspond histopathologically to
a structureless area in shades of brown, blue or gray. cysts filled with keratin. The blue or gray structureless
In less heavily pigmented verrucous types the predom- area of some verrucous seborrheic keratoses is due to
inant feature is often orange, yellow or skin-colored acanthosis of the epidermis, which causes epidermal
clods. melanin to appear blue.
Pattern Analysis – Basic Principles 81
Figure 3.39: Seborrheic keratosis with only one pattern, namely yellow, orange and white clods
Lichen planus-like keratosis There are also other less common patterns of dermatofibro-
Lichen planus-like keratosis is actually a solar lentigo ma. For instance, the central white structureless zone may
(or sometimes a seborrheic keratosis) in regression and be entirely absent. Instead, one may find a few smaller
may therefore show the same dermatoscopic features as eccentrically located structureless zones that may be white
these lesions. Two additional features are clues to lichen or skin-colored. Another uncommon dermatofibroma vari-
planus-like keratosis; erythema (a sign of inflammation) ant has no peripheral lines or circles, consisting entirely
and gray dots and/or clods (3.44). of brown, white and skin-colored structureless zones.
Once the solar lentigo has disappeared and the inflam- Usually dermatofibroma have a symmetric combination of
mation has subsided, complete regression is marked by patterns and colors but exceptions like the one shown in
gray dots and/or clods with no sign of the pre-existing figure 3.46 exist. The firm consistency on palpation is an
lesion. In this case, of course, other differential diag- additional clinical clue to the diagnosis of dermatofibroma.
noses must be considered, including a fully regressed
melanoma. In most cases the distinction is obvious Correlation between dermatoscopy and
because several lichen planus-like keratoses occur dermatopathology
together at typical sites (forearm, dorsum of the hand, Thin reticular lines or circles are caused by elongation
face and back). However, a diagnostic biopsy may be of rete ridges and melanin hyperpigmentation of basal
necessary in some cases. keratinocytes. White structureless zones, thick white
reticular lines and perpendicular white lines are caused
Correlation between dermatoscopy and by dermal fibrosis. Red or pink pigmentation is caused
dermatopathology by inflammation and dilated blood vessels.
On histopathology the gray clods or dots represent
accumulations of melanophages in the papillary dermis. 3.6.5 Melanotic macules
Ink-spot lentigo
3.6.4 Dermatofibroma The characteristic pattern of ink-spot lentigo is reticular
The most common patterns of dermatofibroma on derma- lines (more rarely branched lines) that may be quite
toscopy are reticular lines peripherally and structureless thick, but always have a uniform dark-brown or black
white centrally (3.45). Peripheral reticular lines are pigmentation (3.47). A clue is that the reticular lines
usually brown and always thin – never thick. Instead within the lesion may be interrupted at various sites,
of reticular lines there may be dense light-brown circles and tend to end abruptly at the margin.
or, more rarely, regularly arranged radial lines distrib-
uted over the entire circumference. Thick white lines in Correlation between dermatoscopy and
place of the central structureless zone is also a common dermatopathology
variant. If the center of the dermatofibroma is brown or The reticular pattern of ink-spot lentigo is caused by
red structureless one can also find (polarizing-specific) marked hyperpigmentation (therefore dark-brown or
perpendicular white lines in the center. black) of basal keratinocytes at the rete ridges.
82 Pattern Analysis – Basic Principles
Seborrheic keratosis
Pattern Colors Clues
Typical: Typical: Typical:
Flat: Reticular or curved lines, circles Lines: Brown White dots or clods
Moderately raised: Curved lines, clods, Circles: Brown Sharp border – in cases of flat types, a
circles Clods: White, skin-colored, orange, brown, scalloped border (with multiple concavi-
Verrucous: Clods, thick curved lines, occasionally the pigmentation may be so dense ties). Looped or coiled vessels
structureless that black clods are found.
Structureless area in heavily pigmented types:
Brown, blue and gray.
86 Pattern Analysis – Basic Principles
Dermatofibroma
Pattern Colors Clues
Typical: Typical: Typical pattern:
Reticular and structureless Reticular lines and circles are light-brown, Reticular (or circles) at the periphery, struc-
Variants: structureless zones are either white or tureless (or white reticular lines) or white
Instead of thin brown reticular lines there skin-colored. perpendicular lines (only visible with polar-
may be densely arranged light-brown cir- ized dermatoscopy) in the center.
cles; instead of the structureless white center
there may be thick, white reticular lines.
Rare:
Completely structureless or radial lines at
the periphery (on the entire circumference).
88 Pattern Analysis – Basic Principles
Ink-spot lentigo
Pattern Colors Clues
Typical: Typical: Typical:
Reticular lines Black or dark-brown Reticular lines within the lesion, some inter-
rupted, and abrupt break-off of pigmenta-
tion at the margin
Pattern Analysis – Basic Principles 89
Genital lentigo, labial lentigo stem (branched pattern of vessels). However, while this
Regardless of whether labial or genital lentigines occur pattern is common in nodular basal cell carcinomas it
in isolation or as part of a syndrome, they are charac- is usually absent in superficial basal cell carcinomas,
terized by three different patterns: 1. structureless 2. which are characterized by a polymorphous pattern
curved parallel lines, and 3. circles. The pigmentation of vessels consisting of thin, serpentine vessels that are
ranges from light-brown to dark-brown (3.48). not branched, and occasionally coiled vessels.
Reticular lines and vessels as dots are not seen in basal
Correlation between dermatoscopy and cell carcinoma and when seen constitute a clue against
dermatopathology the diagnosis. Ulceration, which is relatively common
The structureless pattern correlates with hyperpigmentation in basal cell carcinoma can induce the full variety of
of basal keratinocytes in areas where rete ridges are absent polymorphous vessel types including dot vessels, but
or flattened (e.g. on the lip). Patterns of parallel lines and a pattern of dot vessels is not expected.
circles are probably due to the special anatomy of the
epidermis on the vulva and the penis and of the transition Correlation between dermatoscopy and
zone between keratinizing epidermis and mucosa. dermatopathology
Blue, gray and brown clods correspond to pigmented
3.6.6 Pigmented basal cell carcinoma tumor cell aggregates. When they are located deep
Pigmented basal cell carcinomas have a diverse, but they appear gray or blue. In superficial location they
usually characteristic, dermatoscopic appearance, are brown. Radial lines with a common base and radial
showing patterns composed only of radial lines, dots, lines that converge in a central clod arise when several
clods and structureless zones (3.49, 3.50). A common epithelial tumor strands originate from one follicular
pattern and color combination is blue clods that are structure. The histopathological correlate of skin-colored
usually, but not always, of different sizes and shapes. or white structureless zones is the fibrous stroma. In scle-
This may occur in isolation or in combination with brown rosing basal cell carcinoma, this stroma may constitute
clods, gray, blue and/or brown dots, white structure- most of the lesion. Orange clods or orange structureless
less zones, and radial lines. All other arrangements areas are usually a sign of erosion coated with serum.
of lines (reticular, branched, curved and parallel), as
well as pseudopods and circles do not occur in basal 3.6.7 Squamous cell carcinoma
cell carcinoma. These structures are all very strong Invasive cutaneous squamous cell carcinoma is rarely
clues against the diagnosis of basal cell carcinoma. pigmented, but pigmentation is not uncommon in both
The pigmentation of basal cell carcinoma is caused Bowen’s disease and actinic keratosis.
by melanin. The pigmented tumor cell aggregates of
basal cell carcinoma appear brown or gray when they Pigmented actinic keratosis
are superficial, and blue when they lie deeper. An Pigmented actinic keratoses usually occur on the face.
orange-colored structureless area correlates with an On dermatoscopy they may show a variety of patterns
erosion or ulcer coated with serum crusts. Structureless (3.51). Most commonly there are gray and brown
zones are usually central and are skin-colored or white. dots arranged between the follicular openings. Other
Clues include peripheral radial lines, seen segmentally common patterns of facial pigmented actinic keratosis
(as opposed to occupying the entire circumference). are angulated lines, structureless and circles (grey dots
These radial lines may be thin or thick and nearly always arranged around follicular openings). Frequently one
have a common base. Radial lines may also converge can find dermatoscopic criteria of a solar lentigo in
at a central hyperpigmented dot or clod. These latter addition, for example curved lines or a well demarcated,
structures may be seen centrally as well as peripherally. scalloped border (3.51 bottom row).
Both these patterns of radial lines constitute very strong The dermatoscopic pattern of gray dots between or
clues to the diagnosis of basal cell carcinoma. around follicular openings can equally be seen in
Blue clods constitute a relatively specific feature, not melanoma in situ and solar lentigo in regression (lichen
only as a pattern, but also as a clue (when only one planus-like keratosis). Sometimes these three entities
or two blue clods are present). cannot be clearly distinguished from each other on
The pattern of vessels in basal cell carcinoma (both dermatoscopy alone. Clues to pigmented actinic ker-
pigmented and non-pigmented) is an important clue. atosis are scale, white circles and 4 white dots in a
The typical vessel pattern of basal cell carcinoma is square (4-dot clod, 3.51). The latter clue can only be
branched serpentine vessels that originate from a thick seen with polarized dermatoscopy.
90 Pattern Analysis – Basic Principles
Figure 3.53: Dots and coiled vessels arranged in lines in Bowen’s disease.
The dermatoscopic overview on the left shows two patterns (dots and structureless). Brown dots and coiled vessels are arranged in lines in
the periphery. The close-up on the right shows a higher magnification of brown dots arranged in lines.
A C
Figure 3.54: Pigmented Bowen’s disease induced by human papilloma virus (HPV).
Clinical examination (A) reveals a brown plaque with a scalloped border. Dermatoscopically (B) there are two patterns (structureless and
dots). The higher magnification (C) shows brown dots arranged in lines. This clue permits a specific diagnosis.
3.7 Characteristic features of melanocytic lesions greater importance than any single clue to melanoma.
The pattern of vessels in Clark nevi is unremarkable:
3.7.1 Melanocytic nevi it usually consists of a pattern of dots occasionally
Clark nevus interspersed with short vessels, either straight or curved.
The most common acquired nevus is the Clark nevus. Distinguishing between a Clark nevus and a “super-
While on dermatoscopy it shows remarkably diverse ficial” or a “superficial and deep” congenital nevus
morphology, there are specific features which distinguish is usually simple. Clark nevi are flat clinically while
the Clark nevus both from other benign nevi on the “superficial” or “superficial and deep” congenital
one hand and melanoma on the other (3.55–3.59). nevi are raised. The reticular pattern is predominant in
However the range of appearances of Clark nevi does Clark nevus, the clod pattern is usually (but not always)
overlap with both superficial congenital nevi and mel- predominant in congenital nevi. Clark nevi are either
anoma (especially in situ melanoma) to an extent that uniformly pigmented or marked by central or eccentric
differential diagnosis may be quite difficult even with hyperpigmentation, while small congenital nevi are
dermatoscopy. This does not mean that a biological usually centrally hypopigmented or variegate (see
zone of overlap actually exists; rather this should be the section on this entity). Confusion arises because
viewed as a limitation of the method. most dermatopathologists do not make this distinction,
The reticular pattern usually predominates in the Clark describing both Clark nevi and small congenital nevi
nevus; indeed most often thin reticular lines is the only as “dysplastic” junctional or compound nevi or simply
pattern (3.55). In the growth phase, the reticular pat- as a junctional or compound nevus.
tern may be combined with peripheral dots or clods
(3.56). These peripheral structures usually regress as Correlation between dermatoscopy and
the nevus matures. Clark nevi consisting solely of brown dermatopathology
dots or small brown clods are exceptions. Less common The 2-dimensional horizontal projection of the 3-dimen-
again is the combination of the reticular pattern with sional rete ridges causes the characteristic reticular
a structureless zone, nearly always hyperpigmented pattern on the surface of the skin. The pigmented lines
and located centrally. correspond to vertically arranged rete ridges while the
In general – as one expects in benign lesions – combi- hypopigmented center represents the dermal papillae.
nations of patterns are arranged symmetrically in the The brown pigmentation of lines is primarily due to
Clark nevus. The same is not true for arrangements of deposits of melanin in basal keratinocytes. Often the
colors, which may be symmetrical or asymmetrical. As melanocytes themselves are not sufficiently pigmented to
the proliferation of melanocytes in the epidermis is the be visible on dermatoscopy. If they are, they appear as
essential architectural feature of the Clark nevus, the brown dots or clods that correspond to smaller or larger
colors are those of melanin in the epidermis: light-brown, nests of melanocytes at the dermoepidermal junction.
dark-brown and black. The commonest arrangements of
colors in the Clark nevus are uniform brown pigmenta- “Superficial” and “superficial and deep” congenital nevi
tion, and brown peripherally with central hyperpigmen- Like Clark nevi, “superficial” and “superficial and deep”
tation. This pattern of central hyperpigmentation of a congenital nevi are extremely diverse morphologically,
reticular lesion is a clue to the diagnosis of Clark nevus. but have specific clues that usually make diagnosis
Pigmentation may also be variegate, or eccentrically straightforward. Again, it should be remembered that in
hyperpigmented. This pattern of Clark nevi overlaps this context the term “congenital” does not necessarily
morphologically with in situ melanoma. mean that the nevus was visible at birth. As mentioned
Other features, especially those of melanoma, are in chapter 2, “superficial” and “superficial and deep”
usually absent. Occasionally one finds a few gray congenital nevi have a different architecture to Clark
dots (melanophages in the dermis), erythema (a sign nevi and are usually easy to distinguish from the latter
of inflammation) or peripheral radial lines occupying by histopathology.2 As we have described above, this
the whole circumference. is also true – with certain limitations – of dermatoscopy.
Very rarely there may be skin-colored or white reticu- While the reticular pattern is predominant in Clark
lar lines. It should be noted that grey dots and white nevus, the clod pattern is predominant in congenital
lines are also clues to melanoma. In these rare cases it nevi (3.60, 3.61). The clod pattern occurs either alone
may not be possible to confidently reach a diagnosis
of Clark nevus. In general, pattern and color, and 2 Regrettably, this distinction is not made by many dermatopathologists. Both
the symmetry of their combinations, should be given types of nevi are termed “dysplastic compound nevi” or “compound nevi”.
98 Pattern Analysis – Basic Principles
3 4 5
1 2
3 4
Figure 3.61: A patient with several “superficial and deep” congenital nevi.
On dermatoscopy one finds various patterns: (1) Structureless and brown (2) Reticular at the periphery, structureless in the center (3)
Reticular at the periphery, structureless in the center (4) Reticular (differential diagnosis: Clark nevus).
or in combination with other patterns, usually reticular with other patterns in both types of nevus. There also
or less often structureless. When clods are combined is a morphological zone of overlap with melanoma.
with the reticular pattern, the clods are usually found Clues to melanoma, especially gray dots and white
centrally and not, as in the (growing) Clark nevus, reticular lines, may be found in some congenital nevi.
peripherally. Occasionally, congenital nevi with exclu- Occasionally, histopathology is required to make this
sively reticular or curved lines may be found, usually distinction.
on the extremities. As in Clark nevus, combinations of
patterns are usually symmetrical in both “superficial” Correlation of dermatoscopy and dermatopathology
and “superficial and deep” congenital nevi. The histological correlate of reticular lines was explained
in the section on Clark nevi. Brown clods correspond to
A further clue to “superficial” and “superficial and nests of melanocytes at the dermo-epidermal junction,
deep” congenital nevi, especially the more common which are usually larger in congenital nevi than in the
clod or clod-reticular types is that they are either Clark nevus. Skin-colored clods arise due to lightly
uniformly brown or hypopigmented in the center. The pigmented or non-pigmented nests of melanocytes in
less common purely reticular types, on the other hand, the papillary dermis.
often have variegate pigmentation. The dermatoscop- The widened dermal papillae filled with melanocytes
ic presentation of “superficial” and “superficial and cause the epidermis to protrude outward, which gives
deep” congenital nevi is more protean than that of rise in metaphorical terminology of a cobblestone
Clark nevi (3.62). pattern. When the nests of melanocytes are somewhat
The most specific clue to the diagnosis of congenital deeper, i.e. below the dermal papillae, the surface of
nevus is terminal hairs, in greater numbers, or longer and the is seen as a skin-colored or light-brown structure-
darker, than on surrounding skin. This clue is, however, less area.
seen only in a minority of cases. Some “superficial” or
“superficial and deep” congenital nevi show dermato- Combined congenital nevi
scopic features of seborrheic keratosis, most often white Combined congenital nevi are those showing features
dots or clods, and occasionally orange clods between of both a “blue nevus” and either a “superficial” or
skin-colored clods. Occasionally there is also peri-infun- “superficial and deep” congenital nevus (3.63). The
dibular hyperpigmentation (brown circles around the dermatoscopy is exactly what one would expect from
infundibula). Other less specific dermatoscopic clues such a combination. In most cases there is a central
are clods or vessels as dots located in the center of blue structureless area (blue nevus), surrounded by
reticular lines, curved lines (primarily in combination brown reticular lines or brown clods (or both). If the
with reticular or branched lines), and densely arranged blue structureless area is located eccentrically rather
aggregations of small circles. than centrally, it is difficult to distinguish a combined
Occasionally it may be difficult to distinguish between a nevus from a melanoma. Occasionally, instead of a
Clark nevus and a congenital nevus on dermatoscopy, as blue structureless area one sees blue clods, which rarely
the reticular pattern may occur alone or in combination may be distributed over the entire lesion.
Pattern Analysis – Basic Principles 105
Correlation between dermatoscopy and the combination of relatively heavily pigmented nests
dermatopathology of melanocytes and acanthosis of the epidermis.
Refer to the above sections regarding blue nevi and
“superficial” or “superficial and deep” congenital nevi. Reed nevus
Serial dermatoscopic photography shows that an early
Recurrent nevus Reed nevus consists solely of dark-brown clods. The
On dermatoscopy one typically sees a hypopigment- characteristic pattern of radial lines or pseudopods
ed (lighter than surrounding skin) structureless zone, at the periphery only develops during subsequent
corresponding to the scar after excision (3.64). The growth (3.67). The radial lines or pseudopods are
recurrent nevus is within this area. Common patterns symmetrically distributed over the entire periphery
seen are peripheral radial lines, pseudopods, and brown while there is a black, black-gray or dark-brown
clods of different sizes. Radial lines and pseudopods structureless area in the center, or occasionally thick,
are of course clues to melanoma. In contrast to local gray reticular lines. Occasionally there are black dots
recurrence of a melanoma, the recurrent nevus usually or clods peripherally. Once growth ceases, the radial
does not extend beyond the scar. lines and pseudopods disappear. A Reed nevus is
then identical to a darkly pigmented Clark nevus with
Correlation between dermatoscopy and reticular lines peripherally and a structureless hyper-
dermatopathology pigmented center, or reticular lines only. One plausible
The radial lines and pseudopods correspond to fascicles theory suggests this is followed by transepidermal
of pigmented melanocytes at the dermo-epidermal elimination of melanocytes and the disappearance
junction. of the nevus. Combinations of patterns in Reed nevus
are usually symmetrical. If the pseudopods in a Reed
Spitz nevus nevus are only seen in some segments of the circum-
The “classical” Spitz nevus as described by Sophie ference, it cannot be distinguished from a melanoma
Spitz is non-pigmented or only lightly pigmented. dermatoscopically.
On dermatoscopy one most often finds skin-colored
or light-brown clods and perpendicular white lines Correlation between dermatoscopy and
(3.65). Alternatively when the lines seen between the dermatopathology
clods are lighter than the normal skin, they are termed The pseudopods and radial lines at the periphery are
white reticular lines. This pattern is also seen in some fascicles of pigmented melanocytes at the dermo-epi-
melanomas and dermatofibromas. In non-pigmented dermal junction that have spread centrifugally.
or lightly pigmented Spitz nevi one may find vessels
as dots. Blue Nevi
The patterns seen in pigmented Spitz nevi are brown Most blue nevi can be diagnosed easily. As we noted in
clods peripherally; centrally gray or blue-gray clods or chapter 2, the term “blue nevus” includes various entities
structureless (3.66 A, B). This central area is occasionally which can be distinguished by dermatopathology, but
interspersed with thick, light-gray reticular lines and/ not by dermatoscopy or clinical examination. As this
or polarizing-specific white lines. book is primarily focused on dermatoscopy, specific
Spitz nevi are nearly always easily distinguished from sub-classification will not be performed and the general
Reed nevi. Clinically Spitz nevi are nodular or papular term “blue nevus” will be used.
and Reed nevi are flat or only slightly raised. Derma- The dermatoscopic pattern of all blue nevi is struc-
toscopically, the patterns of established Reed nevi are tureless (3.68). Blue nevi usually have only one color,
pseudopods or radial lines. Only in the early stages of most commonly blue or gray. (When assessing color
growth may one see clods in Reed nevi. (3.66 C, D). one exercises latitude: slight variations in shade should
not be interpreted as a separate color.) Occasionally
Correlation between dermatoscopy and one sees variegate blue and gray. Less common again
dermatopathology are blue nevi with shades of gray and blue flanked by
Like the previously described nevi, brown clods corre- brown regions. It may then be difficult or even impos-
spond to pigmented melanocyte nests in the epidermis. sible to distinguish between this entity and a combined
White lines are most likely due to zones of fibrosis in congenital nevus on dermatoscopy.
the papillary dermis. The gray reticular lines in the Occasionally grey lines or dots may be seen against
center of pigmented Spitz nevi are probably due to a blue background. Applying the basic principle that
Pattern Analysis – Basic Principles 107
The same is true for melanomas on the face (lentigo Blue and gray structureless zones are caused by melanin
maligna in common nomenclature when they are in the dermis and/or orthohyperkeratosis (and in most
in situ, and lentigo maligna melanoma when they cases hypergranulosis as well) of the overlying epider-
have become invasive). Flat melanomas on the face mis. White structureless areas are caused by a zone of
often show the pattern of gray circles, or gray dots fibrosis in the dermis, which usually indicates regression.
arranged as circles, or angulated lines. Angulated Gray structures are produced by an accumulation of
lines (polygons) are also a specific clue for non-facial melanophages in the dermis. These melanophages may
flat melanomas on chronic sun damaged skin (3.76). be aggregated to form dots or clods, or be arranged
The polygonal geometric shapes formed by angulated in lines along the rete ridges, or in circles around hair
lines of non-facial lesions are larger than the holes follicles. Black dots or clods correspond to either nests
caused by individual follicular openings, whereas of melanocytes or accumulations of melanin in the
in facial lesions the angulated lines are framing the stratum corneum.
hypopigmented follicular openings. As in Reed nevus, peripheral pseudopods or radial
lines are caused by fascicles of melanocytes at the
Correlation between dermatoscopy and dermo-epidermal junction that have spread centrifugally.
dermatopathology White lines are a sign of fibrosis in the dermis. Thick
Histological correlates of the basic elements and the brown reticular lines correspond to widened rete ridges
colors of melanin have already been addressed. Here filled with pigmented atypical melanocytes.
we will only address some of the clues to melanoma. Parallel lines on the ridges are caused by a tenden-
The histological correlate of an eccentric structureless cy in acral melanoma for melanocytes to proliferate
zone varies according to its color. A black structureless along the crista profunda intermedia. Angulated lines
zone is caused by a dense accumulation of melanin in of facial lesions correspond to deposition of melanin
the epidermis, usually in the stratum corneum. Brown in the papillary dermis around follicular openings and
structureless zones are usually due to lentiginous arrange- proliferation of pigmented melanocytes in follicular
ments of pigmented melanocytes at the dermo-epidermal epithelium. The histopathological correlate of angulat-
junction. ed lines of non-facial lesions is not currently known.
However, this is only seen when the rete ridges are One plausible explanation is that they correspond to
flattened; if the rete ridges were intact there would be angiocentric deposition of melanin in proximity to the
reticular lines instead of the brown structureless zone. vessels of the superficial dermal plexus.
Pattern Analysis – Basic Principles 121
Solar lentigo, seborrheic keratosis and lichen planus like Squamous Cell Carcinoma
keratosis Rosendahl C, Cameron A, Argenziano G, Zalaudek I, Tschandl
Braun RP, Rabinovitz HS, Krischer J, Kreusch J, Oliviero M, P, Kittler H. Dermoscopy of squamous cell carcinoma and
Naldi L, Kopf AW, Saurat JH. Dermoscopy of pigmented keratoacanthoma. Arch Dermatol. 2012 Dec; 148(12):
seborrheic keratosis: a morphological study. Arch Der- 1386–92.
matol. 2002 Dec; 138(12): 1556–60. Zalaudek I, Giacomel J, Schmid K, Bondino S, Rosendahl C et
Zaballos P, Blazquez S, Puig S, Salsench E, Rodero J, Vives al. Dermatoscopy of facial actinic keratosis, intraepider-
JM, Malvehy J. Dermoscopic pattern of intermediate mal carcinoma, and invasive squamous cell carcinoma:
stage in seborrhoeic keratosis regressing to lichenoid a progression model. J Am Acad Dermatol. 2012 Apr;
keratosis: report of 24 cases. Br J Dermatol. 2007 Aug; 66(4): 589–97.
157(2): 266–72.
Clark Nevus, “superficial” and “superficial and deep”
Dermatofibroma congenital nevi
Zaballos P, Puig S, Llambrich A, Malvehy J. Dermoscopy of Clark WH Jr, Reimer RR, Greene M, Ainsworth AM, Mastran-
dermatofibromas: a prospective morphological study of gelo MJ. Origin of familial malignant melanomas from
412 cases. Arch Dermatol. 2008 Jan; 144(1): 75–83. heritable melanocytic lesions. ‘The B-K mole syndrome’.
Kilinc Karaarslan I, Gencoglan G, Akalin T, Ozdemir F. Arch Dermatol. 1978 May; 114(5): 732–8.
Different dermoscopic faces of dermatofibromas. J Am Kittler H, Tschandl P. Dysplastic nevus: why this term should
Acad Dermatol. 2007 Sep; 57(3): 401–6. be abandoned in dermatoscopy. Dermatol Clin. 2013
Oct; 31(4): 579–88
Ink-spot lentigo Rosendahl CO, Grant-Kels JM, Que SK. Dysplastic nevus:
Argenziano G. Dermoscopy of melanocytic hyperplasias: Fact and fiction. J Am Acad Dermatol. 2015 Sep; 73(3):
subpatterns of lentigines (ink spot). Arch Dermatol. 2004 507–12.
Jun; 140(6): 776. Hofmann-Wellenhof R, Blum A, Wolf IH, Zalaudek I, Piccolo
D, Kerl H, Garbe C, Soyer HP. Dermoscopic classification
Genital lentigo, Labial lentigo of Clark’s nevi (atypical melanocytic nevi). Clin Dermatol.
Blum A, Simionescu O, Argenziano G, Braun R, Cabo H, 2002 May-Jun; 20(3): 255–8.
et al. Dermoscopy of pigmented lesions of the mucosa Argenziano G, Zalaudek I, Ferrara G, Hofmann-Wellenhof R,
and the mucocutaneous junction: results of a multicenter Soyer HP. Proposal of a new classification system for mela-
study by the International Dermoscopy Society (IDS). nocytic naevi. Br J Dermatol. 2007 Aug; 157(2): 217–27.
Arch Dermatol. 2011 Oct; 147(10): 1181–7.
Pattern Analysis – Basic Principles 123
Reed Nevi
Bär M, Tschandl P, Kittler H. Differentiation of pigmented Spitz
nevi and Reed nevi by integration of dermatopathologic
and dermatoscopic findings. Dermatol Pract Concept.
2012 Jan 31; 2(1): 13–24.
Marchell R, Marghoob AA, Braun RP, Argenziano G. Der-
moscopy of pigmented Spitz and Reed nevi: the starburst
pattern. Arch Dermatol. 2005 Aug; 141(8): 1060.
Powered by TCPDF (www.tcpdf.org)
125
The classical language of dermatoscopy consists of terms that are poorly defined, of dubious significance,
a large number of mainly metaphoric terms with no obscure, or otherwise unnecessary. The consensus
over-arching structure. It qualifies as a technical lan- conference expert panel proposed a standardized
guage or “jargon” in the sense that it has a specific dictionary including both metaphoric and descriptive
vocabulary, which is incomprehensible outside its con- terms. (1) Although the authors of this book prefer
text. Although metaphors that are apt and colorful stick descriptive terminology we think that teachers of der-
in the memory, their sheer number and the fact that matoscopy should be familiar with both languages
many are ambiguous, redundant, or just bad analogies and should be able to teach both terminologies. The
make them a potential barrier to learning, teaching and aim of this chapter is to help those who are only
research. The metaphoric vocabulary of dermatoscopy familiar with metaphoric terminology. If you do prefer
has expanded so quickly that even experts find it difficult metaphoric terminology, we strongly encourage you
to oversee the plethora of terms (1). to select metaphoric terms that are included in the
In chapter 3 we introduced a simple descriptive termi- standardized dictionary.
nology based on only five geometrically defined basic Descriptive terms are not the definitions of the metaphoric
elements, which, like the letters of the alphabet, are the terms. The descriptive terms are used by those who prefer
building blocks of any new descriptive term. Because descriptive terminology over the metaphoric terms. The
of its simplicity and logic, this descriptive terminology majority of the terms describe features, which, on their
is becoming increasingly popular. A survey of Interna- own, are not very specific, but become meaningful in
tional Dermoscopy Society (IDS) members indicated the context of pattern and color. The method for assess-
that 23.5 % prefer to use descriptive terminology while ment of patterns, colors and clues, the core of pattern
20.1 % prefer metaphoric terminology. Most participants, analysis, is then described in chapter 5.
however, use both terminologies.
In 2015 the IDS initiated a new consensus conference “Angulated lines (polygons)”
with the primary aim of harmonizing metaphoric and Strictly speaking the term “angulated lines” (“polygons”)
descriptive terminology. Another goal was to rationalize is not a metaphoric term. It is composed of two parts;
metaphoric language by eliminating synonyms and one part is “line”, which is a basic element, and the
second part is “angulated”, which describes the spatial “rhomboids” (4), and “zig-zag pattern” (5) under the
arrangement. Originally, the term “polygon” was used umbrella term “angulated lines”.
to describe specific structures of flat melanomas on
non-facial chronic sun damaged skin (2, 3). Polygons “Annular-granular pattern”
were defined as geometric polygonal shapes complete The “annular-granular” pattern is regarded as a char-
or incomplete, bounded by straight lines, or by a straight acteristic feature of in situ melanoma (lentigo maligna)
pigment interface, meeting at angles and larger than on the face (6). It describes the arrangement of gray
the holes caused by individual follicles and larger by or brown dots (granular) around follicular openings
far than the holes bounded by reticular lines (4.1). (annular). The major aspect of this feature is actually
Throughout the book we use the term “angulated lines” the gray color, but this is not part of the term. The fea-
or “polygon” in a broader sense. We use it for straight ture is not particularly specific because it may occur
lines that do not intersect and which meet at angles in pigmented actinic keratoses (7) or lichen planus-like
in such a way that they form complete or incomplete keratoses as well. The equivalent term in the descriptive
polygonal shapes no matter if the skin involved is facial terminology is “gray dots arranged around follicular
or non-facial. We summarize the terms “polygon”, openings” (4.2).
Metaphoric dermatoscopic terms and what they mean 127
“Atypical (or irregular) pigment network” the hypopigmented intermediary spaces. An atypical
The terms “atypical” and “irregular” are subjective. (or irregular) pigment network is a clue to melanoma
According to the dictionary of standardized terms and so are thick reticular lines.
(8) an atypical network is defined as a network with
increased variability in the color, thickness, and spacing “Blotch”
of the lines of the network (4.3). In pattern analysis, The original meaning of “blotch” was a darkly pigmented
we make a distinction between structure and color. structureless area, but only used to describe melanocytic
We speak of eccentric hyperpigmentation if the darker lesions. “Irregular blotches” are a criterion of melanoma
shade is seen at the periphery the lesion. We use the in the 7-point checklist (9). “Irregular” means that several
terms “speckled” or “variegate” if the pigmentation is darkly pigmented structureless areas are irregularly
distributed in such a way that areas of dark pigmen- distributed. In pattern analysis, we make a distinction
tation alternate with areas of light pigmentation. If between color and structure. Structureless areas may
the network lines are broadened we call them thick assume any color and are then termed brown, black,
reticular lines as opposed to thin reticular lines. Thick blue, gray, white, or red structureless zones. A “blotch”
reticular lines are broader than or at least as broad as such as that shown in figure 4.4 would therefore be
for melanoma; compared to seborrheic keratoses or ing” they will be of the flat type, also known as solar
blue nevi it is not. The disadvantages of metaphoric lentigo . In pattern analysis the “brain-like pattern” can
language become evident here. Associative metaphoric be simply described using descriptive terms. We refer
terms may be catchy and easy to remember, but they to thick curved lines, clods and circles. Some “circles”
are also strongly linked to a specific diagnosis. The may be distorted into ellipses.
moment one refers to a blue-white veil, the association
with the diagnosis of melanoma is so strong that all “Branched streaks”
other differential diagnoses are not likely to be even Branched streaks are considered to be specific to mela-
considered. In pattern analysis, any structureless zone nocytic lesions (13). However, they are also found in
in an eccentric location, regardless of its color (except non-melanocytic lesions, such as ink-spot lentigo. In the
skin color), is a clue to melanoma. Thus, the term “blue- language of pattern analysis, they are simply described
white veil” can be replaced by the descriptive term “blue as branched lines.
structureless zone, eccentrically located”.
“Broadened pigment network”
“Brain-like pattern”, “cerebriform pattern”, A broadened pigment network (14) is found in the
“gyri and sulci” presence of melanoma, occasionally melanocytic nevi,
These are archetypal metaphoric terms signifying a seborrheic keratoses, and also ink-spot lentigo. In the
special arrangement of thick, curved, pigmented lines descriptive terminology the synonymous term is “thick
and clods and circles (4.7). This pattern is vaguely reticular lines”. It is a useful clue for in situ melanomas
reminiscent of the surface of a brain with its “gyri” and thin invasive melanomas (4.3).
(the hypopigmented spaces between the thick curved
lines) and “sulci” (hyperpigmented curved lines, clods “Central white patch”
and circles) (11). As a rule, lesions with this pattern are This is defined as a white structureless zone in the center
raised and not flat. of the lesion, which is quite specific for dermatofibroma
When the lesion is flat and the curved lines are not thick (15, 16) (4.8). This feature suggests the presence of
but thin and the circles small, the pattern is referred to a dermatofibroma, but not all dermatofibromas show
as “fingerprinting”(12) (see section on fingerprinting). this feature. The term “central white patch” actually
Both patterns are regarded as being quite specific for denotes two things: the first characteristic is a symmet-
seborrheic keratoses and their use generally causes rical arrangement of two patterns (reticular peripheral
other differential diagnoses to be discarded. In cases of and structureless “central”); the second characteristic is
“brain-like whorls” a seborrheic keratosis will be of the the center of the pattern showing a white structureless
markedly acanthotic type and in cases of “fingerprint- zone (“white patch”).
130 Metaphoric dermatoscopic terms and what they mean
“Cobblestone pattern”
This is a metaphoric term (8) for a pattern of large,
polygonal clods similar to cobblestones (4.10). This pat-
tern is supposed to evoke the impression of a primarily
dermal nevus of the Unna or Miescher type. However,
one may find it in the presence of other congenital nevi
Figure 4.8: “Central white patch”.
Structureless white zone (“central white patch”) located in the
as well, and occasionally even in seborrheic keratoses.
center of a dermatofibroma. The metaphoric term “cobblestone pattern” may be
simply substituted by a description of the pattern. These
are large, polygonal, skin-colored or light-brown clods.
“Chrysalis, Chrysalids, and Crystalline”
The term “chrysalis” was used to describe a pattern of “Comedo-like openings”
straight white lines at right angles to each other seen This is not a descriptive term but an interpretation.
only when a polarizing dermatoscope is used (17). Comedo-like openings are dilated, keratin-filled infun-
Chrysalis is named after a vague resemblance of the dibula of a seborrheic keratosis (4.11). As the keratin is
structure to a wax moth infestation of a beehive. Figure usually contaminated by melanin or exogenous impu-
4.9 shows an example of the structure of white lines rities and affected by oxygenation, colors seen are
in a thick melanoma which has grown rapidly. It can brown, yellow or orange. Comedo-like openings are
be seen that the white lines aligned at right angles are usually seen as clods. Less often they appear as dots
certainly brighter with polarized dermatoscopy. In the or – because the pigment is most dense at the margin
descriptive language these structures are referred to of the keratin plug – as circles. As the term “comedo-like
as “polarizing-specific white lines” or “perpendicular openings” already suggests a diagnosis (8), it is not
white lines”. This pattern is seen in melanoma, Spitz used in pattern analysis. It is better to finish describing
nevus, basal cell carcinoma, and dermatofibroma. a lesion before moving on to interpret these findings
Sometimes the terms “chrysalids” and “crystalline” and reach a diagnosis.
Figure 4.9: Non-polarized (left) and polarized (right) dermatoscopic view of an invasive melanoma (> 1 mm). White lines are seen as a clue
to malignancy in both images but they are seen to be brighter, and perpendicular orientation is more evident, in the polarized image. One
can also see so-called “shiny white blotches and strands” which correspond to white structureless zones and white clods.
Metaphoric dermatoscopic terms and what they mean 131
“Fat fingers”
“Fat fingers” are similar to “crypts”, being another met-
aphoric term for thick curved lines (4.14). However, the
difference between this entity and “crypts” and “sulci” is
that “fat fingers” does not refer to the pigmented curved
lines themselves but the intervening hypopigmented
spaces (12). Like “crypts” and “sulci”, these structures
are mainly found in seborrheic keratoses.
“Fibrillar pattern”
“Fibrillar”(21, 22) means “consisting of fibrils”; fibrils are
very thin fibers. Regrettably, the term does not convey the
principal characteristic of this pattern. According to the
Figure 4.14: “Fat fingers”.
“Fat fingers” (arrows) refer to the broad, hypopigmented inter-
standardized dictionary it consists of “linear pigmented
vening spaces between brown clods, or the thick curved lines of filamentous lines of similar length with one end at the
raised seborrheic keratoses. furrows and oriented at a certain angle to the furrows
and crossing the ridges” (4.15). Because the spatial
“Crown vessels” arrangement of the parallel lines is the characteristic
The metaphoric term “crown vessels” (20) is used for feature of this pattern and not that it is composed of
radial, serpentine or branched vessels at the periphery “fibrils”, the descriptive term for this pattern is “parallel
of the lesion that radiate towards the center but do not lines crossing the ridges”.
cross the midline of the lesion. It is a common finding
of sebaceous hyperplasia and helps to differentiate it “Fingerprinting”
from basal cell carcinoma (4.12). The flat initial stage in the evolution of seborrheic ker-
atoses, also known as solar lentigo, may (especially
“Crypts” on the trunk) show a pattern consisting of long, thin,
Crypts are defined in medicine as small pits or glandular curved lines that are partly arranged in parallel fashion
cavities. In dermatoscopy this term is used to describe (4.16). Together with a few interspersed circles these
invaginations of the epidermis filled with keratin and lines are vaguely reminiscent of dermatoglyphs; hence
melanin (8). It roughly corresponds to the “sulci” of the “fingerprinting”. Describing this pattern as an accumu-
“brain-like pattern” (see the section on this term) of some lation of long curved lines serves the same purpose as
seborrheic keratosis (4.13). In descriptive terminology using the metaphoric term “fingerprinting”, without the
these structures are thick curved lines or elongated clods. obligatory inference that the lesion is a solar lentigo.
Metaphoric dermatoscopic terms and what they mean 133
“Fissures and ridges” pists that similarly shaped structures are given different
“Fissures and ridges” (11) are merely other expressions names to describe their geometry, merely because they
of the previously mentioned “gyri” (i.e. ridges) and are a different color. Aggregated globules (aggregat-
“sulci” (i.e. fissures). ed brown clods) are, in most algorithms, taken to be
indicative of a melanocytic lesion. However, this clue
“Globules” is not very specific as non-melanocytic lesions such as
“Globules” or “Globuli” in common terminology (8) basal cell carcinoma or seborrheic keratoses may also
are small to middle-sized, round or oval, brown clods have brown clods.
(4.17). The term “clod” in pattern analysis is used to
describe any round, oval or polygonal well circum- “Homogeneous pattern”
scribed structure, of any color. The term “homogenous” (8) is often used instead of
Thus, round or oval brown clods known as “globules” in structureless. We do not encourage that. Nothing in
metaphoric language are simply one small sub-group of dermatoscopy is purely homogenous. Even blue nevi
clods. We consider it confusing to novice dermatosco- are not homogenously blue (4.18). Structureless is
134 Metaphoric dermatoscopic terms and what they mean
“Peppering”
The pattern of gray dots on a white structureless back-
ground is known as peppering in metaphoric language
(6) (4.25). The white structureless zone corresponds to
fibrosis of the dermis, interspersed with numerous mela-
nophages (gray dots or small gray clods). Sometimes the
term “regression” is used synonymously, but “peppering”
is not always due to partial or complete regression
of a pre-existing melanocytic lesion. Especially when
the white structureless zone is absent, one should be
cautious in interpreting the gray dots as regression.
Figure 4.23: “Milky red areas”. Like all metaphoric terms, this one is also replaceable
A melanoma with “milky red areas”, which is called pink struc- with descriptive terminology (white structureless zone
tureless zone in the descriptive terminology. with gray dots and/or clods).
also occur in congenital nevi and basal cell carcinoma “Peripheral streaks” or “irregular peripheral exten-
as well as less frequently in many other lesions, including sions”
melanoma. On dermatoscopy, milia-like cysts are seen These terms refer to radial lines or pseudopods at a
as white or yellow dots and/or clods. lesion’s periphery, but only occupying part of the cir-
cumference (4.26). This feature should cause one to
“Milky red areas” suspect a melanoma. However, occasionally Reed nevi
According to the consensus paper (1) the term “milky and recurrent nevi also have radial lines which are not
red area” is used to characterize “a red vascular blush distributed over the entire circumference, and basal cell
with no specific distinguishable vessels”. In descriptive carcinomas may also have radial lines.
terms we simply call it a pink structureless area (4.23).
It is of some value for the diagnosis of hypo- or non-pig- Pigment network
mented types of melanoma. The term “pigment network” refers to reticular lines.
In contrast to the widespread view, the presence of a
“Moth-eaten border” so-called pigment network is not unique to melanocytic
The sharp border with concave or sharp punched-out lesions. Reticular lines are also found in solar lentigines,
invaginations that is frequently found in flat seborrheic seborrheic keratoses, melanotic macules (especially
“Rainbow pattern” are “red puddles”. It is simpler to call them red clods
The rainbow pattern is defined as “circumscribed struc- (4.29). A collection of red clods is primarily found in
tureless areas displaying colors of the whole spec- vascular lesions, especially hemangiomas and recent
trum of visible light”. In the descriptive terminology hemorrhages.
we call it polychromatic structureless zone. Initially it However, red clods may rarely be found in melanoma
was described as a specific clue to Kaposi sarcoma – in which case the advocates of metaphoric language
(26) but later it was found out that it can be found in call them “milky red globules” rather than “red lacunes”.
other diagnosis too (for example in dermatofibroma or This is a further instance of how the diagnosis can alter
vascular lesions (4.28) (27). the description.
Figure 4.31: The structures known in metaphoric terminology as “rosettes” are clearly seen as 4 white dots arranged in a square (4 dot
clod), but only on polarized dermatoscopy (left). With non-polarized dermatoscopy (right) they are simply seen as a white clod.
known as a “central white patch” (see the respective (4.9). If lines are present they are usually oriented
sections). This is a further instance of the diagnosis perpendicular to each other. These structures are seen
influencing the description. In descriptive terminology only under polarized dermatoscopy and can be found
we simply refer to either white structureless zones or in melanoma, basal cell carcinoma, Spitz nevus and
white lines, as appropriate. dermatofibroma.
central structure. Unfortunately, in many other cases (4.34). This pattern is mainly found in Reed nevi and in
they bear little similarity to spoke wheels. Usually the those rare cases in which a melanoma mimics a Reed
radial lines only partly surround the central structure. nevus. Occasionally the term “starburst pattern” is also
Another variant consists of a darkly pigmented dot in used for a combination of clods at the periphery and a
the center of a less heavily pigmented clod, reminiscent structureless hyperpigmented center. This pattern is more
of a wheel, but without spokes. In any case the term commonly found in the pigmented Spitz nevus. When
“spoke wheel” is dispensable because these entities can two different patterns bear the same name, it is quite
also be described by the use of descriptive terminology. natural for the two different diagnoses – Reed nevus
and pigmented Spitz nevus – to be lumped together.
“Starburst pattern”
The “starburst pattern” (8) is the combination of patterns “Strawberry pattern”
consisting of a structureless zone centrally, with radial In metaphorical terminology this refers to a pattern
lines or pseudopods occupying the entire periphery which is sometimes seen in actinic keratoses in which
“String of pearls”
The metaphor of “string of pearls” is used for coiled
vessels that are arranged in serpentine lines (30). In the
descriptive terminology we use the term “serpiginous”
for this specific arrangement of vessels. It is rather
specific for clear cell acanthoma although it has been
described in other lesions too, most notably in prurigo
or lichen simplex chronicus (4.36).
Tabelle 4.1: Translation of metaphoric language into simple terminology based on five simple geometric terms
(five basic elements)
Metaphoric terminology Descriptive terminology
Annular-granular pattern Dots, gray and circles, gray
Atypical pigment network Lines, reticular and thick or reticular lines that vary in color
Blotch Structureless zone, brown or black
Blue-gray ovoid nests Clods, blue, large, clustered
Blue-whitish veil, blue veil Structureless zone, blue
Branched streaks Lines, branched
Broadened network Lines, reticular and thick
Central white patch Structureless zone, white, central
Cerebriform pattern, fissures and ridges, Lines, curved and thick
gyri and sulci, fat fingers
Cobblestone pattern Clods, brown or skin colored, large and polygonal
Comedo-like openings Clods, brown, yellow, or orange (rarely black)
Crown vessels Radial linear vessels, not crossing the center
Crypts Lines, curved and thick, in combination with clods
Delicate network Lines, reticular and thin
Fibrillar pattern Lines, parallel, short, crossing ridges (volar skin)
Fingerprinting Lines, brown, curved, parallel, thin
Globules Clods, small, round or oval
Granularity or granules Dots, any color
Homogenous pattern Structureless, any color
Lattice-like pattern (volar skin) Lines, parallel, thin, in the furrows and crossing the ridges
Leaf like areas Lines, radial, connected to a common base (sometimes variously shaped clods have
been called “leaf like areas”)
Milia like cysts, cloudy or starry Dots or clods, white, clustered or disseminated
Milky red areas Structureless zone, pink
Milky red globules Clods, pink and small
Moth eaten border Sharply demarcated, scalloped border
Negative pigment network (synonyms: Lines, reticular, hypopigmented, around brown clods
inverse network, reticular depigmentation)
Peppering Dots, gray
Pigment network Lines, reticular
Pseudo-network Structureless, brown, interrupted by follicular openings (facial-skin)
Radial streaming Lines, radial, peripheral and segmental
Rainbow pattern Structureless zone, polychromatic
Red lacunes Clods, red or purple
Rhomboids Lines, angulated (facial skin)
Rosettes Dots, white, four arranged in a square, 4-dot clod
Scar-like depigmentation Structureless zone, white
Shiny white blotches and strands Clods, white
Shiny white streaks (synonyms: chrysalis, Lines, white, perpendicular
chrysalids, crystalline structure)
Spoke wheel area Lines, radial, converging to a central dot or clod
Starburst pattern Pseudopods, circumferential or lines, radial, circumferential
Strawberry pattern Structureless, red, interrupted by follicular openings
Streaks Lines, radial (always at periphery)
String of pearls Coiled vessels arranged in serpentine lines
Targetoid dots Dots, brown, central (in the center of hypopigmented spaces between reticular lines)
Zig-zag pattern Lines, angulated (facial skin)
144 Metaphoric dermatoscopic terms and what they mean
References
1 Kittler H, Marghoob AA, Argenziano G, et al. Standard- 14 Soyer HP, Kenet RO, Wolf IH, Kenet BJ, Cerroni L. Clin-
ization of terminology in dermoscopy/dermatoscopy: icopathological correlation of pigmented skin lesions
Results of the third consensus conference of the Inter- using dermoscopy. Eur J Dermatol. Jan-Feb 2000; 10(1):
national Society of Dermoscopy. J Am Acad Dermatol. 22–28.
2016; 74(6): 1093–106. 15 Agero AL, Taliercio S, Dusza SW, Salaro C, Chu P,
2 Jaimes N, Marghoob AA, Rabinovitz H, et al. Clinical and Marghoob AA. Conventional and polarized dermoscopy
dermoscopic characteristics of melanomas on nonfacial features of dermatofibroma. Arch Dermatol. Nov 2006;
chronically sun-damaged skin. J Am Acad Dermatol. Jun 142(11): 1431–1437.
2015; 72(6): 1027–1035. 16 Zaballos P, Puig S, Llambrich A, Malvehy J. Dermoscopy
3 Keir J. Dermatoscopic features of cutaneous non-facial of dermatofibromas: a prospective morphological study
non-acral lentiginous growth pattern melanomas. Der- of 412 cases. Arch Dermatol. Jan 2008; 144(1): 75–83.
matol Pract Concept. Jan 2014; 4(1): 77–82. 17 Marghoob AA, Cowell L, Kopf AW, Scope A. Observation
4 Schiffner R, Schiffner-Rohe J, Vogt T, et al. Improvement of of chrysalis structures with polarized dermoscopy. Arch
early recognition of lentigo maligna using dermatoscopy. Dermatol. May 2009; 145(5): 618.
J Am Acad Dermatol. Jan 2000; 42(1 Pt 1): 25–32. 18 Balagula Y, Braun RP, Rabinovitz HS, et al. The signifi-
5 Slutsky JB, Marghoob AA. The zig-zag pattern of lentigo cance of crystalline/chrysalis structures in the diagnosis
maligna. Arch Dermatol. Dec 2010; 146(12): 1444. of melanocytic and nonmelanocytic lesions. J Am Acad
6 Stolz W, Schiffner R, Burgdorf WH. Dermatoscopy for Dermatol. Aug 2012; 67(2): 194 e191–198.
facial pigmented skin lesions. Clin Dermatol. May-Jun 19 Di Stefani A, Campbell TM, Malvehy J, Massone C,
2002; 20(3): 276–278. Soyer HP, Hofmann-Wellenhof R. Shiny white streaks: An
7 Tschandl P, Rosendahl C, Kittler H. Dermatoscopy of flat additional dermoscopic finding in melanomas viewed
pigmented facial lesions. J Eur Acad Dermatol Venereol. using contact polarised dermoscopy. Australas J Dermatol.
Jan 2015; 29(1): 120–127. Nov 2010; 51(4): 295–298.
8 Argenziano G, Soyer HP, Chimenti S, et al. Dermoscopy 20 Argenziano G, Zalaudek I, Corona R, et al. Vascular
of pigmented skin lesions: results of a consensus meeting structures in skin tumors: a dermoscopy study. Arch
via the Internet. J Am Acad Dermatol. May 2003; 48(5): Dermatol. Dec 2004; 140(12): 1485–1489.
679–693. 21 Malvehy J, Puig S. Dermoscopic patterns of benign volar
9 Argenziano G, Fabbrocini G, Carli P, De Giorgi V, melanocytic lesions in patients with atypical mole syn-
Sammarco E, Delfino M. Epiluminescence microscopy drome. Arch Dermatol. May 2004; 140(5): 538–544.
for the diagnosis of doubtful melanocytic skin lesions. 22 Saida T, Oguchi S, Ishihara Y. In vivo observation of
Comparison of the ABCD rule of dermatoscopy and a magnified features of pigmented lesions on volar skin
new 7-point checklist based on pattern analysis. Arch using video macroscope. Usefulness of epiluminescence
Dermatol. Dec 1998; 134(12): 1563–1570. techniques in clinical diagnosis. Arch Dermatol. Mar
10 Menzies SW, Ingvar C, McCarthy WH. A sensitivity and 1995; 131(3): 298–304.
specificity analysis of the surface microscopy features 23. Pizzichetta MA, Talamini R, Marghoob AA, et al. Nega-
of invasive melanoma. Melanoma Res. Feb 1996; 6(1): tive pigment network: an additional dermoscopic feature
55–62. for the diagnosis of melanoma. J Am Acad Dermatol.
11 Braun RP, Rabinovitz HS, Krischer J, et al. Dermoscopy of Apr 2013; 68(4): 552–559.
pigmented seborrheic keratosis: a morphological study. 24 Botella-Estrada R, Requena C, Traves V, Nagore E, Guillen
Arch Dermatol. Dec 2002; 138(12): 1556–1560. C. Chrysalis and negative pigment network in Spitz nevi.
12 Kopf AW, Rabinovitz H, Marghoob A, et al. “Fat fingers:” Am J Dermatopathol. Apr 2012; 34(2): 188–191.
a clue in the dermoscopic diagnosis of seborrheic kerato- 25 Tschandl P, Rosendahl C, Kittler H. Accuracy of the first
ses. J Am Acad Dermatol. Dec 2006; 55(6): 1089–1091. step of the dermatoscopic 2-step algorithm for pigmented
13 Pizzichetta MA, Argenziano G, Talamini R, et al. Dermo- skin lesions. Dermatol Pract Concept. Jul 2012; 2(3):
scopic criteria for melanoma in situ are similar to those 203a208.
for early invasive melanoma. Cancer. Mar 1 2001; 26 Cheng ST, Ke CL, Lee CH, Wu CS, Chen GS, Hu SC.
91(5): 992–997. Rainbow pattern in Kaposi’s sarcoma under polarized
dermoscopy: a dermoscopic pathological study. Br J
Dermatol. Apr 2009; 160(4): 801–809.
Metaphoric dermatoscopic terms and what they mean 145
We now come to the core of the method, making a spe- and so create their own style. Pattern analysis is the
cific diagnosis. Pigmented skin lesions can be described framework for developing this personal algorithm.
very clearly and reliably using the method (patterns, In every algorithmic method one makes decisions,
colors and clues) described in the previous chapters. An which progressively reduce the number of differential
exact morphological description is like a thread winding diagnoses. In formulating the algorithmic method of
through a labyrinth; the thread leads a wandering or pattern analysis, we attempted to fulfill the following
disoriented clinician safely to the outcome or exit. The basic principles: The criteria used to make decisions
essence of pattern analysis is a structured description must be clearly and unambiguously defined. The criteria
formulated using a clearly defined algorithmic method. should be exclusive, i.e. the properties being assessed
The diagnostic method is structured in such a way that do not overlap, so unequivocal classification is possible.
one starts by describing the most general of features, Decisions should be so simple that even beginners are
then proceeds progressively to finish with the most able to perform the task. Decisions should show a high
specific features. Rules prescribe, at each turn, the degree of concordance when judged by a range of
direction one should take so that the clinician is not clinicians. This decision process should be carried as
misled, as happens easily when using a method where far as possible – but no further!
descriptive terms are subjective, poorly defined, or The first step is to decide whether a pigmented lesion
dependent on diagnosis. Only after a comprehensive is composed of one or more than one pattern.
description and after all observable data have been
taken into account are the findings interpreted and a
specific diagnosis established. 5.1 One pattern
The algorithm always takes the general form: The first step in pattern analysis of a pigmented lesion
is to decide: is there one pattern or more than one
Pattern + Color + Clues = Diagnosis pattern? This decision is nearly always both simple
and unequivocal. When a pigmented lesion consists
The algorithmic method will be presented here in a of just one pattern, then of course the next question
stepwise manner. When you are confronted with an is, which pattern? As we know, a pattern is formed
unknown pigmented lesion you may use the following by an aggregation of one of the five basic elements
pages as a classification reference guide. Using patterns, (lines, pseudopods, circles, clods and dots, 5.1). When
colors and clues, the number of potential diagnoses is basic elements are not seen or there are too few
progressively minimized. Finally just one or a few diag- basic elements to constitute a pattern, the “pattern”
noses remain. In those cases in which it is not possible is termed structureless. When evaluating patterns,
to reach a confident specific diagnosis, the degree of individual dots or clods are not important; at this stage
doubt and the type of possible diagnoses will determine the general impression takes precedence. Once this
whether histopathology is required. decision is made, we progress through the algorithm
Pattern analysis is not an algorithm carved in stone, in a stepwise manner, which divides into smaller and
it is a method by which algorithms are constructed. smaller branches with a progressively smaller differ-
With experience, every investigator will develop their ential diagnosis.
own individual algorithm. Only then does one become
an expert. General criteria, guidelines and concepts 5.1.1 Lines
are certainly required, but personal experience is also Lines may form six different patterns, created by dif-
crucial. Beginners must rely on recipes, but experts ferences both in the form of individual lines, and the
personalize the recipe, refine and develop it further, arrangement of the lines relative to one another. These
148 An algorithmic method for the diagnosis of pigmented lesions
Reticular
Branched
Lines
Angulated
Pseudopods
Parallel
Circles
Radial
One pattern
Clods Curved
Dots
Structureless
> 1 color
patterns are reticular, branched, angulated, parallel, every shade of brown is to be interpreted as a separate
radial, and curved. color. One or two black lines do not render the lesion
multicolored. The normal hypopigmentation around
Reticular lines follicular openings does not create an extra color.
Lesions that consist exclusively of reticular lines are Nearly all pigmented lesions are somewhat lighter at
extremely common. Although many algorithms use the periphery than in the center; again this does not
reticular lines as a criterion to diagnose lesions as constitute an additional color.
melanocytic (1), lesions with reticular lines are not If there is only one color, namely light-brown, and the
always melanocytic (2). In most cases the histological lesion consists of thin reticular lines, the diagnosis is
correlate of reticular lines is hyperpigmentation of basal either junctional Clark nevus or solar lentigo (5.3).
keratinocytes on rete ridges, which may or may not be A Clark nevus is round or oval and the pigmentation
created by an increase in the numbers of melanocytes. does not end abruptly at the periphery. In contrast, the
As the next step the investigator assesses color (5.2). As border of a solar lentigo is usually sharply demarcated
melanin appears brown in the epidermis, reticular lines and scalloped. A few brown dots may be found in
are usually light-brown or dark-brown. If the pigment both diagnoses, but more frequently in Clark nevus.
is very dense, the lines are black. Rarely reticular lines It is not always possible to make a reliable distinction
are gray. Beginners are often too strict in assessing on dermatoscopy, but as both lesions are benign this
color and therefore tend to see too many colors. Not is seldom crucial. Rare differential diagnoses for thin,
An algorithmic method for the diagnosis of pigmented lesions 149
light-brown reticular lines on dermatoscopy are a variant clues are abrupt ending of lines within the lesion and
of dermatofibroma (3) with reticular lines only (5.4) and a sharply demarcated border. Normally no differential
urticaria pigmentosa (4), a type of mastocytosis (5.5). diagnosis needs to be considered. Very rarely a Reed
When brown reticular lines are thick and not thin, nevus may demonstrate this pattern and color combina-
one should first consider a Clark nevus or less often tion, but without the additional clues to “ink-spot lentigo”.
a superficial congenital nevus (5.6). A solar lentigo is For a lesion with only reticular lines but more than
very unlikely. A seborrheic keratosis may present with one color, one first should exclude a solar lentigo or
thick, brown, reticular lines, but in this case one nearly seborrheic keratosis. This is done best by considering
always sees reticular lines in combination with other the clues to solar lentigo (well-demarcated, scalloped
characteristic features of seborrheic keratosis. border) and seborrheic keratosis (white dots or clods,
Black or at least very dark-brown reticular lines are a orange or yellow clods, well-demarcated border, circles,
clue to the diagnosis of ink-spot lentigo (5.7). Additional thick curved lines, vessels as loops or coils). Once a
An algorithmic method for the diagnosis of pigmented lesions 151
solar lentigo or a seborrheic keratosis has been ruled is central hyperpigmentation, i.e. light-brown reticular
out, three differential diagnoses should be considered: lines peripherally and dark-brown or even black lines
a) Clark nevus, b) “superficial” or “superficial and in the center (5.9). This pattern is typical of a Clark
deep” congenital nevus, and c) in situ melanoma (5.8). nevus. All other diagnoses may be safely ruled out.
One proceeds in the usual stepwise manner. The colors When dark-brown or black and light-brown areas are
and their distribution are assessed before the final step, present alternately so that one obtains the impression
resolving the differential diagnosis using clues. For a of a speckled lesion, this type of color distribution is
pattern of reticular lines, only the colors light-brown, termed variegate. The differential diagnosis for a var-
dark-brown, black and very rarely gray will be seen. iegate reticular lesion is: Clark nevus, “superficial” or
In practice, there are only three ways that two colors “superficial and deep” congenital nevus, or an in situ
combine in reticular lesions (in theory, of course, there is melanoma (5.10). While the distinction between a Clark
an infinite number of combinations). The first possibility nevus and a congenital nevus is purely of academic
152 An algorithmic method for the diagnosis of pigmented lesions
1 color
Reticular
Figure 5.8: Continuation of the algorithm for the reticular pattern, when more than one color is present
Figure 5.9: One pattern, lines, reticular, more than one color, central hyperpigmentation.
This pattern and color combination is the typical dermatoscopic appearance of the Clark nevus. The six Clark nevi seen here are all varia-
tions on this pattern.
interest, the differentiation between these and an in situ The third and last color combination seen in reticular
melanoma is of course very significant, and is based pattern lesions is eccentric hyperpigmentation, i.e. the
on the presence or absence of the clues to melanoma more heavily pigmented area is peripheral, not central.
outlined in chapter 3. Only 5 of the 9 clues to melano- This color combination is found in both Clark nevi and
ma are seen in reticular pattern lesions: a) gray dots, in situ melanomas (5.11), but only rarely in congenital
clods, circles or lines; b) radial lines or pseudopods nevi. As in the case of variegate pigmentation, the dif-
seen only in some segments of the periphery; c) black ferential diagnosis is resolved by assessing the lesion
dots or clods at the periphery d) thick reticular lines for clues to melanoma. When no clue to melanoma
and e) angulated lines (polygons). When one of these is present, a Clark nevus is the most likely diagnosis.
clues is present, the diagnosis of melanoma should be A common difficulty is how one should proceed when
seriously considered. the overall assessment shows a symmetrical pattern and
An algorithmic method for the diagnosis of pigmented lesions 153
Figure 5.10: One pattern, lines, reticular, more than one color, variegate.
The first and the second row show Clark nevi and “superficial” or “superficial and deep” congenital nevi with a reticular pattern and
variegate pigmentation. These two types of nevus can be distinguished from each other only on histopathology; the distinction is purely of
academic interest. Clues to melanoma are not seen in any of these lesions. The lesions shown in the third row are in situ melanomas with
thick reticular lines (left and middle) or gray dots and small gray clods (right, at 11 o’clock position) as clues to melanoma.
color combination, but a clue to melanoma is present, nevi). If, on the other hand, there is only a single reticular
e.g. gray dots. This is a situation where experts will make lesion with gray structures the clue should be given more
better decisions than beginners, as the strength of the weight. This helps to increase sensitivity (to detect more
clue must be weighed against one’s level of certainty that melanomas). The management of patients with multiple
the lesion is in fact symmetrical. As a general principle, nevi will be discussed in more detail in chapter 9.
symmetry of pattern and color should be given greater
weight than the clue; in short, “pattern trumps clues”. Branched lines
Nevertheless, some of these lesions must be submitted Branched lines and reticular lines are closely related
for histopathology to confidently exclude malignancy. and often occur together. Sometimes they are difficult to
A clue should also be weighed differently depending distinguish, in which case one should analyze accord-
on the number of lesions with similar features in the ing to the much more common reticular pattern. There
same patient. For example, some patients have multi- are, however, lesions that are exclusively composed
ple reticular lesions with gray dots or gray lines. In this of branched lines. In practice, these lesions are either
context, gray structures have a lower weight as a clue to black or brown, and they are all benign. Lesions that
malignancy. This “comparative approach” (5) helps to have only brown branched lines are either a Clark nevus
increase specificity (to reduce the number of excisions of or a “superficial” or “superficial and deep” congenital
154 An algorithmic method for the diagnosis of pigmented lesions
Figure 5.11: One pattern, lines, reticular, more than one color, peripheral hyperpigmentation.
This color and pattern combination is found in Clark nevi (top row) and in situ melanomas (bottom row). Clark nevi usually have no clues to
melanoma. The melanoma bottom left has thick reticular lines; the melanoma bottom right has peripheral black dots (at 3 o’clock position)
as clues to melanoma.
nevus. The branched lines are probably columns of lines is one of the most specific clues to the diagnosis
melanocytes at the base of rete ridges, which appear of melanoma. On facial skin, however, angulated lines
as nests in the vertical plane of the histopathological are also seen in pigmented actinic keratoses (8, 9).
specimen. Black (or very dark-brown) branched lines Most melanomas with angulated lines are in situ (not
indicate an “ink-spot” lentigo. Here again the pigmen- invasive). The lines are usually brown or gray. Many
tation is at the base of the rete ridges but it is in basal melanomas with angulated lines also have gray dots,
keratinocytes, not in melanocytes as in Clark nevus or but usually too few to be called a pattern. Close inspec-
in superficial congenital nevus. tion of angulated lines may show them to be formed
by densely packed gray dots.
Angulated lines
Angulated lines are the hallmark of flat melanomas Parallel lines
on skin with chronic sun damage, on both facial and The pattern of parallel lines is the typical pigment
non-facial skin (6, 7) (5.12). They often appear in pattern of acral skin. Parallel lines may be arranged
conjunction with another pattern, most often with the in one of three ways; on the ridges (ridge pattern), in
reticular pattern on non-facial skin and with circles on the furrows (furrow pattern), or crossing ridges and
facial skin. On non-facial skin, a pattern of angulated furrows (crossing pattern). Acral lesions that only show
An algorithmic method for the diagnosis of pigmented lesions 155
Crossing ridges
Acral nevus, classical
a furrow pattern or a crossing pattern, and without any bottom right). Satellite clods are a strong clue to the
of the clues to melanoma, may be safely considered to diagnosis of hemorrhage. As both hemorrhage and
be benign (10). These are either classical acral nevi or exogenous pigmentation are found in the stratum corne-
small “superficial” or “superficial and deep” congenital um (i.e. superficially) one can remove the pigmentation
nevi (5.13). by careful paring with a scalpel. This is, of course, not
Assessment of a lesion showing the ridge pattern pro- possible with a melanocytic lesion.
ceeds in the normal stepwise fashion by evaluating color.
If it is brown, in situ melanoma must be considered. Radial lines
Occasionally, acral nevi may also have a ridge pat- As radial lines always occur in combination with another
tern; the same is true for acral lentigines or melanotic pattern, they are discussed in the section dedicated to
macules related to the person’s ethnic origin or found lesions with more than one pattern.
as part of rare diseases such as the Laugier-Hunziker
syndrome (11). However, a biopsy is usually required Curved lines
to confirm such a benign diagnosis. Curved lines usually occur in combination with other
Black, red or purple parallel lines on the ridges indicate patterns. However, some solar lentigines or seborrheic
either hemorrhage or exogenous pigmentation (5.14 keratoses may have only curved lines (5.15). In these
156 An algorithmic method for the diagnosis of pigmented lesions
cases one often finds clues like a curved, sharply demar- circles are formed by melanin pigment arranged either
cated border, or a few circles, to support the diagnosis around the openings of the crater-like infundibula or in
of solar lentigo or seborrheic keratosis. infundibular epithelium. The center of the infundibulum
appears hypopigmented. If infundibula contain kera-
5.1.2 Pseudopods tinized material rather than a hair-shaft, (which is not
Like radial lines, pseudopods occur only in combination unusual) the hypopigmented center is seen as yellow or
with other patterns and are also discussed in the section orange clods. Facial circles, especially when they are
dedicated to lesions with more than one pattern. broad and confluent, are often seen on close inspec-
tion to be small dots arranged as circles around the
5.1.3 Circles openings of the infundibula. On facial skin it is crucial
Just as the pattern of parallel lines is the pattern of acral to differentiate between a pattern of circles (or dots
skin, the pattern of circles is the pattern of facial skin. arranged as circles) formed by pigment, and gaps in
In contrast to parallel lines, the pattern of circles is not other patterns created by the infundibula, as this is of
unusual at other locations on the body. On the face, great diagnostic significance. This is most commonly
An algorithmic method for the diagnosis of pigmented lesions 157
A B
C D
an issue with the pattern of dots and the structureless facial melanoma in situ but usually there is also another
pattern (5.16). clue present. Gray circles on facial lesions indicate
Circles (or dots arranged as circles) may be brown or melanoma in situ. Gray dots arranged in circles, even
gray (5.17). Brown circles not associated with hair folli- when the gray color is only present in some parts of the
cles are usually signs of solar lentigo or flat seborrheic lesion, give rise to the differential diagnosis of lichen
keratosis. Brown circles associated with hair follicles planus-like keratosis, pigmented actinic keratosis and
can also be seen in solar lentigo and flat seborrheic in situ melanoma. Clues that favor pigmented actinic
keratosis. Occasionally, brown circles are present in keratosis (in addition to the non-dermatoscopic clue of
158 An algorithmic method for the diagnosis of pigmented lesions
Solar lentigo
Only brown Seborrheic keratosis
circles (Rarely: Clark nevus, dermatofibroma,
facial melanoma in situ)
Circles
Melanoma in situ
Some circles
Lichen planus-like keratosis
are gray
Pigmented actinic keratosis
palpable roughness) are white circles, scale, and 4-white 5.1.4 Clods
dots in a square (the latter visible only with polarized After reticular lines, the pattern of clods is the second
dermatoscopy). In lichen planus-like keratosis one can most common. Proceeding according to the method,
usually see remnants of a solar lentigo. The differential color is assessed next.
diagnosis of facial lesions with grey circles can be
challenging (8) and is discussed further in chapter 8. One color predominates
In figure 5.18 we show a potpourri of facial lesions When a lesion consists exclusively of clods, the color of
with circles or dots arranged as circles. the clods determines the diagnosis (5.21). When white
A biopsy is often needed to diagnose facial lesions and/or yellow clods predominate, seborrheic keratosis
with gray circles, as dermatoscopy cannot always dif- is the most common diagnosis. Dilated infundibula and
ferentiate between melanoma in situ, lichen planus-like inclusion cysts (“milia”) filled with keratin are clearly
keratosis and pigmented actinic keratosis. Confocal seen as yellow or white clods (5.22). It is important to
laser scanning microscopy may prove useful in this remember that the white clods of seborrheic keratosis
special case. may not be accurately appreciated when using polarized
The pattern of circles is not confined to facial skin. On dermatoscopy, only becoming clearly visible when a
the trunk or the extremities, the pattern of thin brown non-polarized instrument is used (12).
circles has the same differential diagnoses as thin retic-
ular lines, i.e. the pattern of circles may be a variant White and/or yellow clods are also found in cases of
of the reticular pattern. In both cases there is melanin sebaceous gland hyperplasia. These clods are located
hyperpigmentation in the basal keratinocytes. In the centrally and are all of similar size and shape. Radial
reticular pattern, the rete ridges are narrow so that vessels, which do not cross the center of the lesion,
the lines touch each other, thus creating the impression are a strong clue to sebaceous gland hyperplasia. The
of a network pattern. In a pattern of circles, however, central clods are subtle and easily overlooked if one
the rete ridges are broad so that the lines arranged is distracted by the more obvious peripheral vessels.
as circles around the papillae do not touch each other The color of clods in a seborrheic keratosis depends
and on dermatoscopy one sees discrete circles (5.19). on the quantity of the melanin mixed with keratin and
For both brown circles and brown reticular lines, the may range from white or yellow to orange, brown or
differential diagnosis includes a junctional Clark nevus black. While orange clods are a sign of seborrheic
and a solar lentigo. Dermatofibroma is an additional keratosis, one should include basal cell carcinoma in
diagnosis for the pattern of circles (5.20). Rarely, a the differential diagnosis because orange clods may
dermatofibroma may consist of just one pattern, i.e. also be due to ulceration (serum crust). Whereas
thin brown circles. multiple orange clods are seen in seborrheic keratosis
An algorithmic method for the diagnosis of pigmented lesions 159
Purple
Hemangioma/vascular malformation
Yellow/white
Seborrheic keratosis, Sebaceous gland hyperplasia
Clods
1 color Skin colored Congenital nevus (Unna or Miescher nevus)
Seborrheic keratosis
Black
Hemangioma, thrombosed, Hemorrhage
Blue (gray)
Basal cell carcinoma
Figure 5.21: Continuation of the decision tree for one pattern, clods, one color
An algorithmic method for the diagnosis of pigmented lesions 161
(5.23), in basal cell carcinoma one usually finds one or as the foreground and hence constituting structure. In
two orange clods with traces of red due to red blood practice this assumption usually works, but it is incorrect
cells in the serum crust. A further clue is the pattern and is a common cause of errors in dermatoscopy.
of vessels: in basal cell carcinoma serpentine vessels, The correct general principle is that structure is defined
often branched; in seborrheic keratosis, looped or by pigment. For example, the spaces between reticular
coiled vessels (only rarely serpentine). White dots or lines are not clods because the (more heavily pigmented)
clods may be found in both diagnoses. lines represent the structure and the (less pigmented)
Unconscious rules often affect how patterns and mor- spaces are merely the background against which the
phology are perceived. Understanding these unconscious lines are defined. It takes experience and deliberate
rules is important in all of dermatoscopy, but it is partic- training of the eye to (when necessary) override uncon-
ularly relevant to the interpretation of white structures. scious rules and correctly make this distinction between
In familiar settings one has no difficulty in establishing foreground and background.
which features constitute foreground and which con- By definition, a structure is called “white” only when it is
stitute background, but this is not always the case in clearly lighter than the normal perilesional skin. White
unfamiliar settings, for example when one is learning structures (lines, circles, dots or clods) are exceptions
dermatoscopy. The unconscious tendency is to interpret to the general principle that pigment defines structure.
what one perceives to be the most prominent features That is, when white structures are seen one should
162 An algorithmic method for the diagnosis of pigmented lesions
reverse normal practice and interpret the more heavily Large, polygonal skin-colored clods are usually found
pigmented structures as constituting the background in exophytic congenital nevi with a papillomatous sur-
against which the hypopigmented white structures are face, such as Unna nevus or Miescher nevus, and also
defined. occasionally in verrucous seborrheic keratosis (5.25).
For the sake of completeness it should be mentioned In all of these lesions one may also find smaller orange
that, in rare cases, Bowen’s disease (especially when it clods interspersed between the skin-colored clods.
occurs in conjunction with a seborrheic keratosis) may When a pigmented lesion consists exclusively of brown
have only white, yellow and/or orange clods. Invasive clods, various types of melanocytic nevi must be con-
squamous cell carcinomas are usually non-pigmented sidered in the diagnosis (5.26). Large, polygonal light-
but may have white circles or clods as a clue to the brown clods are primarily signs of an Unna nevus or
correct diagnosis (13). The diagnosis of non-pigmented Miescher nevus, especially when the clinical appearance
lesions is discussed in greater detail in chapter 6. is papillomatous. Quite often typical curved vessels
Red or purple clods are characteristic of hemangioma are found in the center of the clods, but the vascular
or vascular malformations (5.24). Thrombosed vessels morphology of these nevi may be highly polymorphous.
are seen as black clods. Hemorrhage may be seen In general, pigmented lesions should be diagnosed
as red clods. The differential diagnosis of blue clods on the basis of their structure and color. A diagnosis
is quite different from that of purple or black clods, so made on the basis of the pattern and color should not
this distinction must be made carefully. be discarded because the corresponding pattern of
An algorithmic method for the diagnosis of pigmented lesions 163
vessels is absent. The pattern of vessels should, at most, center. However, these clues are not sufficiently specific to
be used to confirm the diagnosis. always distinguish Spitz nevi from small congenital nevi.
Small to medium-sized, round and oval brown clods Blue clods are characteristic of pigmented basal cell
are characteristic features of small congenital nevi, of carcinoma (5.28). While melanoma and combined
both the “superficial” and “superficial and deep” types. congenital nevus may also show a pattern of blue clods,
Some pigmented Spitz nevi may also have only brown they nearly always also have clods of other colors, or
clods, or brown clods peripherally may combine with another pattern in addition to blue clods. Therefore,
gray clods or lines centrally (5.27). Central hyperpig- when confronted with only blue clods one should first
mentation is also common in pigmented Spitz nevi, and look for clues to support or refute a diagnosis of basal
peripheral clods are usually smaller than those in the cell carcinoma.
164 An algorithmic method for the diagnosis of pigmented lesions
Seborrheic keratosis
Orange clods predominate
Unna nevus
Other pigment Basal cell carcinoma (rarely)
Clods
>1 color
Congenital nevus
Unna nevus
Melanin Spitz Nevus
Melanoma
Basal cell carcinoma
(Seborrheic keratosis, clonal type)
Figure 5.29: Continuation of the decision tree for one pattern, clods, more than one color
166 An algorithmic method for the diagnosis of pigmented lesions
More than one color clods appear together with brown clods it is likely that
When one finds clods of different colors it is helpful to the pigmentation is due to melanin.
distinguish between the colors of melanin and the colors White, yellow or orange clods signify keratin with
of other pigments (5.29). Clods whose pigmentation (orange) or without (white or yellow) inclusions of
is due to melanin are brown, blue or gray. Although melanin. However, orange clods may also result from
melanin may also appear black on dermatoscopy, a ulceration (serum crust). White or yellow clods are
pattern of black clods is nearly always a sign of the mainly found in seborrheic keratoses and less often
blood pigment hemoglobin. Accumulations of melanin in Unna nevi (5.30). Orange clods in large numbers
appear black in the stratum corneum but usually appear also indicate a seborrheic keratosis or less often an
as dots rather than as clods. When one finds black Unna nevus. In addition, just a few orange clods can
melanin clods, there are almost never enough to form be produced by ulceration in a basal cell carcinoma.
a pattern. To differentiate between melanin and hemo- When any combination of red, purple or black clods
globin one may also consider the other colors present is seen, the differential diagnoses are the same as
in the lesion. If black clods appear together with red for a pattern of clods with only one of these colors;
or purple clods it is almost always hemoglobin. If black hemangioma, vascular malformation or hemorrhage.
An algorithmic method for the diagnosis of pigmented lesions 167
Figure 5.32: Seborrheic keratosis (clonal type) with brown and gray clods
Dots
Clark nevus
Only brown Solar lentigo
dots present Pigmented Bowen’s disease
Pigmented purpuric dermatosis
Figure 5.33: Continuation of the decision tree for one pattern, dots
may find residual features of the original solar lentigo brown dots. Distribution may be random, or (mainly
or seborrheic keratosis e.g. curved lines or the typical on the face) dots may be arranged as circles or as
sharply defined and scalloped border. Lichen planus-like angulated lines around the openings of the infundibula
keratosis occurs usually – but not exclusively – on chronic (15). On the trunk or the extremities, the gray and brown
UV-exposed sites such as the face or the dorsum of the dots of an in situ melanoma may also form angulated
hand, surrounded by other solar lentigines and other lines (polygons). These angulated lines on non-facial
signs of UV-related aging of the skin. Quite often one skin are much larger than structures formed around
may find several lesions simultaneously. infundibular openings (7).
Pigmented actinic keratoses occur predominantly on the Finally it should be mentioned that some inflammatory
face whereas pigmented Bowen’s disease preferentially skin diseases are associated with melanophages in the
occurs on the trunk and the extremities. Pigmented Bow- dermis and so may also have gray dots. In most such
en’s disease is notorious for mimicking other lesions, but cases, one must rely on the clinical signs to reach the
usually the clues of coiled vessels and dots arranged as correct diagnosis.
lines lead to the correct diagnosis (14). Finally, in situ The exclusive presence of brown dots usually indicates
melanomas may also have gray dots, usually mixed with a Clark nevus. Rarely a solar lentigo or pigmented
An algorithmic method for the diagnosis of pigmented lesions 169
Bowen’s disease may also have only brown dots. The then includes heavily pigmented melanocytic lesions
combination of brown and red dots may also occur in like Reed nevus, Clark nevus or melanoma.
inflammatory skin diseases associated with extravasation A blue structureless pattern is quite specific for blue
of red blood cells, such as various forms of pigmented nevi of all types. One should keep in mind the fact that
purpuric dermatosis. As in other inflammatory skin a blue nevus may – in addition to blue structureless
diseases, clinical signs rather than dermatoscopy lead zones – have gray or even brown areas which make
to the correct diagnosis. the nevus appear variegate. Within blue or dark-gray
structureless zones one may also find light-gray areas
5.1.6 Structureless that could be interpreted as structures – usually lines or
When no basic elements are seen, or there are too few clods. In accordance with general principle that struc-
to constitute a pattern, or the visible structures cannot tures are defined by pigment, these light-gray lines or
be reliably assigned to just one of the five basic ele- clods should be ignored, as they have less (not more)
ments, this is termed a structureless pattern. However, pigment than the surrounding area and so should not
“structureless” does not mean “featureless” – whether be considered to be structures.
due to different shades of color within the lesion or In exceptional cases, melanomas and metastases of
a type of “granularity”. One should therefore avoid melanomas may be blue and structureless. However,
the term “homogeneous”. The structureless pattern is even in these exceptional cases one finds additional
the least specific pattern, thus giving rise to a long list clues to the correct diagnosis. These clues include black
of differential diagnoses. In the absence of structure, dots and gray lines, which should not be present in a
color may be the only clue. Even clues are often absent blue nevus. Unlike blue nevus, there will be a history of
because most clues are based on some kind of “struc- progressive growth. Metastases of melanoma can usually
ture”. In summary, lesions that only have a structureless be diagnosed on the basis of their clinical features in
pattern are difficult to diagnose using dermatoscopy combination with the past history of melanoma. Apo-
and therefore often require histopathology. crine hidrocystomas and exogenous pigmentation (for
example tattoos) can be structureless blue. Structureless
One color predominates over all others blue pigmented basal cell carcinomas have also been
The colors black, blue, brown and red are of practical reported, but these are excessively rare. A history of
relevance (5.35). Black and structureless usually indi- progressive growth or the presence of clues to basal
cates the presence of hemoglobin (not melanin) and its cell carcinoma may alert the clinician to this possibility.
degradation products. Hemorrhagic crusts, hemorrhages The brown and structureless pattern indicates solar
in the epidermis in general, and thrombosed vessels all lentigo, flat seborrheic keratosis, pigmented Bowen’s
appear black and structureless. In exceptional cases, disease, or melanocytic nevus, usually of the “superfi-
melanin in the stratum corneum can entirely cover all cial” or “superficial and deep” congenital type. A red
other structures and colors. The differential diagnosis structureless lesion is created by a recent hemorrhage
170 An algorithmic method for the diagnosis of pigmented lesions
Hemorrhage
Black Hemangioma, thrombosed
Reed nevus or Clark nevus (rarely)
Melanoma (rarely)
Blue nevus
Blue Apocrine hidrocystoma
Exogenous pigmentation
Melanoma or melanoma metastasis (rarely)
Structureless
1 color
Solar Lentigo/seborrheic keratosis
Brown Pigmented Bowen’s disease
Congenital nevus
Clark nevus
Red
Hemorrhage
Figure 5.35: Continuation of the decision tree for one pattern, structureless, one color
in the stratum corneum. This will become a black struc- only one large contiguous area is seen, the lesion
tureless lesion as the hemoglobin degrades, before it should be interpreted as structureless and not as a
entirely disappears due to transepidermal elimination. large clod. Skin color and white are not regarded as
pigment. Lesions consisting only of these two “colors”
More than one color are discussed in chapter 6 as non-pigmented lesions.
Sometimes for lesions with more than one color it is When the colors of keratin, namely yellow and orange
difficult to decide whether the pattern is one of clods are predominant, one should first consider keratinizing
or structureless. The difference is that clods are well lesions such as seborrheic keratosis. Structureless lesions
circumscribed, and always occur in numbers. When that are only yellow or orange are rare. Occasionally a
An algorithmic method for the diagnosis of pigmented lesions 171
Stepwise Procedure
1. Lines+
2. Pseudopods+
4. Clods+
5. Dots+
Figure 5.37: Continuation of the algorithm for more than one pattern
basal cell carcinoma may have a large, orange struc- one pattern. To constitute a pattern, multiple repeti-
tureless area, corresponding to an erosion. When the tions of a given basic element must be found, in an
colors of hemoglobin, namely red and purple predom- area occupying a significant part of a lesion. Two or
inate, the only diagnoses to consider are hemorrhage, more such areas must be found before a lesion can
or hemorrhage in a pre-existing lesion such as a nevus. be classified as having more than one pattern. A few
Structureless lesions whose pigmentation is primarily isolated lines, dots, clods, circles or pseudopods in
due to melanin may have black, brown, gray or blue a pattern of another basic element does not mean
areas (5.36). Black zones in a structureless lesion can there is more than one pattern, and such isolated
also result from thrombosis. As a rule of thumb, black basic elements should be ignored at this stage. When
should be attributed to blood when it appears together appropriate, they can be taken into account when
with red or purple and attributed to melanin when it one is weighing clues.
appears together with brown, blue or gray. The exact number of patterns is unimportant; no more
When the colors of melanin are symmetrically distributed diagnostic accuracy is achieved by counting patterns
in a structureless lesion, this is most likely a nevus but it than by simply distinguishing between one and more
could be practically any type of nevus. A specific classi- than one pattern. Requiring only this simple judgment
fication is usually not possible by dermatoscopy. When improves agreement between observers.
the colors of melanin are distributed asymmetrically in As lesions become more complex, it becomes increas-
a structureless lesion, one should consider melanoma, ingly likely that more than one interpretation could
a metastasis of a melanoma, and seborrheic keratosis. reasonably be considered by the investigator. In part
The distinction is made on the basis of specific clues. this reflects the skill of the investigator, but it also
In nodular structureless lesions that are blue and black reflects (often poorly understood) variations in basic
a melanoma should be ruled out (16). The color black, perception between observers. Two features of the
which usually indicates melanin in the stratum corneum, algorithm reduce errors in diagnosis due to differing
is not expected in blue nevi. Exceptionally, a pigmented interpretations. Firstly, as will be detailed below,
basal cell carcinoma or a dermatofibroma may show descriptions are generated in a defined stepwise
a structureless pattern with blue, brown, or grey areas. fashion. Secondly, the algorithm is constructed in
such a way that various interpretations, as long as
they are plausible, lead to the same diagnosis. The
5.2 More than one pattern result is that the algorithmic method generally leads to
Although it is true that the majority of pigmented the same conclusion regardless of which algorithmic
lesions have more than one pattern, beginners tend pathway is followed. Of course, this redundancy in
to classify far too many lesions as having more than the algorithm is not infallible, so when the investiga-
172 An algorithmic method for the diagnosis of pigmented lesions
Stepwise Procedure
1. Reticular or branched
2. Angulated
Lines+ 3. Parallel
4. Radial
5. Curved
Figure 5.38: Continuation of the algorithm for more than one pattern, lines
tor is uncertain which pathway to follow, it is good and symmetry is judged purely on arrangement of
practice to follow all the plausible pathways and patterns. The more patterns there are, the less is the
consider all the differential diagnoses that the different likelihood of symmetry.
pathways offer. Structural symmetry has been defined in chapter 3.
In contrast to lesions with only one pattern, for which In theory there are an infinite number of ways that
all patterns are regarded as being equally important, two patterns can combine symmetrically. In practice,
the algorithm for pigmented lesions with more than there are only three symmetrical arrangements of two
one pattern is constructed in a hierarchical manner. patterns in pigmented skin lesions: a) One pattern is
If there is more than one pattern, one looks for the in the center and the other at the periphery, b) the
individual patterns in a stepwise manner, following a opposite is the case, and c) the basic elements of
sequence established on the basis of pattern specificity one pattern (e.g. dots) are regularly spread over a
(5.37). The sequence starts with the pattern of lines, second pattern (e.g. reticular lines). All other combi-
the most specific pattern in dermatoscopy, and ends nations of two patterns are, by definition, considered
with the structureless pattern, which is the least specific. asymmetrical.
Thus the description begins not with the most prominent When assessing symmetry we should keep in mind that
pattern present, but with the most specific. we are dealing with biological structures. Assessment
When a lesion consists of more than one pattern of symmetry is therefore a matter of judgment as to
the investigator first determines whether a pattern the type and degree of variation that is expected in
of lines is present or not. When a pattern of lines nature, rather than a strict application of the propo-
is present, one follows the algorithm for patterns of sitions of geometry. At times there will be uncertainty
lines. When no pattern of lines is present, one looks as to whether a lesion should be judged symmetrical
next for pseudopods, then for circles, then clods, and or asymmetrical.
finally for dots. When none of these patterns of basic This is an important role of experience in derma-
elements are seen, the lesion logically must consist toscopy; experts can confidently call more lesions
of only one pattern, namely structureless, and the symmetrical than beginners, reducing the need for
analysis is performed in accordance with the known exhaustive assessment to exclude malignancy. In
rules for this pattern. cases of uncertainty, it is prudent to consider all the
The most important decision in the analysis of pig- differential diagnoses at the ends of both applicable
mented lesions with more than one pattern is the pres- branches of the algorithm. Considering all applica-
ence or absence of structural symmetry. Symmetry in ble branches of the algorithm is appropriate in any
lesions with one pattern is judged on the distribution situation when one reaches a decision point and is
of colors within the lesion. In lesions with more than uncertain of the correct pathway. This ensures that no
one pattern, the distribution of color is not assessed, potential diagnosis is discarded prematurely.
An algorithmic method for the diagnosis of pigmented lesions 173
Symmetric combinations:
Reticular and clods
Figure 5.39: Symmetrical combinations of the two patterns – reticular and clods
Symmetric combinations:
Reticular and structureless
Figure 5.42: Symmetrical combinations of reticular at the periphery and structureless skin-colored in the center.
Two congenital nevi with reticular lines at the periphery and a structureless skin colored zone in the center.
tureless zone centrally (5.41). The central structure- (5.43). One should not be misled by the histopatho-
less zone may be skin-colored or light brown. If the logical finding of a “dysplastic junctional nevus”
central structureless zone is skin-colored and raised because this is just a different name for Clark nevus.
or papillomatous, the diagnosis is most commonly a When the center is structureless blue, this is usually
“superficial and deep” congenital nevus (5.42). If the a combined congenital nevus (5.44). If the center is
central structureless zone is light brown and flat it could white, i.e. lighter than the surrounding skin, and the
be a superficial or superficial and deep congenital adjacent reticular lines are light-brown and thin, the
nevus or a Clark nevus (5.43). An accessory nipple most likely diagnosis is dermatofibroma (5.45).
may also have this pattern. A symmetrical combination of pseudopods and/or
If the central structureless area is black, the diagnosis radial lines with the reticular pattern (i.e. the pseu-
is nearly always Clark nevus, or rarely a Reed nevus dopods/radial lines are seen occupying the entire
176 An algorithmic method for the diagnosis of pigmented lesions
circumference) indicates a Reed nevus. A symmetrical needs careful assessment and should specifically be
combination of peripheral radial lines with a reticular assessed for clues to melanoma (5.46). As a gen-
pattern in the center is also rarely seen with a Clark eral rule when assessing a lesion for clues, a clue
nevus. In practice, when any of these patterns are is only considered to be present when it is clearly
seen it is difficult to reliably exclude melanoma, so present. Imagination and fantasy have no place in
such lesions (in adults at least) should be submitted the search for clues. The diagnosis is nevus (Clark
for histopathology. nevus, combined congenital nevus, “superficial” or
Symmetrical combinations of three patterns are seen, “superficial and deep” congenital nevus) only when
though less often than symmetrical combinations of there are no clues to melanoma (5.47). The diagnosis
two patterns. One example would be structureless is melanoma when (by these standards) at least one
in the center and a combination of reticular lines clue to melanoma is present (5.48).
and dots or clods at the periphery. These three-fold
combinations are usually found in “superficial” and Angulated lines
“superficial and deep” congenital nevi. The main differential diagnosis of lesions with angu-
Any lesion with an asymmetrical combination of lated lines is flat melanoma on chronic sun-damaged
patterns that includes reticular or branched lines skin (including facial and non-facial skin). Many
178 An algorithmic method for the diagnosis of pigmented lesions
Clue to melanoma
Melanoma
Asymmetric
Clark nevus
No clue to melanoma
Congenital nevus, "superficial" or "superficial and deep"
Combined congenital nevus
Figure 5.46: Continuation of the algorithm for more than one pattern, reticular, with asymmetrical combination of patterns
Figure 5.47: More than one pattern, reticular, asymmetrical, without clue to melanoma.
Top: More than one pattern, reticular and clods, combined asymmetrically, more than one color (light-brown and dark-brown), but no clue
to melanoma – “superficial and deep” congenital nevi. Bottom left: More than one pattern, reticular and structureless, combined asymmet-
rically, more than one color (light-brown and dark-brown), but no clue to melanoma (the eccentric structureless area is skin-colored and
therefore not a clue) – a Clark nevus. Bottom right: More than one pattern, reticular and clods, combined asymmetrically, one color (brown)
and no clue to melanoma – a “superficial and deep” congenital nevus.
An algorithmic method for the diagnosis of pigmented lesions 179
Figure 5.48: More than one pattern, reticular, asymmetrical, with clues to melanoma.
Top left: More than one pattern, reticular and structureless, combined asymmetrically, more than one color, white eccentric structureless zone
as a clue to melanoma – a melanoma. Top right: More than one pattern, reticular and structureless, combined asymmetrically, more than
one color, white eccentric structureless zone and gray structures and black dots as clues to melanoma – a melanoma. Bottom left: More than
one pattern, reticular and structureless, combined asymmetrically, more than one color, and an eccentric structureless zone with multiple
colors as a clue to melanoma – a melanoma. Bottom right: More than one pattern, reticular, structureless and pseudopods, combined asym-
metrically, more than one color, with pseudopods occupying only some segments of the periphery as a clue to melanoma. Histopathology
shows this is actually a Reed nevus and not a melanoma; nevertheless, melanoma was still the best diagnosis on the dermatoscopy.
Figure 5.49: A flat melanoma on chronic sun damaged skin with angulated lines
Figure 5.50: Pigmented actinic keratosis on facial skin with angulated lines and white circles
the ridges is in the colors of melanin, the diagnosis of seen at other sites (5.52 right). As a general rule,
melanoma must be considered, even in the absence of the thicker a melanoma at an acral site, the more it
other clues to melanoma. Hemorrhage or exogenous resembles melanoma at other locations.
pigmentation are the likely diagnoses when colors It may be difficult to distinguish between the structure-
other than those of melanin are seen. less pattern and the parallel ridge pattern, when the
When the pattern is the furrow- or crossing-pattern, lines in the ridge pattern are wide enough to almost
a distinction is made between symmetrical and asym- occupy the furrows. When the distinction cannot be
metrical combinations of patterns. Symmetrical com- made with certainty, both possibilities should be
binations are found in classical acral nevi and all followed in the algorithm.
other nevi, such as Reed nevi or “superficial” and
“superficial and deep” congenital nevi (5.52 left). Radial lines
Not all melanomas on acral skin have a parallel ridge Radial lines always occur in combination with other
pattern. In asymmetrical combinations involving the patterns. When the radial lines occupy the entire cir-
pattern of furrows or the crossing pattern, the clues cumference of the lesion the combination of patterns is
to melanoma are the same at acral locations as those symmetrical. When assessing symmetry, pseudopods
An algorithmic method for the diagnosis of pigmented lesions 181
>1 Color
Melanoma
Parallel +
symmetric
Acral nevus
2. Furrows or crossing
furrows and ridges No clue to
Acral nevus
melanoma
asymmetric
Clue to melanoma
Melanoma
Figure 5.51: Continuation of the algorithm for more than one pattern, lines, parallel
Figure 5.53: More than one pattern, radial lines at the periphery.
Top: Radial lines at the periphery, distributed over the entire circumference, are typical of Reed nevi. Bottom: The peripheral radial lines
are not regularly distributed over the entire circumference, but are present only in some segments. This is an asymmetrical combination of
patterns. This pattern is seen in basal cell carcinoma and melanoma. The absence of reticular lines, a few blue clods (left) and serpentine
vessels (right) are more indicative of basal cell carcinoma than melanoma. Histopathology confirmed basal cell carcinoma in both cases.
and radial lines are considered equivalent. In practice, between these diagnoses is made on the basis of clues.
radial lines peripherally are found in combination with Two arrangements of radial lines are strong clues to
only two different patterns in the center; clods and basal cell carcinoma. Peripheral radial lines in basal
structureless. When the center is structureless and white, cell carcinoma usually have a common base, which
the lesion is usually a dermatofibroma. When the center is not usually the case in melanoma. Also, in basal
is structureless and brown, black or gray, the lesion cell carcinoma, radial lines are not only seen at the
is usually a Reed nevus (5.53, top row). In the latter periphery, as in melanoma, but also within the lesion.
case one may find brown or gray clods instead of the In this case the radial lines do not just converge at the
structureless center. center but do so at a dot or a clod. These structures are
Asymmetry is necessarily created when peripheral usually multiple, and are possibly the most specific clue
radial lines do not occupy the entire circumference of in dermatoscopy. Another feature often seen in basal
a lesion but are present only in some segments. The cell carcinoma but only rarely in melanoma is radial
primary differential diagnosis is then melanoma versus lines extending from a hypopigmented structureless
basal cell carcinoma (5.53 bottom row). The distinction area (5.53 bottom right). Usually in melanoma radial
An algorithmic method for the diagnosis of pigmented lesions 183
Circles+
Melanoma
Partly gray
Lichen planus-like keratosis
Pigmented actinic keratosis
Figure 5.57: Continuation of the algorithm: more than one pattern, circles
Symmetric combinations:
Clods and structureless
Structureless skin colored central, Structureless black or dark Structureless blue central,
clods peripheral brown central, clods peripheral clods peripheral
Symmetric
Yellow or
Seborrheic keratosis
white
Clods+
Orange Basal cell carcinoma
Seborrheic keratosis
Other pigment
Hemangioma
Red or purple
Melanoma,
Asymmetric primary or metastatic
Figure 5.61: Continuation of the algorithm for more than one pattern, clods, asymmetrical
nevus. In cases of a blue structureless center, the first are sticky, fibers of clothing may become adherent
entity to be considered is a combined congenital to them and serve as an indirect sign of ulceration.
nevus (5.60 right). Red or purple clods usually indicate a hemangioma
Asymmetrical combinations of patterns containing or a vascular malformation, but may also signify a
clods but no lines, pseudopods or circles are ana- melanoma or metastasis of melanoma when occurring
lyzed differently depending on whether the colors are in combination with another pattern. In particular,
predominantly those of melanin, or those of another hemangioma should not be diagnosed when vessels
pigment (5.61). as lines or dots are seen within the red clods.
Lesions pigmented by melanin are predominantly When melanin is the predominant pigment, the color
black, brown, gray or blue. Black pigmentation can of the whole lesion is assessed, and not just the clods.
be caused by coagulated blood as well as melanin, When the lesion is one color, brown, the diagnostician
so the interpretation of the color black depends on should consider a “superficial and deep” congenital
what other colors are present. When the colors brown, nevus or a Spitz nevus. When other colors are also
blue or gray are predominant in the remainder of the present the diagnostician should consider a basal
lesion, the color black is best interpreted as melanin cell carcinoma or a melanoma in addition to a sebor-
in the stratum corneum. However, when red or purple rheic keratosis and its variants (5.62). The distinction
is predominant, the color black is best interpreted as between these three differential diagnoses is made,
coagulated blood. as mentioned earlier, on the basis of additional clues.
Lesions with no pigment, or with pigments other than
melanin predominant, are assessed based on the color 5.2.5 Dots
of the clods, and not on the color of the rest of the When all other patterns have been excluded, only
lesion. A predominance of white or yellow clods is dots remain (5.63). In the algorithm we are currently
indicative of a seborrheic keratosis. Orange clods in in the category of “more than one pattern”. Thus, all
large numbers also indicate a seborrheic keratosis but lesions that now follow consist of dots and a struc-
when only a very few are present, basal cell carcinoma tureless zone. As structureless is the least specific
must also be considered. The orange clods of basal pattern, the diagnostic process is mainly based on
cell carcinoma are simply serum crusts arising from the color of dots, meaning that the algorithm for
erosion or ulceration, so red inclusions (representing “dots and structureless” differs only slightly from the
blood) within the clods are common. As serum crusts algorithm for “dots”.
188 An algorithmic method for the diagnosis of pigmented lesions
Stepwise procedure
Lichen planus-like keratosis
Pigmented actinic keratosis
1. Gray dots
Pigmented Bowen‘s disease
Melanoma, regressive
Basal cell carcinoma
2. Blue dots
Basal cell carcinoma
Dots+
Clark nevus
4. Brown dots Congenital nevus
Solar lentigo
Pigmented Bowen‘s disease
Figure 5.63: Continuation of the algorithm for more than one pattern, dots
Gray dots may signify a melanoma (5.64 right), a for basal cell carcinoma (5.64 left). Usually there also
lichen planus-like keratosis, a pigmented superficial will be additional clues to support this diagnosis. Black
squamous cell carcinoma (actinic keratosis or Bow- dots are uncommon but should cause the investigator
en’s disease) or a basal cell carcinoma. Of these to think of melanoma, and prompt a search for other
differential diagnoses, the basal cell carcinoma can clues to this diagnosis. Brown dots are found in solar
be differentiated most easily from the others on the lentigo and pigmented Bowen’s disease. Very rarely
basis of additional clues. Blue dots are quite specific a Clark nevus may have just brown dots and a struc-
190 An algorithmic method for the diagnosis of pigmented lesions
tureless area. The distinction between solar lentigo 5.4 Chaos and Clues
and pigmented Bowen’s disease can be made quite No diagnostic system will detect every pigmented
easily when the pattern is brown dots plus structureless. skin malignancy. Melanomas, pigmented basal cell
In pigmented Bowen’s disease, the dots are usually carcinomas, and pigmented squamous cell carcino-
arranged as lines. These lines are often radial, and mas including pigmented Bowen’s disease, must all
may include coiled vessels. Coiled vessels are also start as minute lesions at which time dermatoscopic
common in the structureless zone. features of malignancy may not be recognizable.
Unlike benign lesions, however, malignant lesions
will grow continuously, and with increasing size clues
5.3 Applying pattern analysis to clinical practice to malignancy can be expected to become visible to
The basics of pattern analysis are easy to learn, the dermatoscopist.
but its application is sometimes complex and needs Several diagnostic methods have been developed
experience. Gaining experience requires time and the for pigmented skin lesions based on dermatoscopic
opportunity to work regularly with dermatoscopy; i.e. analysis. Classical pattern analysis was the original
regular use in one’s medical practice. The spectrum method published by Pehamberger, Steiner and Wolff
of pigmented skin lesions is not very large yet to per- in 1987 (19) and it is still widely used by experienced
sonally examine the full gamut of pigmented lesions dermatoscopists. The method we present in this book is
of the skin, including rare diagnoses and unusual nothing but pattern analysis presented using an objec-
appearances of common diagnoses, takes some time tive, geometric language and with a clear, stepwise
even at specialized centers, and proportionately longer path to generate descriptions and reach a diagnosis.
in small practices. Fortunately, seeing photographs of Not every clinician has the time or the desire to
unusual or rare lesions in dermatoscopy atlases or on become an expert in dermatoscopy. Many simply
various internet websites can speed up this process. wish to have uncomplicated and easily assimilated
Familiarity with the spectrum of common diagnoses is guidelines for daily use. The ABCD rule was designed
at least as important as knowing about rare diagnoses. specifically for melanocytic lesions for the detection
It is absolutely essential to be aware of the morpho- of melanomas (20). The 7-point checklist (21), Men-
logical spectrum of the Clark nevus and the seborrheic zies’ method (22), and the CASH algorithm (23)
keratosis. This knowledge is best acquired by first-hand involve a 2-step process where the first step attempts
experience gained in the course of regular – ideally to determine whether a lesion is melanocytic before
everyday – application of dermatoscopy to one’s own an algorithm is applied to determine whether it should
patients. The use of the algorithmic method will become be biopsied to exclude melanoma. Finally, the 3-point
quite natural over time. checklist was developed in 2000 to detect pigmented
The algorithm need not be learned by heart, but malignancy (24).
it must be explored. After all, one of the roles of These algorithms may be easy to learn but, with the
experience is to discover one’s own pathway while exception of Menzies’ method, they are not easy
traversing the road to expertise. With time, experts to apply. No one actually calculates scores for all
become so familiar with the pathway that they do lesions assessed, because it takes too long to fit into
not require a map, appearing to arrive at their goal normal clinical routine.
blind – i.e. without an algorithm. In actual fact they In structure, Menzies’ method can be seen to be a
make decisions so rapidly that it can appear that simplified version of the algorithms of pattern analysis
they follow no method other than their own intuition. as it assesses in sequence pattern, color and clues.
This illusion is so strong that some experts do actually Unlike pattern analysis, Menzies’ method is limited
believe in their own intuition. The disadvantage of to melanocytic lesions.
intuitive diagnosis is that the method cannot be taught. Fortunately, simple and easily learned rules of thumb
A method that cannot be taught is barely a method at based on pattern analysis can be formulated, that fulfill
all. Beginners should beware of intuitive diagnoses: the demand for a rapid and uncomplicated algorithm,
without experience they are frequently incorrect. but without this restriction to melanocytic lesions. We
now present one such method.
“Chaos and Clues” is designed to be applied to any
pigmented skin lesion (25) to detect any type of malig-
nancy and to achieve this rapidly in the setting of a
busy practice (26). In essence lesions are examined
An algorithmic method for the diagnosis of pigmented lesions 191
* Exceptions to no intervention
1. Changing lesions on adults At least one of:
2. Nodular or small lesions with any clue 1. Gray or blue structures
3. Head/Neck: Pigmented circles or dermatoscopic gray 2. Eccentric structureless area
4. Acral: Parallel ridge pattern
3. Thick lines reticular or branched
Biopsy (unless unequivocal
Clue present 4. Black dots or clods, peripheral
diagnosis of seborrheic
5. Lines radial or pseudopods, segmental
keratosis by pattern analysis)
6. White lines
7. Lines parallel, ridges (acral) or chaotic (nails)
Chaos present
8. Polymorphous vessels
9. Angulated lines (polygons)
Chaos absent
No intervention*
clinically and dermatoscopically for chaos (defined as study was based on 463 consecutive pigmented skin
asymmetry of pattern or color) and only when this is lesions from a primary care skin cancer practice (25).
discovered does the clinician pause to search for one This included 29 melanomas (20 in situ), 72 pigmented
of nine clues to malignancy. If there are both chaos basal cell carcinomas and 37 pigmented squamous
and at least one clue to malignancy then (excision) cell carcinomas (including pigmented Bowen’s dis-
biopsy is indicated (5.65). Lesion descriptions in the ease and pigmented actinic keratosis). Diagnostic
chaos and clues algorithm are formulated using the sensitivity was 90.6 % and specificity was 62.7 % for
same method and language as for pattern analysis. the diagnosis of malignancy and significantly better
“Chaos and Clues” is designed to detect malignancy than with the unaided eye. The specificity increased
rather than to make a specific diagnosis. In other to 77 % when solar lentigines/seborrheic keratoses
words, it guides the clinician in the decision whether were diagnosed by pattern analysis.
or not to submit a lesion for histopathology. We do
not believe that attempting to determine melanocytic Chaos
status is a useful part of this process (2). Chaos is defined as asymmetry of pattern and/or color
Sometimes this may be obvious but we believe this is within a lesion, the shape of a lesion is not relevant
rightly the domain of the pathologist, who is actually (5.66). By definition a lesion with one pattern and
able to see melanocytes. The exact diagnosis is also one color, regardless of its shape, is symmetrical and
left to the pathologist. Of course, with increasing therefore does not exhibit chaos. If any line drawn
experience and expertise, the clinician may attempt through the center of a lesion has different colors
to reach a specific diagnosis by applying pattern or patterns on opposite sides it is asymmetrical and
analysis. exhibits chaos. Any color other than skin color at the
Unlike previously proposed “simplified” algorithms, edge of a lesion (such as white) should be regarded
“Chaos and Clues” has been evaluated in the normal as part of the lesion.
clinical situation which requires the detection of all Lesions without chaos, subject to four exceptions, are
pigmented malignancies and not just melanoma. This not analyzed any further.
192 An algorithmic method for the diagnosis of pigmented lesions
While natural laws such as gravity, surface tension, should be assessed further. These include changing
electromagnetic forces and biological feedback mech- lesions on adults, nodular or small lesions with any
anisms favor symmetry, malignant tissue tends to not clue, dermatoscopic pigmented circles or gray struc-
be restrained by feedback mechanisms and this is tures on the head or neck and a parallel ridge pattern
a plausible explanation for dermatopathologic and on palms or soles. The beginning dermatoscopist can
therefore dermatoscopic chaos. reasonably be expected to assess more lesions as
While natural laws do favor symmetry it is very rare asymmetrical than experts. By refining this judgement
to find perfect symmetry in nature, and so judge- with accumulated experience, one reduces the number
ment is required in deciding whether deviations from of lesions requiring full assessment.
geometrically perfect symmetry fall within normal
biological variation. It can be useful when assessing Clues to Malignancy
equivocal chaos to consider whether what is observed In pattern analysis, a clue is a feature which favors
is consistent with the chaotic behavior of malignant one diagnosis over another, when analysis of pat-
tissue. If a decision cannot be made the lesion should terns and colors has not led to a specific diagnosis.
be assessed as exhibiting chaos and fully assessed. In the chaos and clues algorithm, a clue is simply a
There are four exceptions where a lesion without chaos feature which, when present, indicates that a lesion
An algorithmic method for the diagnosis of pigmented lesions 193
requires a biopsy to exclude malignancy. One clue rotated. Both types of light are suitable for dermatos-
is sufficient. Both chaos and clue can be produced copy, but the dermatoscopist should be aware of the
by the same feature. different information each can give. Unless otherwise
Because some clues depend on the colors gray, blue specified, dermatoscopic images in this chapter were
and white it is important to recognize that the type of taken with non-polarized dermatoscopes.
dermatoscope used can influence the way these colors
are observed. As a general rule, gray and blue struc- The Nine Clues
tures in all skin lesions and the white dots and clods 1. Gray or blue structures (dots, clods, circles, or
in seborrheic keratoses are seen more vividly with lines, 5.67):
non-polarized light. Certain white structures are only Gray dots may be seen in pigmented basal cell
seen when using polarized light. Polarizing-specific carcinoma, in pigmented Bowen’s disease or pig-
white lines — bright white lines at right angles to each mented actinic keratosis, as well as in melanoma.
other (but not crossing each other) — can be seen in Gray circles occur in facial in situ melanomas. Dense
certain lesions, most notably melanomas, Spitz nevi, deposition of melanin in the dermis causes blue
basal cell carcinomas and dermatofibromas. These clods in pigmented basal cell carcinomas and inva-
structures vary in intensity as the dermatoscope is sive melanomas. Gray lines occur in melanomas.
194 An algorithmic method for the diagnosis of pigmented lesions
4. Black dots or clods, peripheral (5.70): are often formed by dots in linear arrangement.
Black dots and clods are generally produced by Radial lines (and pseudopods) in melanomas are
melanin in keratinocytes or pigmented melanocytes expected to be connected to either a pattern of
close to, or at the level of, the stratum corneum. reticular lines or to a pigmented structureless area
Central black dots frequently occur in Clark nevi, whereas in basal cell carcinoma they frequently
but when they are peripheral and not located on extend from a hypopigmented area.
reticular lines, they are a clue to malignancy. 6. White lines (5.72, 5.73): To be considered “white”
5. Lines radial or pseudopods, segmental (5.71): and therefore a clue to malignancy, lines must be
Peripheral pseudopods or radial lines are a feature clearly whiter than normal perilesional skin. This
of Reed nevus when they occupy the entire periph- clue is not restricted to lines in a reticular pattern,
ery (“circumferential”), but when only seen in part any pattern of white lines seen with either polar-
of the periphery (“segmental”) they are a clue to ized or non-polarized dermatoscopy constitutes
malignancy. When radial lines converge to a central a clue. As polarizing-specific white lines may
dot or clod they are highly specific for pigmented occasionally be the only clue to malignancy in
basal cell carcinoma. Radial lines which converge melanomas, we believe that examination with
also occur in pigmented SCC in-situ; these lines polarized dermatoscopy should be routine.
An algorithmic method for the diagnosis of pigmented lesions 197
9. Angulated lines (Polygons) (5.76) around follicular openings and therefore border a
Angulated lines or polygons were first described smaller zone than angulated lines on non-facial skin.
by Keir in flat melanomas on non-facial skin with While angulated lines can be seen in some benign
chronic sun-damage (7). Angulated lines on non-fa- lesions and particularly in facial pigmented actinic
cial skin form complete or incomplete polygonal keratosis, we have found it to be a valuable clue
shapes which are larger than the holes caused to melanoma. The sensitivity and specificity have
by individual follicles and larger by far than the not yet been formally assessed but author CR has
holes bounded by reticular lines. These lines meet found it to be present in 20 % of consecutively
but do not cross. excised melanomas (unpublished data). The clue
Angulated lines may also appear in flat facial mel- of angulated lines (polygons) is usually, but not
anomas. Angulated lines of facial skin are situated always, associated with dermatoscopic grey color.
200 An algorithmic method for the diagnosis of pigmented lesions
Figure 5.76: Angulated lines (polygons) as a clue to flat melanomas on chronic sun-damaged skin.
Polygons bounded by straight, angulated lines in two flat melanomas on non-facial chronic sun damaged skin. In both cases there are
geometric shapes, some complete and some incomplete, formed by lines meeting at angles and larger than the holes caused by individual
follicles and larger by far than the holes bounded by reticular lines.
References
1 Marghoob AA, Braun R. Proposal for a revised 2-step 13 Rosendahl C, Cameron A, Tschandl P, Bulinska A,
algorithm for the classification of lesions of the skin using Zalaudek I, Kittler H. Prediction without Pigment: a
dermoscopy. Archives of dermatology 2010; 146: 426–8. decision algorithm for non-pigmented skin malignancy.
2 Tschandl P, Rosendahl C, Kittler H. Accuracy of the first Dermatology practical & conceptual 2014; 4: 59–66.
step of the dermatoscopic 2-step algorithm for pigmented 14 Cameron A, Rosendahl C, Tschandl P, Riedl E, Kittler H.
skin lesions. Dermatology practical & conceptual 2012; Dermatoscopy of pigmented Bowen’s disease. Journal
2: 203a08. of the American Academy of Dermatology 2010; 62:
3 Zaballos P, Puig S, Llambrich A, Malvehy J. Dermoscopy 597–604.
of dermatofibromas: a prospective morphological study 15 Stolz W, Schiffner R, Burgdorf WH. Dermatoscopy for
of 412 cases. Archives of dermatology 2008; 144: facial pigmented skin lesions. Clin Dermatol 2002; 20:
75–83. 276–8.
4 Akay BN, Kittler H, Sanli H, Harmankaya K, Anadolu 16 Argenziano G, Longo C, Cameron A, Cavicchini S,
R. Dermatoscopic findings of cutaneous mastocytosis. Gourhant JY, Lallas A et al. Blue-black rule: a simple
Dermatology 2009; 218: 226–30. dermoscopic clue to recognize pigmented nodular mel-
5 Argenziano G, Catricala C, Ardigo M, Buccini P, De anoma. The British journal of dermatology 2011; 165:
Simone P, Eibenschutz L et al. Dermoscopy of patients with 1251–5.
multiple nevi: Improved management recommendations 17 Blum A, Hofmann-Wellenhof R, Marghoob AA, Argenzia-
using a comparative diagnostic approach. Archives of no G, Cabo H, Carrera C et al. Recurrent melanocytic nevi
dermatology 2011; 147: 46–9. and melanomas in dermoscopy: results of a multicenter
6 Jaimes N, Marghoob AA, Rabinovitz H, Braun RP, Cam- study of the International Dermoscopy Society. JAMA
eron A, Rosendahl C et al. Clinical and dermoscopic dermatology 2014; 150: 138–45.
characteristics of melanomas on nonfacial chronically 18 Tschandl P. Recurrent nevi: report of three cases with
sun-damaged skin. Journal of the American Academy dermatoscopic-dermatopathologic correlation. Derma-
of Dermatology 2015; 72: 1027–35. tology practical & conceptual 2013; 3: 29–32.
7 Keir J. Dermatoscopic features of cutaneous non-facial 19 Pehamberger H, Steiner A, Wolff K. In vivo epilumines-
non-acral lentiginous growth pattern melanomas. Der- cence microscopy of pigmented skin lesions. I. Pattern
matology practical & conceptual 2014; 4: 77–82. analysis of pigmented skin lesions. Journal of the American
8 Tschandl P, Rosendahl C, Kittler H. Dermatoscopy of Academy of Dermatology 1987; 17: 571–83.
flat pigmented facial lesions. Journal of the European 20 Nachbar F, Stolz W, Merkle T, Cognetta AB, Vogt T,
Academy of Dermatology and Venereology: JEADV Landthaler M et al. The ABCD rule of dermatoscopy.
2015; 29: 120–7. High prospective value in the diagnosis of doubtful mela-
9 Akay BN, Kocyigit P, Heper AO, Erdem C. Dermatoscopy nocytic skin lesions. Journal of the American Academy
of flat pigmented facial lesions: diagnostic challenge of Dermatology 1994; 30: 551–9.
between pigmented actinic keratosis and lentigo maligna. 21 Argenziano G, Fabbrocini G, Carli P, De Giorgi V,
The British journal of dermatology 2010; 163: 1212–7. Sammarco E, Delfino M. Epiluminescence microscopy
10 Saida T, Miyazaki A, Oguchi S, Ishihara Y, Yamazaki Y, for the diagnosis of doubtful melanocytic skin lesions.
Murase S et al. Significance of dermoscopic patterns in Comparison of the ABCD rule of dermatoscopy and a
detecting malignant melanoma on acral volar skin: results new 7-point checklist based on pattern analysis. Archives
of a multicenter study in Japan. Archives of dermatology of dermatology 1998; 134: 1563–70.
2004; 140: 1233–8. 22 Menzies SW, Ingvar C, McCarthy WH. A sensitivity and
11 Sendagorta E, Feito M, Ramirez P, Gonzalez-Beato M, specificity analysis of the surface microscopy features
Saida T, Pizarro A. Dermoscopic findings and histolog- of invasive melanoma. Melanoma research 1996; 6:
ical correlation of the acral volar pigmented maculae in 55–62.
Laugier-Hunziker syndrome. The Journal of dermatology 23 Henning JS, Dusza SW, Wang SQ, Marghoob AA, Rab-
2010; 37: 980–4. inovitz HS, Polsky D et al. The CASH (color, architecture,
12 Benvenuto-Andrade C, Dusza SW, Agero AL, Scope A, symmetry, and homogeneity) algorithm for dermoscopy.
Rajadhyaksha M, Halpern AC et al. Differences between Journal of the American Academy of Dermatology 2007;
polarized light dermoscopy and immersion contact der- 56: 45–52.
moscopy for the evaluation of skin lesions. Archives of
dermatology 2007; 143: 329–38.
202 An algorithmic method for the diagnosis of pigmented lesions
As originally described, pattern analysis is a method 6.1 Clues used in the diagnosis of
for assessing pigmented lesions. This restriction to pig- non-pigmented (amelanotic) lesions
mented lesions is not an accident: patterns formed by
melanin pigment are more prominent and more specific In the absence of melanin pigment, other clues must
diagnostically than other features seen at dermatoscopy. be used in the diagnosis of non-pigmented lesions. In
The absence of melanin structures in non-pigmented general, these clues may also be seen in pigmented
lesions restricts the range of features available for lesions, but being less specific, they are of less signif-
diagnosis. Furthermore, because structures pigmented icance when diagnosis can proceed on the basis of
by melanin are the building blocks of current diagnostic structures pigmented by melanin.
algorithms, alternative methods are required to diagnose The pattern of non-pigmented lesions is usually struc-
non-pigmented lesions. tureless. However, non-pigmented lesions may show
For the purposes of this chapter, we have defined as patterns of white, yellow, orange, pink, skin colored,
“pigmented” any lesion containing any area at all or red clods; and white lines, circles or dots. Lesions
pigmented by melanin (i.e. black, brown, gray, or with orange or red clods, which have been addressed
blue). While a “non-pigmented” lesion lacks these in detail in the section on pigmented lesions, may be
colors, white, yellow, orange, pink, or red may be seen, analyzed as either pigmented or non-pigmented.
either singly or in combination. As these lesions only
lack melanin pigment, “amelanotic” is a more accurate Ulceration
term than “non-pigmented”. The latter term, although Ulceration is not a strong clue to a specific diagnosis,
less precise, remains more popular. In this chapter we but in the absence of a clear history of trauma it is a
use both terms interchangeably. good clue to malignancy. As an over-riding principle,
The assessment of amelanotic lesions is challenging in the absence of a clear and convincing history of
and a specific diagnosis is not always possible even trauma, any solitary ulcerated non-pigmented lesion
when dermatoscopy is added to clinical examination. should be submitted for histopathology. Ulceration is
For some non-pigmented lesions, diagnosis with the usually manifested dermatoscopically as an orange or
unaided eye is easier than with dermatoscopy. Other yellow structureless area, which represents dried serum
non-pigmented lesions that are difficult to diagnose with crust (6.1). Bleeding due to ulceration will be seen at
the unaided eye have specific dermatoscopy features. dermatoscopy as either red clods or red structureless
However, in most cases a satisfactory level of diagnostic zones (6.2). Sometimes ulceration appears together
accuracy is only achieved by supplementing dermato- with necrosis. Necrosis can be white, yellow or black
scopic features with clinical findings. (6.3). Especially when a contact fluid is used, ulcer-
Because many inflammatory diseases have to be includ- ation may not be apparent dermatoscopically with the
ed in the list of differential diagnoses, the number of appearance of ulceration often mimicking compacted
possible diagnoses is higher than for pigmented lesions. keratin. However, adherent fiber, either clothing fab-
In this chapter we focus on neoplastic non-pigmented ric or loose hair, is an indirect dermatoscopic clue to
lesions and discuss inflammatory diseases only as dif- ulceration. Fibers adhere to ulcerated surfaces because
ferential diagnoses. The dermatoscopy of inflammatory of the sticky consistency of the serum.
diseases is dealt with in more detail in chapter 8.
Surface scale
The multiple air-tissue interfaces created by scale means
that more of the light incident on the skin surface is
reflected, making scale appear white. Like ulceration,
scale is usually better assessed clinically. Dermatoscopes
are designed to make surface scale more transparent,
204 Non-pigmented (amelanotic) lesions
i.e. invisible, to allow better visualization of pigment When visible, scale is seen on dermatoscopy as white
structures and vessels. or silvery polygonal clods that are not entirely homo-
If serum is present in the stratum corneum (e.g. in a geneous (6.4).
spongiotic dermatitis), scale appears yellow and not
white. Scale may be present both in inflammatory dis- Keratin
eases (e.g. psoriasis) and in neoplasms (e.g. Bowen’s While scale indicates mild hyper- and parakeratosis,
disease). In both cases, scale is produced by hyper- and keratin corresponds to prominent hyperkeratosis. Subsur-
parakeratosis of the stratum corneum, with conglomer- face keratin (e.g. in the “milia-like cysts” of seborrheic
ates of corneocytes (keratinocytes that form the stratum keratosis) has no contact with air and usually appears
corneum) with preserved nuclei remaining adherent to white on dermatoscopy. Surface keratin that has con-
the skin surface because of incomplete desquamation. tact with air usually appears yellow or orange (6.5).
Non-pigmented (amelanotic) lesions 205
This color is emphasized by any serum inclusions. In differential diagnosis of amelanotic lesions is further
non-pigmented (amelanotic) seborrheic keratoses the discussed in section 6.3.
infundibular keratin plugs (“comedo-like openings”)
are usually yellow. If melanin is admixed with keratin, There are two types of white lines. One is seen only with
keratin plugs can be even brown or black as in some polarizing dermatoscopes; the other is seen regardless
hyperpigmented seborrheic keratoses. of instrument type. Polarizing-specific white lines are
arranged as two groups of parallel lines at right angles
White clues to each other (6.7). These lines may be short or long,
In dermatoscopy, “white” is defined as lighter than but do not cross each other. They are most commonly
surrounding normal skin. All basic elements except found in melanoma, Spitz nevus, basal cell carcinoma,
pseudopods have a white variant. The various white and dermatofibroma (1–5), but can also occasionally
structures are listed in table 6.1, and their role in the be seen in a wide range of other lesions including scar
206 Non-pigmented (amelanotic) lesions
tissue (6). Polarizing-specific white lines correspond to and dermatofibromas but they are absent in basal cell
fibrosis and sclerosis in the dermis. Usually the overlying carcinomas.
epidermis is devoid of rete ridges (i.e. flat). While they can be seen in benign lesions, white lines
White lines are also produced by fibrosis of the papillary of any type in amelanotic lesions should be regarded
dermis when the rete ridges are intact. These white lines as a clue to malignancy (7).
often (but not always) form the reticular pattern (7), White circles are the most specific clue to actinic kera-
and are seen regardless of the type of dermatoscope toses (flat lesions) (8) and well-differentiated squamous
used. They are most often seen in melanomas, nevi, cell carcinomas/keratoacanthomas (raised lesions) (9)
Non-pigmented (amelanotic) lesions 207
but can also be found in lesions of cutaneous lupus ery- matoscopy (12, 13). Their white color is duller than
thematosus (10). White circles correspond to acanthosis polarizing-specific white clods or milia (6.11). Single
of follicular epithelium with prominent hypergranulosis white dots or clods may also represent pus in skin
(6.8). abscesses such as folliculitis (6.12) or furuncles, but
White dots or clods usually correspond to keratin filled also in myiasis (14).
cysts (milia). These are better seen with non-polarizing Dots and clods that are only visible with polarized
contact dermatoscopes (11). If multiple white dots or dermatoscopy (polarizing-specific white clods) do not
small round clods are present they usually point to a correspond to milia. Polarizing-specific white clods have
seborrheic keratosis (6.9) but are also seen in con- the same significance as polarizing specific white lines
genital nevi. Although multiple white dots and clods and correspond to dermal fibrosis and sclerosis. They
representing keratin filled cysts are a clue to seborrheic nearly always appear together with polarizing specific
keratosis they are not unusual in basal cell carcinoma white lines and are found in some basal cell carcinomas
(6.10) and are also occasionally found in melanoma. and melanomas (6.13, 6.14). Four white dots arranged
The white clods of sebaceous gland hyperplasia are in a square (four-dot clod) are a polarizing specific
visible with polarized and with non-polarized der- structure seen particularly in actinic keratoses, but also
208 Non-pigmented (amelanotic) lesions
in other lesions, and even on severely sun-damaged skin epidermal hyperplasia that surrounds the overgrowth
without any discrete lesion (15). With non-polarized of granulation tissue. For the sake of completeness it
dermatoscopes, this structure may be seen less clearly should be mentioned that necrotic tissue and calcinosis
as a single circle or clod. (6.16) may appear white on dermatoscopy.
A white structureless zone in a flat lesion usually cor-
responds to fibrosis or sclerosis. It can be found in Other clues
flat melanomas and basal cell carcinomas but also Scale, keratin, ulceration and white clues are not the
in other flat lesions including inflammatory conditions only clues that help diagnose non-pigmented lesions.
such as lupus erythematosus and in flat scars. A white Most clues have already been mentioned in other
structureless zone in a raised lesion usually represents chapters but the color yellow deserves special atten-
subsurface keratin as in well-differentiated squamous tion. Yellow color usually corresponds to keratin or
cell carcinomas/keratoacanthomas or in pilomatrixoma a serum crust, which indicates ulceration. A yellow
(6.15). structureless zone can also be found when there is a
A peripheral white rim can be found in some pyogenic dermal accumulation of macrophages that are replete
granulomas (16, 17). It corresponds to the reactive with lipids (xanthoma cells). These xanthoma cells can
Non-pigmented (amelanotic) lesions 209
Figure 6.10: Two basal cell carcinomas with white dots and clods.
Although multiple white dots and clods are a clue to seborrheic keratosis they may be found in basal cell carcinomas.
Figure 6.13: Basal cell carcinoma with polarizing specific white Figure 6.14: Basal cell carcinoma with four-dot clod (for white
lines and clods dots arranged in a square)
6.2 Vascular patterns serpentine, helical or coiled. When one vessel type
predominates, this is called a “monomorphous” pattern
In dermatoscopic assessment of pigmented lesions, of vessels. When more than one type of vessel is seen,
blood vessel morphology is only ever accorded the the pattern is called “polymorphous”. In addition to the
status of being a clue to diagnosis, as patterns formed type of vessels, their arrangement – both how vessels
by vessels (20) are less specific and hence less important are arranged relative to each other, and how vessels
than pigment patterns and colors. The patterns formed are distributed throughout the lesion – may also be of
by blood vessels are no more diagnostically specific diagnostic significance (6.20).
in amelanotic lesions, but in the absence of melanin In the majority of cases, vessels appear to be distributed
pigment, analysis of vessel patterns must assume greater randomly, i.e. not arranged in any specific manner
importance. throughout the lesion. Vessels as dots or coils may
The pattern of vessels is assessed using the principles be arranged in straight lines (linear arrangement) or
detailed in chapter 3. Vessels may be seen as dots, clods in serpentine lines (serpiginous arrangement). When
or lines (6.19). Lines may be straight, curved, looped, vessels as dots or coils are not uniformly distributed but
212 Non-pigmented (amelanotic) lesions
are denser at some sites than others, this arrangement er, there is a tendency for more vessels to be viewed
is termed “clustered”. Linear vessels of any type at the obliquely and thus seen as loops. As malignant neo-
periphery that are oriented towards but do not cross plasms become thicker, neovascularization becomes
the center are termed “radial”. The arrangement of more common.
linear vessels (most commonly curved, sometimes ser- This variation means the same vessel morphology may
pentine or looped) in the center of skin colored or light have different diagnostic significance in nodules com-
brown clods is termed “centered”. Straight linear vessels pared to flat lesions.
that intersect each other nearly at right angles have a
“reticular” arrangement. Finally, serpentine vessels may
be arranged such that multiple vessels originate from 6.3 Differential diagnosis of non-pigmented
one common vessel; the derivative vessels typically lesions
originate from a thicker vessel. This arrangement is
termed “branched”. General principles
Vessel morphology varies with lesion thickness. The As a general principle, even in a largely non-pigmented
capillary loops that rise from the superficial vascular lesion, if there is any pigment at all that can be attribut-
plexus and extend towards the surface of the skin may ed to melanin (black, brown, blue or gray) one should
appear as dots or curved or looped lines, depending first attempt to diagnose a lesion using a pigmented
on the angle from which they are viewed (6.21). In flat lesion algorithm (6.22). Only if there truly is no pigment
lesions, most vessels are viewed end on and so appear or if the pigmented features present are non-specific,
as dots or short curved lines. As a lesion becomes thick- should a non-pigmented algorithm be used. The meth-
Non-pigmented (amelanotic) lesions 213
od we present requires the integration of clinical and Scale is an important hallmark of Bowen’s disease
dermatoscopy features to reach an acceptable level and actinic keratosis but is obviously also found in
of diagnostic accuracy. inflammatory conditions. On the rare occasions scale
As a matter of convenience, clinical features are usually is seen in superficial basal cell carcinoma or melano-
assessed before dermatoscopy. These findings are then cytic lesions, it is usually a consequence of irritation
included in the diagnostic process as one proceeds such as rubbing or scratching. Occasionally nevi show
with dermatoscopy. a spongiotic reaction that leads to scaling (21). With
The main features assessed clinically are whether the severe chronic sun damage, the entire skin surface may
lesion is flat or raised; whether it is solitary or one be scaly, including that overlying lesions.
of many; and the presence or absence of ulceration, If keratin is present, the main differential diagnoses
scale, and keratin. include well-differentiated squamous cell carcinomas/
It is critical to differentiate between flat and raised keratoacanthomas, seborrheic keratoses and viral warts
non-pigmented lesions. When we speak of flat lesions (22). Keratin can also be found in Unna or Miescher
we do not mean that the lesion must be so flat as to nevi (keratin plugs on the surface between papillomatous
be impalpable. Rather, a lesion is termed flat when the invaginations), in keratinizing adnexal proliferations
horizontal diameter greatly exceeds height. Macules, such as pilomatrixoma (subsurface keratin) (23), in
flat papules and patches are flat whereas elevated keratinizing cysts (subsurface keratin), and in angioker-
papules and nodules are raised. atoma (surface keratin) (24). After clinical assessment
While it is true that neoplasms tend to be solitary and one then proceeds to dermatoscopy.
inflammatory conditions tend to be multiple, this is
not always the case. In particular, actinic keratosis is Dermatoscopy of non-pigmented lesions
often multiple. Most critically, a solitary non-pigmented The first step in assessing non-pigmented lesions is to
neoplasm – most commonly Bowen’s disease, but rarely decide whether they are flat or raised. The vascular
even a non-pigmented melanoma – may be concealed pattern is more diagnostically significant in flat lesions
amongst multiple patches of psoriasis. than in raised lesions. In raised lesions, other clues
As already mentioned, ulceration does not suggest a (ulceration, keratin, and white clues) take priority over
specific diagnosis but should (in the absence of trauma) vessel pattern analysis, just as pigmented structures
prompt the consideration of malignancy. take priority for pigmented lesions.
214 Non-pigmented (amelanotic) lesions
Melanocytic nevus
Dots
Inflammatory skin diseases (e.g. psoriasis)
(Bowen’s disease)
Hemangioma
Clods
Vascular malformation
Hemorrhage
Serpentine
Basal cell carcinoma
Monomorphous
vascular pattern Coiled Bowen’s disease
(inflammatory skin diseases)
Flat
Vessels as
Polymorphous Exclude melanoma
dots present
vascular pattern
Figure 6.23: Algorithm for flat non-pigmented lesions with visible blood vessels
Flat non-pigmented lesions Next, one decides whether the vascular pattern is
Because nearly all pigmented lesions may also appear monomorphous or polymorphous.
in a non-pigmented form, the differential diagnosis for When there is a monomorphous pattern of vessels, a
flat non-pigmented lesions encompasses nearly the distinction is made between vessels as dots, clods, and
entire spectrum of melanocytic and non-melanocytic vessels as lines. In cases of vessels as dots, the differen-
lesions discussed in chapter 2. In addition to this spec- tial diagnosis comprises Bowen’s disease, inflammatory
trum of neoplasms, various inflammatory skin diseases skin diseases such as psoriasis, and benign melanocytic
must also be considered. Melanocytic lesions that may lesions. There are exceptional cases of flat amelanotic
appear as non-pigmented skin-colored to red macules melanomas on chronic sun-damaged skin that have a
or patches are Clark nevi, “superficial” or “superficial monomorphous pattern of dots, but a flat amelanotic
and deep” congenital nevi, Spitz nevi and, of course, melanocytic lesion with a monomorphous vascular
melanoma. Keratinocytic cancers (actinic keratosis, pattern of dots is nearly always a nevus.
Bowen’s disease, superficial basal cell carcinoma) and Scale (not always visible on dermatoscopy) is usually
many inflammatory skin diseases, for example psoriasis, present in inflammatory lesions and Bowen’s disease
nummular dermatitis, porokeratosis, lupus erythematosus but not in melanocytic lesions. Further differentiation
and lichen planus occur mainly as flat pink lesions. is then performed as far as possible on the basis of
We will discuss the dermatoscopic appearance of the clinical context and additional clues. The scale of
inflammatory lesions in greater detail in chapter 8. psoriatic lesions is practically always white, whereas
Rarely, even seborrheic keratosis and dermatofibroma the scale of different types of dermatitis (for example
may be flat and non-pigmented. The most common nummular dermatitis or seborrheic dermatitis) is mixed
flat non-pigmented lesions and their appearance on with serum and appear yellow or orange (25). The scale
dermatoscopy are shown in table 6.2. of porokeratosis usually presents as a peripheral rim (26,
If vessels are seen, there is a relatively simple algo- 27) that should not be confused with delicate peripheral
rithm for flat non-pigmented lesions, starting with the pigmentation of some flat basal cell carcinomas. The
assessment of vascular patterns (6.23). While it has differentiation between psoriasis and Bowen’s disease
proved useful in the hands of the authors, it lacks the (intraepidermal carcinoma) can be challenging. The
specificity of algorithms to assess pigmented lesions. It vessels of Bowen’s disease are usually coils (28) and
should also be seen as evolving, rather than an algo- those of psoriasis usually dots (29) (6.24). Sometimes,
rithm carved in stone. however, the coils of Bowen’s disease are so small
Non-pigmented (amelanotic) lesions 215
that the vessels appear as dots and in long standing, As well as Bowen’s disease, coiled vessels may be
elevated lesions of psoriasis the vessels may appear seen in psoriasis, lichen simplex chronicus, and other
as coils. In psoriasis the vessels tend to be randomly inflammatory diseases. Serpentine vessels are seen in
distributed over the lesions whereas in Bowen’s disease superficial basal cell carcinoma (6.25), but this diagnosis
they tend to be arranged in clusters or lines. In contrast is more reliable when other specific clues to basal cell
to Bowen’s disease (often dull white or yellow scales) carcinoma are present.
the scales of psoriasis tend to be shiny white. The reticular pattern of vessels is not included in the algo-
As discussed in Chapter 3, a monomorphous pattern rithm because it is too unspecific (it occurs, for instance,
of red clods (vessels as clods) indicates a hemangioma on chronic sun-damaged skin). Thin reticular vessels
or a vascular malformation. are found in lesions of a specific type of mastocytosis,
216 Non-pigmented (amelanotic) lesions
telangiectasia macularis eruptiva perstans (30–32). lesions with more than one pattern. The investigator
Thick reticular vessels are found in “spider nevus” (nevus first determines whether vessels as dots are present.
araneus). A central dot vessel (the supplying arteriole) As in all vascular patterns, a few vessels as dots are
is commonly seen, sometimes with visible pulsations. not significant; to constitute a pattern they must cov-
Dermatofibromas are occasionally non-pigmented (4). er a significant part of the lesion. In 6.26 and 6.27
If they are flat they usually show a vascular pattern we show flat non-pigmented lesions with (6.26) and
of dots or coils, and may also show a typical central without (6.27) vessels as dots. If the vascular pattern
white structureless area or central polarizing-specific is polymorphous and includes vessels as dots (figure
white lines. 6.26, middle and bottom row) a melanoma cannot
When the vascular pattern is polymorphous, one should be excluded with certainty and the lesion should be
proceed in a stepwise manner, as in cases of pigmented submitted for histopathology. If it is difficult to decide
Non-pigmented (amelanotic) lesions 217
whether the pattern is one of dots or small coils, it is cell carcinomas, scale is more common in Bowen’s
prudent to assume they are dots and thus keep mela- disease. Seborrheic keratosis may have all types of
noma in the differential diagnosis. vessels as lines, including looped vessels (34). However,
When there are no vessels as dots but polymorphous in most cases there will be one or more of white dots
linear vessels (including coiled, serpentine, and looped and white, yellow or orange clods.
vessels) the investigator should consider superficial When no vessels are visible or when a diffuse erythema
basal cell carcinoma, Bowen’s disease or seborrheic is seen in a flat lesion, dermatoscopy is of no significant
keratosis. In cases of superficial basal cell carcinoma benefit unless other clues are present. Actinic keratoses
the predominant structures are thin serpentine vessels (6.28) often have an erythematous background without
(33), whereas Bowen’s disease is marked by coiled discernable vessels. Facial actinic keratosis may have
vessels (28). Ulceration is more common in flat basal white circles (8), but white circles should also lead to
220 Non-pigmented (amelanotic) lesions
a consideration of invasive squamous cell carcinoma The principal nodular non-pigmented lesions and their
(9). Erythema and white circles are also found in lupus dermatoscopic appearances are listed in table 6.3.
erythematosus (10). When sclerosis becomes prominent, Based on the descriptions of individual lesions it becomes
lesions of lupus erythematosus show white structureless clear that the morphology of vessels assists little in
zones, especially in the center. The polarizing specific diagnosis. Unlike flat lesions, the distinction between
“four dot clod” can be found in actinic keratosis and monomorphous and polymorphous patterns of vessels
in lupus erythematosus (15). has no diagnostic significance for nodules. With the
exception of some specific arrangements discussed
Nodular non-pigmented lesions below, vascular patterns have poor specificity for nodu-
The assessment of non-pigmented nodular lesions is lar non-pigmented lesions. Therefore, ulceration, keratin
one of the biggest challenges in dermatoscopy. The and white clues are given priority over vessel pattern
differential diagnosis to be taken into account is vast. analysis when assessing nodules.
Benign nodular non-pigmented non-melanocytic lesions As stated previously, in the absence of a clear and
include seborrheic keratoses (6.29), dermatofibromas convincing history of trauma, any solitary ulcerated
(4), warts (22) (including molluscum contagiosum, 6.30 non-pigmented nodule should be submitted for histo-
middle), infectious nodules (for example Leishmaniasis pathology. In most ulcerated non-pigmented nodules
(19)), all forms of prurigo (35) (for example picker’s it will not be possible to come to a specific diagnosis
nodule), angioma (including pyogenic granuloma, 6.3), by dermatoscopy. Nodular basal cell carcinomas are
sebaceous gland hyperplasia (12, 13) (6.11), clear cell the commonest nodular malignancy and they are fre-
acanthoma (36) (6.31), common benign neoplasms quently ulcerated, but any other malignant neoplasm
such as pilomatrixoma (23) (6.15) and rare ones such like Merkel cell carcinoma and melanoma can also
as eccrine poroma (37, 38) (6.32) or trichoepithelioma be ulcerated. Benign conditions that may present as
(39) (6.33), and all kind of cysts including epidermal ulcerated nodules include pyogenic granulomas, nodular
cysts or mucoid (myxoid) finger cysts (40) (6.34). prurigo (picker’s nodule) and some infectious diseases
Non-melanocytic malignancies include poorly (41) (6.30 (e.g. leishmaniasis).
top, 6.35 bottom left) and well differentiated squamous If surface keratin is present in an amelanotic nodule,
cell carcinomas (9) (including keratoacanthoma, 6.35, the principal diagnoses are well-differentiated squa-
top left), basal cell carcinoma (33) (6.36), rare cuta- mous cell carcinoma/keratoacanthoma, warts, and
neous malignancies like Merkel cell carcinoma (42) seborrheic keratosis. The presence of keratin helps to
(6.37), and cutaneous metastases of any malignancy. distinguish well-differentiated squamous cell carcinomas
In melanocytic lesions the differential diagnoses are or keratoacanthomas from other raised non-pigmented
limited to Unna or Miescher nevus (6.38), Spitz nevus neoplastic lesions such as basal cell carcinomas, Merkel
and melanoma (6.35 bottom right, 6.39). cell carcinomas and amelanotic melanomas. Unna or
Miescher nevus or papillomatous parts of congenital Blood spots in keratin are a good clue to the diagnosis
nevi may show surface keratin in the epidermal invagi- of well-differentiated squamous cell carcinomas and
nations. Pilomatrixoma is typified by subsurface keratin, keratoacanthomas (6.6).
which is characterized by a white structureless zone While keratin is a good clue to well differentiated squa-
dermatoscopically. The subsurface keratin of epidermal mous cell carcinomas, poorly differentiated squamous
cysts is hardly ever visible by dermatoscopy. Keratin cell carcinomas usually do not produce keratin.
may also be present in viral warts and angiokeratoma. Occasionally dermatoscopy does not permit a clear
A single central plug of keratin usually indicates a distinction between a central keratin plug and a cen-
keratoacanthoma, but it is not possible to reliably dis- tral ulcer, particularly when a keratin plug contains
tinguish between well-differentiated squamous cell bloods spots. This is an important distinction because
carcinoma and keratoacanthoma dermatoscopically. radially arranged vessels may be found around an
Non-pigmented (amelanotic) lesions 225
ulcer, regardless of the lesion’s etiology. For example, Miescher nevus. Centered vessels may also be found in
ulcerated basal cell carcinomas may show radial vessels seborrheic keratoses. The clods only pattern is specific
(6.35 top, right). for hemangioma. It is crucial that no vessels as lines are
Polarizing specific white lines and white clods can be present as this can be a clue to malignancy.
found in malignant neoplasms (e.g. basal cell carcino- The differential diagnoses of the other specific vessel
ma, melanoma) and in some benign conditions (e.g. arrangements include both benign and malignant tumors.
dermatofibroma, Spitz nevus, pyogenic granuloma). Radially arranged vessels are found in sebaceous gland
Reticular white lines may be present in melanoma or hyperplasia (additional clue: several white or yellow
Spitz nevi and in some dermatofibromas. As a general clods in central location), in molluscum contagiosum
rule, non-pigmented nodules with white lines should (additional clue: singular skin-colored or orange clod
be biopsied or excised unless an unequivocal benign in central location) and in keratoacanthoma (additional
diagnosis can be made based on clues beyond der- clue: a keratin plug in central location, which may be
matoscopy. white, yellow or orange). The radially arranged vessels
White circles in a nodular lesion point to keratoacan- in keratoacanthoma may be looped, linear, curved or
thoma/well-differentiated squamous cell carcinoma. branched-serpentine. The specificity of branched ser-
Rarely, other lesions may show white circles, for example pentine vessels for basal cell carcinoma is frequently
some basal cell carcinomas. overestimated. Of course basal cell carcinoma is the
If ulceration, keratin and white clues are absent the most common diagnosis in cases of non-pigmented
specific diagnosis of nodular lesions becomes even nodular lesions with branched vessels, but in princi-
more challenging. ple any invasive tumor in the dermis lying below the
In some cases, a specific vascular arrangement will superficial vascular plexus may have this pattern of
help to narrow down the differential diagnosis. Three vessels. This is true for cysts, benign adnexal tumors
arrangements have high specificity for benign condi- (e.g. trichoblastoma or poroma), keratoacanthoma/
tions, allowing these lesions to be excluded from further well differentiated squamous cell carcinoma and other
assessment. types of neoplasms. It is true for the rare Merkel cell
The serpiginous arrangement of coiled vessels is so carcinoma, and cutaneous metastases regardless of the
specific for clear cell acanthoma that no other diagnosis origin of the primary malignancy. In the absence of a
needs to be considered. confident specific benign diagnosis, non-pigmented
Linear vessels seen in the center of polygonal skin-col- nodules with branched serpentine vessels should be
ored clods is termed a centered arrangement. These submitted for histopathology.
vessels may be polymorphous. When this includes thick Eccrine poroma is characterized by a combination of
curved vessels, this is indicative of an Unna nevus or a coiled, serpentine and branched vessels. The peculiar
226 Non-pigmented (amelanotic) lesions
References
1 Agero AL, Taliercio S, Dusza SW, Salaro C, Chu P, 17 Zaballos P, Llambrich A, Cuellar F, Puig S, Malvehy
Marghoob AA. Conventional and polarized dermoscopy J. Dermoscopic findings in pyogenic granuloma. Br J
features of dermatofibroma. Arch Dermatol. 2006; 142: Dermatol. 2006; 154: 1108–1111.
1431–1437. 18 Song M, Kim SH, Jung DS, Ko HC, Kwon KS, Kim MB.
2 Arpaia N, Cassano N, Vena GA. Dermoscopic patterns of Structural correlations between dermoscopic and histo-
dermatofibroma. Dermatol Surg. 2005; 31: 1336–1339. pathological features of juvenile xanthogranuloma. J Eur
3 Puig S, Romero D, Zaballos P, Malvehy J. Dermoscopy Acad Dermatol Venereol. 2011; 25: 259–263.
of dermatofibroma. Arch Dermatol. 2005; 141: 122. 19 Llambrich A, Zaballos P, Terrasa F, Torne I, Puig S, Mal-
4 Zaballos P, Puig S, Llambrich A, Malvehy J. Dermoscopy vehy J. Dermoscopy of cutaneous leishmaniasis. Br J
of dermatofibromas: a prospective morphological study Dermatol. 2009; 160: 756–761.
of 412 cases. Arch Dermatol. 2008; 144: 75–83. 20 Argenziano G, Zalaudek I, Corona R, et al. Vascular
5 Zaballos P, Puig S, Malvehy J. Dermoscopy of atypical structures in skin tumors: a dermoscopy study. Arch
dermatofibroma: central white network. Arch Dermatol. Dermatol. 2004; 140: 1485–1489.
2006; 142: 126. 21 Meyerson LB. A peculiar papulosquamous eruption
6 Balagula Y, Braun RP, Rabinovitz HS, et al. The signifi- involving pigmented nevi. Arch Dermatol. 1971; 103:
cance of crystalline/chrysalis structures in the diagnosis 510–512.
of melanocytic and nonmelanocytic lesions. J Am Acad 22 Tschandl P, Rosendahl C, Kittler H. Cutaneous human
Dermatol. 2012; 67: 194 e191–198. papillomavirus infection: manifestations and diagnosis.
7 Lozzi GP, Piccolo D, Micantonio T, Altamura D, Peris Curr Probl Dermatol. 2014; 45: 92–97.
K. Early melanomas dermoscopically characterized by 23 Zaballos P, Llambrich A, Puig S, Malvehy J. Dermoscopic
reticular depigmentation. Arch Dermatol. 2007; 143: findings of pilomatricomas. Dermatology. 2008; 217:
808–809. 225–230.
8 Zalaudek I, Giacomel J, Argenziano G, et al. Dermoscopy 24 Zaballos P, Daufi C, Puig S, et al. Dermoscopy of solitary
of facial nonpigmented actinic keratosis. Br J Dermatol. angiokeratomas: a morphological study. Arch Dermatol.
2006; 155: 951–956. 2007; 143: 318–325.
9 Rosendahl C, Cameron A, Argenziano G, Zalaudek 25 Lallas A, Argenziano G, Apalla Z, et al. Dermoscopic
I, Tschandl P, Kittler H. Dermoscopy of squamous cell patterns of common facial inflammatory skin diseases.
carcinoma and keratoacanthoma. Arch Dermatol. 2012; J Eur Acad Dermatol Venereol. 2014; 28: 609–614.
148: 1386–1392. 26 Delfino M, Argenziano G, Nino M. Dermoscopy for the
10 Lallas A, Apalla Z, Lefaki I, et al. Dermoscopy of discoid diagnosis of porokeratosis. J Eur Acad Dermatol Venereol.
lupus erythematosus. Br J Dermatol. 2013; 168: 284–288. 2004; 18: 194–195.
11 Benvenuto-Andrade C, Dusza SW, Agero AL, et al. 27 Pizzichetta MA, Canzonieri V, Massone C, Soyer HP.
Differences between polarized light dermoscopy and Clinical and dermoscopic features of porokeratosis of
immersion contact dermoscopy for the evaluation of skin Mibelli. Arch Dermatol. 2009; 145: 91–92.
lesions. Arch Dermatol. 2007; 143: 329–338. 28 Zalaudek I, Argenziano G, Leinweber B, et al. Dermos-
12 Bryden AM, Dawe RS, Fleming C. Dermatoscopic features copy of Bowen’s disease. Br J Dermatol. 2004; 150:
of benign sebaceous proliferation. Clin Exp Dermatol. 1112–1116.
2004; 29: 676–677. 29 Pan Y, Chamberlain AJ, Bailey M, Chong AH, Haskett
13 Zaballos P, Ara M, Puig S, Malvehy J. Dermoscopy of M, Kelly JW. Dermatoscopy aids in the diagnosis of the
sebaceous hyperplasia. Arch Dermatol. 2005; 141: 808. solitary red scaly patch or plaque-features distinguishing
14 Bakos RM, Bakos L. Dermoscopic diagnosis of furuncular superficial basal cell carcinoma, intraepidermal carci-
myiasis. Arch Dermatol. 2007; 143: 123–124. noma, and psoriasis. J Am Acad Dermatol. 2008; 59:
15 Liebman TN, Scope A, Rabinovitz H, Braun RP, Marghoob 268–274.
AA. Rosettes may be observed in a range of conditions. 30 Akay BN, Kittler H, Sanli H, Harmankaya K, Anadolu
Arch Dermatol. 2011; 147: 1468. R. Dermatoscopic findings of cutaneous mastocytosis.
16 Zaballos P, Carulla M, Ozdemir F, et al. Dermoscopy Dermatology. 2009; 218: 226–230.
of pyogenic granuloma: a morphological study. Br J 31 Arpaia N, Cassano N, Vena GA. Lessons on dermoscopy:
Dermatol. 2010; 163: 1229–1237. pigment network in nonmelanocytic lesions. Dermatol
Surg. 2004; 30: 929–930.
234 Non-pigmented (amelanotic) lesions
As we have seen in previous chapters, clues are import- Branched fine lines in flat acral lesions
ant hints that help to solve the differential diagnosis Brown or gray branched fine lines sprinkled with dots
produced by analysis of pattern and color. They usually in a flat acral lesion is a very distinctive pattern. When
point towards a diagnosis, but very few clues are spe- uniformly distributed over the whole lesion (7.2), it is
cific for a particular diagnosis in all contexts. Studies pathognomonic for tinea nigra (1). As tinea nigra com-
which state that a particular clue has a given sensitivity monly occurs on the feet where surgery is technically
and specificity should be interpreted with caution as difficult, the diagnosis is best confirmed by a successful
there is always a selection bias (usually admitted) in trial of treatment with topical antifungal cream. This
the choice of included lesions, the series may not be leads to resolution of the lesion within 3 weeks and
large, and all too often the clue is only evaluated in a avoids a biopsy to exclude melanoma.
limited context, most commonly of distinguishing nevus
from melanoma. Clues must always be interpreted in Branched serpentine vessels adjacent to keratin
context, otherwise a good clue may become a cliché. The presence of branched serpentine vessels adja-
This chapter will look at a selection of clues that deserve cent to keratin is a strong clue to keratoacanthoma
special attention, and at some common clichés. (7.3). Commonly keratoacanthomas are symmetrical
lesions with a central keratin plug and vessels (linear,
looped, serpentine, coiled or polymorphous) arranged
7.1 Clues in a radial pattern and with this morphology they are
easily identified (2). The clue of branched serpentine
Adherent fiber vessels adjacent to keratin is particularly useful when
Adherent fiber is a dermatoscopic clue to ulceration. the presentation is not typical, but it is also often present
Ulceration may of course be caused by trauma to normal in the more typical cases. There is an ongoing contro-
skin or benign lesions, but malignant neoplasms and versy among different schools of dermatopathologists
especially basal cell carcinomas may ulcerate after whether keratoacanthomas are benign lesions or highly
trivial irritation. The serum or blood that leaks onto the differentiated variants of squamous cell carcinoma (3).
skin has adhesive properties which persist when it dries However, even the proponents of the concept that a
out and this may trap fibers of clothing fabric, other keratoacanthoma is a benign neoplasm agree that
exogenous debris or the patient’s own dislodged hair. lesions that appear as keratoacanthomas clinically
Adherent fiber may be found on basal cell carcinomas or dermatoscopically should be excised to rule out
even when ulceration was not observed prior to der- squamous cell carcinoma.
matoscopy. As ulceration is often an important clue to
malignancy, so is the presence of dermatoscopically
observed adherent fiber (7.1).
236 Clues and Clichés
Circles, ovals, and distorted circles it usually indicates a benign lesion. It usually indicates
A pattern of circles on non-facial skin, not correlating a junctional Clark nevus (7.6). If, on the other hand,
to infundibulae and in conjunction with ovals and dis- the rete ridges are broadened and filled with pigment
torted circles is a strong clue to a seborrheic keratosis (filled with neoplastic melanocytes) the hypopigment-
(7.4). The circles, ovals and distorted circles are pro- ed space becomes pigmented and instead of double
duced by elongation and broadening of rete ridges lines one sees thick reticular lines, which is a clue to
due to acanthosis of the epidermis (7.5). Pigmentation melanoma. Occasionally one finds thick reticular lines
of these structures is due to hyperpigmentation of basal in a seborrheic keratosis. In this case the rete ridges
keratinocytes. are not filled with neoplastic melanocytes but with
pigmented keratinocytes.
Double reticular lines
Occasionally it is apparent on higher magnification Four dots in a square (four-dot clod)
that the individual “lines” forming a reticular pattern These structures were first described by Marghoob and
are actually double lines enclosing a hypopigmented Cowell (4) who called them “rosettes” (see also chapter
space. The double line corresponds to pigmented basal 4). They are only seen with a polarizing dermatoscope
keratinocytes. The hypopigmented space between the (7.7), and are composed of 4 white dots arranged in
pair of lines indicates that the rete ridges are not filled a square or as a rhomboid. With a non-polarizing
with pigment. If this pattern is found throughout a lesion dermatoscope, this structure is seen as a simple white
Clues and Clichés 239
Figure 7.4: Circles, ovals and distorted circles as a clue to seborrheic keratosis.
Circles, ovals, and distorted circles in two seborrheic keratoses. In the bottom lesion one can see the transformation of circles to parallel
curved lines.
Figure 7.7: Four dots in a square (four-dot clod) as clue to actinic keratosis or superficial squamous cell carcinoma.
Left: With polarized light the 4-dots are seen very clearly on the pigmented portion of this actinic keratosis. Right: Four dots in a square as
a clue to superficial squamous cell carcinoma (Bowen’s disease) in this case in collision with a solar lentigo which accounted for the reticular
lines.
Clues and Clichés 241
A C
clod. In the appropriate context it is a clue to actinic keratosis and pigmented actinic keratosis. When the
keratosis or superficial squamous cell carcinoma but gray is seen as thin gray circles, this is a far stronger
it is not as specific as initially thought. It can even be clue to melanoma than gray dots, even when the dots
found on normal skin. are arranged as circles (7.8).
white lines have also been termed “reticular depigmen- Angulated lines (polygons)
tation” (6) or “negative pigment network” (7) but for Angulated lines (polygons) were first described as a clue
reasons of clarity we use only the objective language to flat melanomas on non-facial, chronic sun-damaged
of pattern analysis. skin by Keir (13) (7.12). These melanomas often mimic
Most white lines correspond to fibrosis or sclerosis in solar lentigo and may lack other, more conventional,
the dermis, some to hypergranulosis (for example the melanoma clues. As defined by Keir, angulated lines
white lines seen in lichen planus), and some to a com- (polygons) form multi-sided geometrical shapes which
bination of both (8). Superficial fibrosis (i.e. fibrosis in may be completely or incompletely enclosed. In a
the papillary dermis) and hypergranulosis are seen as recent study by Jaimes and Keir (14) angulated lines
white lines regardless of whether the dermatoscope were found in 44 % of flat melanomas on non-facial,
uses polarized or non-polarized light. In other situa- chronic sun-damaged skin. Facial angulated lines have
tions, white lines are only seen when dermatoscopes been termed rhomboids (15) or zig-zag pattern (16)
with a polarizing light source are used (9). Polarizing by others. Like angulated lines on non-facial skin, they
specific white lines are seen as two groups of parallel are a clue to melanoma. Angulated lines on facial skin
lines, with the groups at right angles to each other can also be found in pigmented solar keratosis (5).
(perpendicular white lines), and correspond to fibrosis
in deeper parts of the dermis. White circles
In the short “Chaos and Clues” algorithm presented in White circles are a clue to actinic keratosis and squa-
chapter 5, any white lines seen either with polarizing mous cell carcinoma (including keratoacanthomas) (2)
or non-polarizing dermatoscopy are a clue to malig- especially on, but not limited to the face (7.13). In flat
nancy if they are whiter than normal skin. In pattern pigmented lesions on the face which have evenly dis-
analysis, the interpretation of white lines depends on tributed gray dots (differential diagnosis: melanoma in
the context. White lines are not highly specific, being situ, lichen planus-like keratosis, and pigmented actinic
seen commonly in melanomas (7) and Spitz nevi (10, 11) keratosis) the presence of white circles helps to make
(7.9), and basal cell carcinomas and dermatofibromas the diagnosis of pigmented actinic keratosis (5) (7.14).
(12) (7.10). White lines are seen occasionally in a wide
variety of other lesions. White reticular lines rule out
a basal cell carcinoma with a similar high degree of
certainty as pigmented reticular lines (7.11).
Clues and Clichés 243
Figure 7.11: White reticular lines as clue to differentiate melanoma from pigmented basal cell carcinoma.
The clinical image on the left may be interpreted as pigmented basal cell carcinoma or as melanoma. On dermatoscopy the lesion is cha-
otic with both polarizing specific and reticular white lines (6 o’clock). Reticular white lines rule out basal cell carcinoma. The diagnosis is
melanoma (invasive, < 1 mm).
from further assessment to exclude melanoma. While White dots and clods (“milia”) indicate a seborrheic
in most cases this clue does correctly identify solar keratosis
lentigines or flat seborrheic keratoses, this pattern may White dots and clods (milia) are produced by small
also be seen in melanoma. The example shown in intraepidermal accumulations of keratin. Although white
figure 7.16 very emphatically illustrates the danger dots and clods are most commonly found in seborrhe-
of interpreting patterns and clues without taking into ic keratosis, they also occur in malignancies such as
account the whole context. basal cell carcinoma and melanoma (7.17). Seborrheic
keratosis should never be diagnosed on the basis of
white dots and clods alone.
246 Clues and Clichés
Figure 7.16: Curved lines (“fingerprint like structures”) as a misleading clue to solar lentigo/flat seborrheic keratosis.
This is a lesion with curved lines (“light-brown fingerprint like structures” in the metaphoric language) that could be easily discarded as a
solar lentigo/flat seborrheic keratosis. However, it is asymmetric with a clue to malignancy by virtue of gray dots. Furthermore, it lacks a
well-demarcated, scalloped border. This is melanoma in situ.
Figure 7.17: White dots (“milia”) can occur in any type of lesion, not only seborrheic keratosis.
Melanoma with multiple white dots and clods (“milia”). This lesion could easily be mistaken for a seborrheic keratosis if it is not assessed in
context. Clinically (left) this is a solitary lesion, whereas seborrheic keratoses usually are seen in large numbers. This anomaly should arouse
suspicion and lead to careful assessment. Dermatoscopically (right) there are no yellow or orange clods, the border is poorly demarcated,
and the vessels are neither looped nor centered in hypopigmented clods. Therefore there is little support for a proposed diagnosis of sebor-
rheic keratosis, even if white dots are the standout feature of this lesion. The presence of gray dots always raises the possibility of melanoma,
and this was confirmed on histopathology (invasive < 1 mm).
Clues and Clichés 247
Blue structureless area (“blue veil”) indicates Segmental pseudopods or segmental radial lines
melanoma indicate a melanoma
A blue structureless area is a good clue to differentiate Although pseudopods are admittedly a very strong clue
a melanoma from a nevus. It is not a good clue to to melanoma they can be found in other lesions too.
differentiate a melanoma from a seborrheic keratosis Figure 7.19 shows a seborrheic keratosis with pseudo-
because acanthotic seborrheic keratoses frequently pods. Segmental radial lines can be found in basal cell
show a blue structureless area (7.18). Other clues are carcinoma, pigmented Bowen’s disease, recurrent nevi,
needed in this context. Reed nevi, and rarely also in Clark nevi.
248 Clues and Clichés
A parallel furrows pattern indicates a benign acral vascular plexus may show serpentine branched vessels
lesion in dermatoscopy including other neoplasms (19), cysts
The absence of a parallel ridge pattern and the pres- (20), deposits, and inflammatory lesions (21) (7.21).
ence of a parallel furrow pattern in acral lesions does
not exclude acral melanoma (18) (7.20). The general Malignant neoplasms are chaotic
principles of pattern analysis also apply to acral lesions. Although most pigmented cutaneous malignant neo-
If there is chaos and a clue to melanoma then the lesion plasms are chaotic by dermatoscopy there are important
is suspicious for melanoma regardless of whether par- exceptions: Small melanomas, nodular melanomas, and
allel lines are situated in the furrows or on the ridges. flat facial and acral melanomas are often symmetrical
(7.22). Whilst in large and flat lesions chaos takes
Serpentine branched vessels indicate a basal cell precedence over clues, clues are more important than
carcinoma chaos in small (5 mm and less) and nodular lesions.
While it is true that most basal cell carcinomas, espe- Very rarely chaos may be absent even in large and flat
cially when they are nodular, have serpentine branched melanomas (7.23). In these cases, information beyond
vessels (“arborizing vessels”), it is not true that this dermatoscopy like the clinical context (“ugly duckling”)
arrangement of vessels is highly specific for basal cell or the history given by the patient (“changing lesion”)
carcinomas. Any tumor underneath the superficial may draw the attention of the examiner to the lesion.
Clues and Clichés 249
Figure 7.21: Benign and malignant neoplasms and cysts with serpentine branched vessels on dermatoscopy.
Serpentine branched vessels can be found in any tumor that is situated underneath the superficial vascular plexus, not only in basal cell
carcinomas. Four examples are trichoepithelioma (top left), eccrine hidrocystoma (top right), pilar sheath acanthoma (bottom left), Merkel
cell carcinoma (bottom right). Images courtesy of Nisa Akay, Jean-Yves Gourhant, Iris Zalaudek and Giuseppe Argenziano).
250 Clues and Clichés
Figure 7.22: Flat facial melanomas, nodular melanomas, and small melanomas may be symmetrical, not chaotic.
Clinical view on the left, dermatoscopic view on the right. Top row: Flat facial melanomas (usually in situ) may be symmetrical. The discrete
gray circles (not dots arranged as circles) on dermatoscopy allow the diagnosis of melanoma in situ with confidence. Middle row: Melano-
mas that grow quickly may present as nodules and often lack chaos. The diagnosis of melanoma can be suspected because of white lines
and a blue structureless area. In nodular lesions, clues are more important than chaos. Bottom row: Small melanoma (5 mm in diameter)
that is not chaotic on dermatoscopy but has gray and blue clods in the center. In small lesions clues are more important than chaos.
Clues and Clichés 251
Figure 7.23: The clinical context helps to diagnose melanomas that are not chaotic.
On the clinical overview (top) this pigmented lesion is different than the other pigmented lesions in this area (“ugly duckling”). The clinical
close-up image (bottom left) reveals a symmetrical brown plaque. On dermatoscopy there is no chaos (bottom right). There are, however,
thick reticular lines as clues to melanoma.
252 Clues and Clichés
References
1 Piliouras P, Allison S, Rosendahl C, Buettner PG, Weedon 15 Lallas A, Tschandl P, Kyrgidis A, et al. Dermoscopic clues to
D. Dermoscopy improves diagnosis of tinea nigra: a differentiate facial lentigo maligna from pigmented actinic
study of 50 cases. Australas J Dermatol. 2011; 52(3): keratosis. Br J Dermatol. 2016; 174(5): 1079–1085.
191–194. 16 Slutsky JB, Marghoob AA. The zig-zag pattern of lentigo
2 Rosendahl C, Cameron A, Argenziano G, Zalaudek maligna. Arch Dermatol. 2010; 146(12): 1444.
I, Tschandl P, Kittler H. Dermoscopy of squamous cell 17 Tschandl P, Rosendahl C, Kittler H. Accuracy of the first
carcinoma and keratoacanthoma. Arch Dermatol. 2012; step of the dermatoscopic 2-step algorithm for pigment-
148(12): 1386–1392. ed skin lesions. Dermatol Pract Concept. 2012; 2(3):
3 Selmer J, Skov T, Spelman L, Weedon D. SCCs and KAs 203a208.
are biologically distinct and can be diagnosed by light 18 Lallas A, Kyrgidis A, Koga H, et al. The BRAAFF check-
microscopy. Histopathology. 2016; 69(4): 535–41. list: a new dermoscopic algorithm for diagnosing acral
4 Marghoob AA, Cowell L, Kopf AW, Scope A. Observation melanoma. Br J Dermatol. 2015; 173(4): 1041–1049.
of chrysalis structures with polarized dermoscopy. Arch 19 Tschandl P. Dermatoscopic pattern of a spiradenoma.
Dermatol. 2009; 145(5): 618. Dermatol Pract Concept. 2012; 2(4): 204a209.
5 Tschandl P, Rosendahl C, Kittler H. Dermatoscopy of flat 20 Salerni G, Gonzalez R, Alonso C. Dermatoscopic pattern
pigmented facial lesions. J Eur Acad Dermatol Venereol. of digital mucous cyst: report of three cases. Dermatol
2015; 29(1): 120–127. Pract Concept. 2014; 4(4): 65–67.
6 Lozzi GP, Piccolo D, Micantonio T, Altamura D, Peris 21 Rudnicka L, Olszewska M, Rakowska A, Slowinska M.
K. Early melanomas dermoscopically characterized by Trichoscopy update 2011. J Dermatol Case Rep. 2011;
reticular depigmentation. Arch Dermatol. 2007; 143(6): 5(4): 82–88.
808–809.
7 Pizzichetta MA, Talamini R, Marghoob AA, et al. Nega-
tive pigment network: an additional dermoscopic feature
for the diagnosis of melanoma. J Am Acad Dermatol.
2013; 68(4): 552–559.
8 Pizzichetta MA, Canzonieri V, Soyer PH, Rubegni P,
Talamini R, Massone C. Negative pigment network
and shiny white streaks: a dermoscopic-pathological
correlation study. Am J Dermatopathol. 2014; 36(5):
433–438.
9 Braun RP, Scope A, Marghoob AA. The “blink sign” in
dermoscopy. Arch Dermatol. 2011; 147(4): 520.
10 Botella-Estrada R, Requena C, Traves V, Nagore E,
Guillen C. Chrysalis and negative pigment network in
Spitz nevi. Am J Dermatopathol. 2012; 34(2): 188–191.
11 Moscarella E, Lallas A, Kyrgidis A, et al. Clinical and
dermoscopic features of atypical Spitz tumors: A mul-
ticenter, retrospective, case-control study. J Am Acad
Dermatol. 2015; 73(5): 777–784.
12 Balagula Y, Braun RP, Rabinovitz HS, et al. The signifi-
cance of crystalline/chrysalis structures in the diagnosis
of melanocytic and nonmelanocytic lesions. J Am Acad
Dermatol. 2012; 67(2): 194 e191–198.
13 Keir J. Dermatoscopic features of cutaneous non-facial
non-acral lentiginous growth pattern melanomas. Der-
matol Pract Concept. 2014; 4(1): 77–82.
14 Jaimes N, Marghoob AA, Rabinovitz H, et al. Clinical and
dermoscopic characteristics of melanomas on nonfacial
chronically sun-damaged skin. J Am Acad Dermatol.
2015; 72(6): 1027–1035.
8 Special situations
8.1 Nails
Pigmentations of the nail plate are frequently – prob-
ably too frequently – biopsied in order to exclude a
melanoma of the nail matrix. However, the decision to
biopsy nail pigmentation should not be made lightly –
biopsies of the nail matrix are not only painful, they also
may cause irreversible abnormalities of nail growth. By
keeping a few basic principles in mind, the number of
biopsies can be reduced without increasing the risk of
missing a melanoma.
In principle, melanomas may occur anywhere in the
nail unit, but in practice almost all originate in the
nail matrix. Nail matrix pigmented neoplasms create
longitudinal pigmentation of the nail plate, known as
Figure 8.1: Anatomy of the nail plate.
longitudinal melanonychia. Melanomas that do not arise (Adapted from Ackerman and Boer, Histologic diagnosis of
in the nail matrix but in the nail bed do not produce inflammatory skin diseases. An algorithmic method based on pat-
longitudinal melanonychia. Initially the nail changes tern analysis, 3rd edition).
produced by nail bed melanoma may be mistaken for
“nail dystrophy” or “onychomycosis” and thus remain
unrecognized. Fortunately these are exceedingly rare Dermatoscopy of nail pigmentation
neoplasms. Dermatoscopy of the nail plate requires a high viscosity
The following applies exclusively to melanomas that (“stiff”) contact fluid like ultrasound gel. Thinner fluids
arise from the nail matrix. like paraffin oil and alcohol do not stay in place due
to the highly irregular contour of the nail organ. A con-
Anatomy of the nail tact fluid also improves visualization when performing
In order to understand nail pigmentation, one must dermatoscopy with polarized light (1–4).
understand the anatomy of the nail organ. The nail plate The most commonly seen pattern on dermatoscopy is
is a convex 0.5-mm-thick plate composed of keratin. It a parallel pattern of lines (8.2). In longitudinal mela-
is the final product of keratinocytes in the nail matrix. nonychia, these parallel lines extend continuously from
The nail plate lies on the nail bed. Its lateral margins the lunula to the free end of the nail plate. Longitudi-
are surrounded by the lateral nail fold while its proximal nal melanonychia usually occupies just a part, and
end is spanned by the proximal nail fold. These margins less frequently the entire width, of the nail plate. It is
are also known as the nail wall or the paronychium. The attributable to the increased production of melanin in
narrow cornified portion of proximal nail fold which the nail matrix, which may or may not be caused by
glides on 1 to 2 mm of the nail plate is known as the a proliferation of melanocytes.
cuticle. The nail matrix itself lies below the lunula, the Gray pigmentation of lines is usually attributable to an
white arc at the proximal end of the nail plate (8.1). increase of melanin, but it is not associated with prolifer-
ation of melanocytes in the nail matrix. The most common
causes of gray parallel lines are lentigines (melanotic
macules) of the nail matrix, ethnic hyperpigmentation,
drug induced hyperpigmentation, hyperpigmentation
during pregnancy, and traumatic or inflammatory hyper-
pigmentation (4). Traumatic hyperpigmentation occurs
on finger nails mainly due to manipulation of the nail
254 Special situations
A B C
Figure 8.6: History of a congenital nevus of the nail matrix in a prepubescent child.
The clinical image is top left, the other images show a chronologic sequence of dermatoscopic images. At the first visit (top right) the pig-
mentation is broader in the proximal nail plate than in the distal nail plate, which indicates that the lesion is still growing. One year later
(bottom left) the pigmentation in the proximal and distal nail plate have the same width, which indicates that the lesion stopped growing.
Three years later the nevus is much smaller.
Special situations 257
not occur in children (5). The exceptional reports of Fungal infections may be accompanied by a struc-
melanoma of the nail matrix in prepubescent children tureless brown discoloration of the nail; in very rare
are most probably congenital nevi that have been mis- cases there may also be parallel lines (8.8). This brown
diagnosed as melanomas. Congenital nevi of the nail pigmentation is due to the fact that fungi may produce
matrix (which may not be visible at birth) commonly melanin. Usually there are additional clues to fungal
show clues to melanoma including signs of growth. infections such as thickening of the nail, subungual
In growing lesions the pigmentation is broader in the hyperkeratoses and whitish discoloration of the nail,
proximal part of the nail plate and smaller in the distal which usually extends from proximal to distal. Sometimes
part (8.6). While this feature would be regarded as fungal microscopy or culture may be the only way to
a clue to malignancy in adults one should not be too establish the diagnosis. There are also pigment-forming
concerned if it occurs in children. Usually this sign of bacteria like for example Pseudomonas aeruginosa
growth will disappear if the lesion is monitored for (8.9), which is characterized by green discoloration
some months. of the nail plate (chloronychia).
Therefore, even if clues to malignancy are present, Rarely, melanomas of the nail organ do not produce
biopsy of the nail matrix in prepubertal children should longitudinal melanonychia. Melanomas of the nail
be performed very rarely, and only after careful con- bed and nail fold may produce other patterns of pig-
sideration. mentation, and amelanotic melanomas are of course
non-pigmented. Dermatoscopy features of these rare
Apart from parallel lines, nail pigmentation may also situations have not yet been described, and diagnosis
be seen as dots, clods or a structureless pattern. These must be based on clinical features.
patterns occur in isolation or in combination with par-
allel lines. In most cases dots, clods and structureless Biopsies of the nail matrix
areas are a sign of bleeding and therefore not caused Longitudinal melanonychia originates in the nail matrix,
by melanin but by hemoglobin (8.7). The distinction which means any biopsy must be performed not from
between hemorrhage and pigmentation of the nail due the nail bed, but from the matrix, at the proximal end
to melanin is usually made quite easily. Hemorrhage of the pigmentation. Dermatoscopy of the free end of
tends to be red or purple and the pigmentation appears the nail helps to identify the zone of the matrix that is
structureless. Pigmentation of the nail plate caused by responsible for the pigmentation of the nail plate (6).
melanin is usually brown or occasionally black, and When the pigmentation is mainly in the deeper layers
usually seen as longitudinal melanonychia, i.e. paral- of the nail plate, the biopsy should be performed in the
lel lines extending the whole length of the nail plate. distal part of the nail matrix. Pigmentation in the more
Parallel lines may also be found in hemorrhage, but superficial layers of the nail plate requires a biopsy from
not only will they be a different color, the lines do not the proximal part of the matrix. These biopsies from
extend continuously from the proximal to the free distal the proximal nail matrix are associated with a higher
end of the nail plate. Growth of the nail plate causes risk of disrupted nail growth.
pigmentation to advance distally. Pigmentation from In narrow lesions (less than 3 mm in diameter) biopsy
hemorrhage advances very slowly; in toe nails it may can be performed by punch excision. In these cases it
be as little as 1 mm per month. Over time, a growing, will be adequate to insert the punch through the nail
non-pigmented healthy piece of nail will be found plate into the nail matrix after folding back the proximal
between the pigmentation and the proximal nail fold. portion of the nail fold. For slightly larger lesions (3
The pigmentation produced by hemorrhage normally to 6 mm) located in the center of the nail, biopsy can
resolves spontaneously, regardless of the pattern seen. be by transverse excision or shave of the nail matrix,
Renewed bleeding may occur in case of repetitive trauma ideally including the entire lesion (1).
and prevent complete outgrowth of the pigmentation. For this purpose the nail plate is fenestrated and the
This is especially true for toe nails; at this site the trauma nail matrix exposed.
tends to remain unnoticed. In lesions larger than 6 mm diameter, usually one is
Because neoplasms may bleed, signs of hemorrhage only able to perform a partial biopsy – in most cases
do not entirely rule out a malignancy of the nail unit a punch biopsy – to confirm the diagnosis. Partial
(8.3). For this reason it is advisable to perform a fol- biopsies have a major disadvantage, as they make the
low-up examination after three months in patients with already demanding histological investigation even more
subungual hemorrhage who are unable to recall an difficult. False negative histopathology reports due to
explicit incident of trauma. inadequate partial biopsies are not rare.
258 Special situations
8.2 Acral lesions pattern in acral nevi of all types (8.11, 8.13); several
The diagnosis of acral lesions by dermatoscopy is types of furrow pattern exist (7, 9). Usually there is
based on the same principles as those at other loca- just one pigmented line in every furrow; but sometimes
tions, but the common patterns seen do vary by site. there are two. These single or double lines may be
While circles predominate on the face and reticular composed of dots, rather than being continuous. In
lines on the trunk and the proximal extremities, the another variant, the parallel lines in the furrows are
predominant pattern of acral melanocytic lesions is connected by crosswise lines (8.11, middle row, 8.13
parallel lines (8.10). middle row). The furrow pattern may be combined
Clinicians and dermatopathologists rarely agree on with dots or clods, more rarely even circles. In lesions
nomenclature, but both call a benign acral melano- with more than one pattern, the same principles apply
cytic proliferation an “acral nevus” without further as for other sites of the body, i.e. pattern and color
differentiation. In fact, while some nevi preferentially plus clues lead to the diagnosis.
occur at specific anatomical locations (Clark nevi are Beginners may find it difficult to distinguish between
found nearly exclusively on the trunk and the proximal ridges and furrows (10). However, with some practice
extremities, but very rarely on the face or acral skin), the distinction can be made quite easily because ridges
nearly all types of nevi may occur on acral skin (8.11, are wider than furrows, and the eccrine ducts (visible
8.12, 8.13). Whenever possible, a specific diagno- as hypopigmented dots) open out onto the ridges. In
sis such as Spitz, Reed, “superficial” or “superficial heavily pigmented acral lesions it may be difficult to
and deep” congenital nevus should be used instead identify the pattern because pigmentation is found
of the term “acral nevus”. There is only one type of both on ridges and in furrows. In these cases, the
nevus which has no counterpart at other locations, pattern can usually be identified as ridge or furrow at
the “classical acral nevus” (8.11). Clinically it is seen the lesions’ periphery, where pigmentation is lighter.
as a uniformly light-brown macule, at dermatopathol-
ogy one finds small melanocyte nests mainly at the
dermo-epidermal junction.
Although parallel lines are the most common pattern
of acral melanocytic proliferations, other patterns
may also be seen. Dots, clods and the structureless
pattern are seen quite frequently, circles and reticular
lines less often, radial lines and pseudopods rarely.
Parallel lines may be seen in the furrows, on the ridg-
es, or crossing both furrows and ridges. One pattern
may merge into another, most commonly the furrow
pattern merging with the crossing pattern. The cross-
ing pattern preferentially occurs at sites of greatest
pressure, i.e. the heels and the lateral part of the sole
(7, 8). While the furrow pattern is the most common
Special situations 261
The general principles to distinguish nevi from mela- (more than one pattern and more than one color
nomas at other sites also apply to acral lesions (8.14). arranged asymmetrically) with at least one additional
Analogous to in situ melanomas on the face or the clue to melanoma. In general, as acral melanomas
trunk, which may only demonstrate circles or reticular become thicker, they more closely resemble melanomas
lines, acral in situ melanomas may show only one found at non-acral sites. Importantly, when clues to
pattern, namely parallel lines on the ridges (11, 12). melanoma are seen in large lesions with more than
This pattern is the most important melanoma-specific one pattern, parallel lines in the furrows do not rule
clue on acral skin, and is so specific that even in the out an acral melanoma (8.14 bottom, 8.16).
absence of chaos one should consider biopsy of acral In addition to melanocytic proliferations, hemorrhage
lesions with parallel lines on the ridges (8.15, top row). may create pigmented lesions on acral skin. When
Of course any other melanoma-specific clue may also blood flows out of the dermis it usually reaches the
be seen in acral melanomas (8.15, middle and bottom epidermis along the outside of the eccrine duct open-
row). Invasive acral melanomas usually are chaotic ings. A small bleed will be limited to the ridges (the
Special situations 265
eccrine ducts open out onto the ridges) and so cre- or structureless, hemorrhages are nearly always red
ates a parallel ridge pattern. A larger hemorrhage or black and can therefore be easily differentiated
overflows from the ridges and creates a structureless from (brown) melanocytic proliferations on the basis
pattern. Red clods at the lesion’s periphery (“satellite of their color. However, rarely hemorrhage may be
clods”) are characteristic of hemorrhage, but are not brown – which makes it more difficult to distinguish
always seen (13). Whether the pattern is lines, clods from melanocytic lesions. The sharp demarcation from
Special situations 267
Figure 8.18: Acral melanoma versus periungual pigmented Bowen’s disease (intraepidermal carcinoma).
Top: Acral melanoma in situ on an index finger with the parallel ridge pattern (between 12 and 3 o’clock) on dermatoscopy (top right).
Bottom: Periungual pigmented Bowen’s disease (intraepidermal carcinoma) with the structureless pattern on dermatoscopy (bottom right).
the surrounding skin and the typical satellite clods of by human papilloma virus infection. The most common
a hemorrhage usually point to the correct diagnosis. pattern of periungual pigmented Bowen’s disease is
If doubt still remains, one may try to carefully shave (as at other sites) structureless brown. Less commonly,
off the blood with a scalpel. With bleeding that is dots arranged in lines are also visible. The parallel
mainly confined to the stratum corneum this will be ridge pattern is usually absent in Bowen’s disease.
easy. With melanocytic lesions, of course, it will be The typical vascular pattern of Bowen’s disease is also
impossible (13). usually absent. Despite some distinguishing features,
In addition to acral hemorrhage, exogenous pigmen- it can be very challenging to differentiate periungual
tation may also have a parallel ridge pattern. These pigmented Bowen’s disease from acral melanoma
lesions may also be difficult to distinguish from mela- (8.18).
nocytic proliferations on the basis of their morphology
alone, as in the case shown in figure 8.17. The correct
diagnosis of exogenous pigmentation can usually be
established by taking a careful history.
A rare differential diagnosis of acral melanoma is
acral (usually periungual) pigmented Bowen’s disease
(intraepidermal carcinoma), which is usually induced
268 Special situations
Figure 8.19: Most invasive facial melanomas are chaotic and have clues to malignancy.
Top: Invasive facial melanoma with regression. Dermatoscopy (right) shows chaos and gray structures. Middle: Invasive facial melanoma.
Dermatoscopy shows chaos and grey structures and angulated lines (polygons). Bottom: Invasive facial melanoma: Dermatoscopy shows
chaos and a pattern of clods with black clods at the periphery.
Special situations 269
8.3 The face Although any pattern may be seen, the predominant
Dermatoscopic diagnosis of pigmented lesions on dermatoscopic patterns for flat facial lesions are
the face follows the same principles as those used to circles, dots, structureless, angulated lines, curved
diagnose lesions at other locations. For many lesions, lines, and reticular lines (8.23). The anatomy of the
the face is not a “special situation” at all. Advanced facial skin explains the relatively high frequency of
neoplasms of all types show the same features on the circles and the structureless pattern in flat facial lesions
face as elsewhere. Advanced melanomas on facial (15). Hair follicles are both more frequent and more
skin are usually chaotic and have the same clues to prominent on the face than at other sites.
malignancy seen at other locations (8.19). The diagno- Furthermore, in the course of one’s life the rete ridg-
sis of facial basal cell carcinomas is not different than es tend to flatten. If the rete ridges are flattened,
at any other anatomic site, even when they are in an hyperpigmentation of basal keratinocytes is seen
embryonic stage (8.20). Other lesions that commonly not as reticular lines, but as a structureless brown
occur on the face (8.21) are Miescher nevi (a type of area interrupted by round non-pigmented follicular
largely dermal, probably congenital nevus), blue nevi, openings (8.23C). We do not recommend the term
and Spitz nevi (typically the non-pigmented type) in “pseudo-network” as it is important to recognize this
younger persons, and of course seborrheic keratosis pattern as structureless.
(8.22) in elderly persons (14). On the face one may Many solar lentigines have a structureless pattern
also find proliferations with differentiation of seba- and are usually only brown. In some solar lentigines,
ceous glands (e.g. sebaceous gland hyperplasia) or however, the rete ridges may be elongated which
benign neoplasms with follicular differentiation (e.g. results in a pattern of reticular or curved lines. Com-
trichoepithelioma), and of course inflammatory con- mon patterns of solar lentigines are shown in figure
ditions like rosacea or lupus erythematosus but these 8.24. Contrary to common belief, a reticular pattern
lesions are usually not pigmented. The dermatoscopy on chronic sun-damaged facial skin is therefore a clue
of inflammatory diseases is discussed in more detail for a non-melanocytic and not for a melanocytic lesion.
later in this chapter. A pattern of circles in flat pigmented facial lesions
Facial location assumes critical importance in the requires careful assessment, especially if the circles are
differential diagnosis of flat pigmented neoplasms; gray. A pattern of thin gray circles is the most specific
flat nevi, early melanoma, solar lentigo especially clue to early facial melanoma, but only when the
when in regression (lichen planus-like keratosis), and pigmentation is confluent, and not grey dots arranged
pigmented actinic keratosis/Bowen’s disease. The as circles (8.25, 8.26). As a facial melanoma in situ
remainder of this section is devoted to the challenging progresses, circles may thicken and in some areas
task of differentiating these lesions. coalesce to create a structureless zone, (8.27) but
270 Special situations
A B
C D
E F
Figure 8.26: Three facial melanomas in situ with circles as the main pattern.
Dermatoscopy on the right. Top, middle: In situ melanomas with gray and brown circles. Bottom: An in situ melanoma with brown circles
only.
274 Special situations
the pattern of circles usually remains visible in other highly specific for melanoma; they are also found in
parts. Sometimes the circles in melanoma in situ are pigmented actinic keratoses and lichen planus-like
brown and not gray (8.26 bottom) and then it can keratosis. Facial angulated lines probably correspond
be very difficult to differentiate melanoma in situ from to what has been previously described as rhomboids
solar lentigo. In solar lentigo, however, other clues are by Stolz et al. (15)
usually present such as reticular and curved lines and Although gray pigmentation may appear in benign
a sharply demarcated, scalloped border. Although facial lesions (as in lichen planus-like keratosis) it is
some say incomplete circles (“asymmetric follicular an important clue to malignancy in flat lesions (8.31).
openings”) are a clue to melanoma in situ, we con- In a study by Tschandl et al (16) the sensitivity of gray
sider this feature of limited value because incomplete pigmentation for melanoma in situ was 96 % and the
circles are commonly found in solar lentigines and in specificity was 31 %.
pigmented actinic keratoses (i.e. the clue has poor Because of the rather poor specificity, the decision to
specificity). A circle in a circle (concentric circles) is biopsy a lesion should not be based on the clue of
a quite specific clue to facial melanoma in situ, but gray pigmentation alone. Other clues including the
it is rarely present (i.e. the clue has poor sensitivity) presence or absence of benign clues (for example
(8.28). Furthermore, this clue is usually found in larger clues to solar lentigo) should be also considered.
lesions with other (less subtle) clues which make the However, if gray circles are present, especially if
diagnosis of melanoma obvious. In a study by Tschandl the circles are thin, the diagnosis of melanoma in
et al (16) the clue of “circle in a circle” was found in situ (lentigo maligna) should be favored over the
only 4.2 % of facial flat melanomas. other diagnoses.
Similar to flat melanomas on chronic sun damaged Pigmented actinic keratoses are common on facial
non-facial skin, the pattern of angulated lines is also skin and can mimic melanoma in situ (8.32). Clues
common in facial melanoma in situ (8.29, 8.30). to pigmented actinic keratosis are scale, background
(16) However, on the face, angulated lines are not erythema, white circles, and, although less specific,
Special situations 275
Figure 8.28: Two facial melanomas with the clue of “circle within a circle”.
Dermatoscopy on the right. Melanoma with concentric circles (circle within a circle) on dermatoscopy (black arrows).
four dots in a square. Pigmented actinic keratosis may in the part of the lesion in which regression occurs
show many patterns including circles, structureless, (8.33 top).
dots and even reticular and curved lines. Reticular While Clark nevi may mimic melanoma in situ, they
and curved lines in a pigmented actinic keratosis are very rare on facial skin. Clinicians should be very
usually indicate a collision with solar lentigo. Lichen reluctant to accept a histopathologic diagnosis of Clark
planus-like keratosis may also mimic facial melanoma nevus (often referred to as “dysplastic” nevus) on facial
in situ. Lichen planus-like keratosis is a solar lentigo skin, especially if the diagnosis is based on a partial
in the stage of regression (8.33). As a consequence biopsy. Dermatopathologists may err too.
of inflammation, melanophages accumulate in the Collisions are very common on the face. Melanoma
papillary dermis and appear as grey dots under the in situ, solar lentigo and actinic keratosis are all com-
dermatoscope. Hence these lesions are frequently mon lesions on chronic sun damaged skin, meaning
biopsied. The gray dots in lichen-planus like keratosis collisions will occur by chance.
are usually distributed between the follicular openings Histopathologically, solar lentigines are commonly
(8.33 bottom) and not arranged in circles around found adjacent to facial melanomas in situ, but this
follicular openings. does not prove a causal relationship between solar
Other parts of the lesion may show remnants of a lentigo and melanoma in situ, as has been proposed
solar lentigo. Angulated lines may occur in lichen in the concept of “unstable lentigo”. Solar lentigines
planus like-keratosis but only rarely and usually only are also found in conjunction with actinic keratoses,
276 Special situations
Figure 8.30: Melanoma in situ with angulated lines as the only clue.
Melanoma in situ with angulated lines as the main pattern. The color is brown. Note that there are no clues that point to solar lentigo (such
as a sharply demarcated scalloped border). In more difficult lesions, the absence of benign clues may raise suspicion that one is dealing
with a melanoma in situ.
especially in pigmented actinic keratosis, without being clinical recurrence seem to be young age and loca-
precursor lesions. tion on the trunk. Usually nevi recur after a superficial
Even with dermatoscopy it is sometimes difficult or even shave biopsy of a small congenital nevus or a dermal
impossible to distinguish between facial melanoma in nevus (Miescher nevus or Unna nevus), with the shave
situ, lichen-planus like keratosis and pigmented actinic removing only the epidermal and superficial dermal
keratosis. If dermatoscopy does not allow a specific parts but not the deeper dermal parts of the nevus. If
diagnosis with confidence, biopsy is required. In the these residual melanocytes repopulate the epidermis,
specific context of flat facial pigmented lesions with this produces visible pigmentation inside the scar.
equivocal dermatoscopy findings, consideration may Recurrent nevi may have any pattern but the most com-
be given to using reflectance confocal microscopy as an mon patterns are radial lines and pseudopods (8.35,
adjunctive diagnostic procedure to increase specificity top and middle). When radial lines and/or pseudopods
(reduce the number of biopsies required). (17, 18) are found over the entire circumference, a recurrent
nevus mimics a Reed nevus. More worryingly, radial
lines and/or pseudopods may be seen only in some
8.4 Mucosal lesions segments of the periphery, creating a recurrent nevus
Lentigines (melanotic macules) are the most common which mimics melanoma dermatoscopically.
pigmented lesions on the lip, the oral mucosa, and There are clues to help distinguish recurrent nevus from
the genital area. Lentigines are typically light-brown melanoma. Most important is the relationship of the
or dark-brown spots (8.34 top row). On the lip they pigmentation to the scar of the original procedure. In
often occur as single lesions, usually on the lower lip. recurrent nevi the pigmentation is nearly always confined
“Mucosal lentiginosis” refers to the presence of multiple to the scar. If a melanoma recurs the pigmentation is
lesions. Genital lentiginosis is much more diverse in also present outside the scar (8.35 bottom). While a
terms of morphology than labial or oral lentiginosis recurrent nevus with a pattern of radial lines and/or
(8.34 middle row). pseudopods may mimic a melanoma, these patterns
As a reliable distinction between this entity and a are usually not seen in an actual recurrent melanoma.
mucosal melanoma cannot be made on the basis of In contrast to recurrent nevi, which are more common
clinical investigation alone, these lesions are frequently on the trunk, recurrent melanomas more frequently
biopsied. The most common dermatoscopic patterns of occur on the face and on acral skin, simply because
mucosal pigmented lesions are structureless, curved melanomas are more often removed incompletely at
lines, and circles, but any pattern may be seen. Color these sites. Not only is obtaining surgical margins more
may be more important than structure in diagnosing difficult, facial and acral melanomas commonly have
mucosal lesions. A recent study suggests any blue, a lentiginous growth pattern (single melanocytes in
gray or white area should raise the suspicion for a the epidermis predominate over nests), making it more
malignant neoplasm especially when combined with difficult for both clinician and pathologist to distinguish
structureless areas (19). the margins of the melanoma.
Mucosal melanomas are usually diagnosed late when Fortunately, a lentiginous growth pattern is correlated
the clinical diagnosis is already obvious (8.34 bottom with slow growth rate, so if these melanomas recur,
row). Dermatoscopy is of limited additional value at this they recur slowly. This explains the apparent paradox
stage, but may be useful in less advanced cases (20, 21). that nevi usually recur faster than melanomas. In a
study by Blum et al (22) the median time to treatment
of recurrence was 8 months for nevi and 25 months
8.5 Recurrent melanocytic lesions for melanoma.
Nevi and melanomas often recur after incomplete
removal. Recurrent nevi may mimic melanoma derma-
toscopically and pathologically, creating a diagnostic 8.6 Difficult lesions
challenge for both the clinician and the dermatopathol- What is difficult? The answer to this question depends
ogist. When submitting recurrent pigmented lesions, on the investigator. Beginners find a lot of things difficult
the clinician must alert the pathologist to the previous which experts find easy, but that is not the focus of this
surgical procedure and (when available) provide the chapter. There are, however, some types of lesions
original histopathologic diagnosis. that challenge even a well-trained investigator (8.36).
Not every incompletely removed nevus becomes a These lesions may be equally consistent with several
clinically apparent recurrent nevus. Risk factors for differential diagnoses, because they demonstrate insuf-
Special situations 281
ficient or contradictory clues. In the extreme case there In theory, any type of collision may occur. However,
are lesions with misleading clues, which are difficult to some collisions occur more frequently than others. For
reconcile with the diagnosis even retrospectively, i.e. instance, solar lentigines are very common on chronic
after the histopathology is known. UV-damaged skin, meaning that these solar lentigines
are seen to collide with all manner of other lesions simply
Too many differential diagnoses by chance. In our previous example, one would observe
Some patterns give rise to a long list of differential diag- the clues of basal cell carcinoma as well as those of
noses that cannot be clearly and reliably distinguished solar lentigo (reticular lines!); the different parts can
from each other, even with the presence of additional usually be quite easily distinguished from each other.
clues. For example, when one follows the algorithm Whether one is confronted with a simple coincidence or
for the structureless pattern, one is confronted with a a causal link, i.e. one lesion (usually a malignant one)
large number of potential diagnoses. A large number arising from another lesion – is usually mere speculation.
of differential diagnoses are also generated when the The presence of a coincidence is especially likely when
investigator is unable to make a confident decision at the two lesions are fundamentally different. For instance,
a certain point in the algorithm. For instance, when the it would be difficult to explain how a melanocytic lesion
investigator cannot decide whether two patterns are (a melanoma) could arise from an epithelial lesion (a
arranged symmetrically or asymmetrically, all diag- solar lentigo).
noses at the ends of both symmetric and asymmetric Collisions with solar lentigines are common, regardless
branches of the algorithm must be considered. This is of whether the second lesion is a basal cell carcinoma,
a major reason why beginners perform more excisions a melanoma, or a superficial squamous cell carcinoma.
– they distrust their observations more frequently than These combinations should not be a reason to construe
experts do, and the list of the potential diagnoses is a causal association. Collisions between seborrheic
therefore longer. The difficulty of too many differential keratoses and superficial squamous cell carcinomas of
diagnoses is unpleasant but rarely causes any serious the Bowen’s type and collisions of dermatofibromas and
error because diagnostic uncertainty usually results in basal cell carcinomas are less common, but do occur. Of
a histopathological investigation. course one should also interpret melanomas that arise
in a pre-existing nevus as collision lesions. Last but not
No specific clues least, it should be mentioned that pathologists tend to
Non-pigmented lesions are considered difficult because mention in their reports only those diagnoses that they
usually the pattern of vessels is the only clue to diagnosis consider relevant. For instance, when there is a collision
(23). Without pigment, a lesion lacks not only color but between a basal cell carcinoma and a solar lentigo,
also structure (= pattern). The specificity of the patterns the latter is not always mentioned. In these cases the
of vessels is very poor compared to that of pigmented dermatoscopist’s findings may remain inexplicable.
structures, meaning there is greater diagnostic uncer-
tainty for non-pigmented lesions (24). The consequence Misleading clues
is a more frequent need for histopathology. Clues that suggest an incorrect diagnosis are misleading.
They commonly give rise to serious errors. Dermatos-
Contradictory clues copy is not a perfect method, even in the hands of an
We call the clues seen in a lesion “contradictory” when experienced investigator. Sometimes the histological
some of the clues favor one particular diagnosis, but diagnosis does not concur at all with the dermatoscopic
other clues favor another diagnosis. For instance, a lesion appearance.
which has some clues such as blue clods, radial lines Nevertheless, one’s confidence in the method should
and serpentine vessels indicating a basal cell carcino- not be shaken by such instances. On the contrary, when
ma, but other clues like reticular lines which render this dermatoscopic findings cannot be made to concur with
diagnosis unlikely, one is confronted with a conflict of the histopathological diagnosis, one should speak to
clues. Handling such apparent contradictions is made one’s pathologist and not to one’s psychiatrist! Never
easier when the potential causes of contradictory clues forget that even pathology reports may be incorrect.
are kept in mind. Errors are very rare in specimen handling and report
The commonest cause of contradictory clues is misin- transcription, but they do occur. More commonly, a
terpretation on the part of the investigator. Once this different diagnosis is reached after the pathologist
possibility has been ruled out, the investigator should reviews the specimen or seeks a second opinion from
consider the presence of a collision lesion (25–28). a colleague.
284 Special situations
Non-infectious inflammatory diseases Scale in psoriasis is usually white (8.40). The scale of
The dermatoscopic appearance of psoriasis has already spongiotic dermatitis (e.g. nummular dermatitis, sebor-
been discussed in detail in chapter 6. Psoriasis is marked rheic dermatitis, contact dermatitis) is more frequently
by a monomorphous pattern of vessels as either dots or yellow because the stratum corneum contains serum.
small coiled vessels. A further clue is white scale (30). The scale of pityriasis rosea is found peripherally.
These features mean it is often necessary to differentiate In lesions of chronic cutaneous lupus erythematosus
psoriasis from Bowen’s disease. In Bowen’s disease, one finds, in addition to erythema, small, white, yellow
vessels are usually coiled. The distribution of vessels is or orange clods on dermatoscopy. These correspond
usually random in psoriasis, whereas in Bowen’s dis- to the follicular keratin plugs seen on histology (8.41).
ease they tend to be arranged in clusters (31). These In addition, scale may occur. In older lesions of lupus
characteristics, however, are relatively non-specific and erythematosus one finds white circles around follicular
may not permit differentiation from Bowen’s disease openings and, as a consequence of sclerosis, white
based on dermatoscopy alone. structureless zones (8.42).
288 Special situations
Porokeratosis, a disorder of keratinization with various seen even in very small lesions (8.45). In sarcoidosis
phenotypes, is typified by the so-called cornoid lamella (including the sarcoidal variant of rosacea) one finds
– a column of parakeratotic cells in the stratum corne- a yellow or orange structureless zone and serpentine
um, visible only on histopathology. On dermatoscopy vessels (8.46). Cutaneous amyloidosis is characterized
this appears as a white or yellow hyperkeratotic circle by a red structureless zones on dermatoscopy (8.47).
defining the border of the lesion (8.43). Centrally there Vessels are typically absent. Transient acantholytic
may be a monomorphous pattern of vessels as dots, dyskeratosis, also known as Grover’s disease, is char-
or a brown and white structureless area as a sign of acterized by a central yellow clod that is often star-
atrophy (32). shaped, and most likely corresponds to the superficial
In lichen planus one finds structureless red to brown erosions topped with a serum crust that are typical for
areas intersected by thick white branched lines or white this itchy disease (8.48).
clods (8.44). The white branched lines correspond to the
so-called Wickham striae (30, 33). The dermatoscopic
features of lichen planus are quite specific and can be
290 Special situations
Infections and Infestations are almost never pigmented. Verrucae planae (plane
warts) have vessels as dots on a light-brown or yellow
Viruses structureless background. Verrucae vulgares or common
Of the large number of skin lesions that may be caused verrucae (8.49, bottom) have skin-colored or pink clods
by viruses we will confine ourselves here to a small with vessels in the center (centered vessels). When a
selection of those lesions whose diagnosis is facilitated viral wart occurs at the sole of the foot or the palm it is
by dermatoscopy. Molluscum contagiosum is caused by known as a plantar or palmar verruca. These merely
a poxvirus. These lesions are manifested as skin-colored have a yellow structureless zone that corresponds to
papules that reveal, on dermatoscopy, white or yellow the hyperkeratosis with red or black dots and short
clods and/or a structureless area in the center (8.49, lines, corresponding to hemorrhage and vessels, and
top). At the periphery of the lesion one finds radially can be differentiated from a callus or a clavus (36).
arranged vessels that do not cross the center of the
lesion (29, 34, 35). Various human papillomaviruses Fungi
(HPV) are the cause of viral warts (verrucae). Viral Tinea nigra is a rare disease in Europe, North America
warts have many different clinical appearances, but and southern Australia, but more common in tropical
Special situations 291
areas. It is caused by the fungus Hortaea werneckii. the axillary hair (29). In most other bacterial infections
Clinically this infection is easily confused with an acral one only finds unspecific hyperemia or pus.
nevus or a melanoma because it is seen as a more or
less homogeneous, dirty brown or gray macule (8.50). Parasites
On dermatoscopy the differential diagnosis can be Scabies is a very common disease caused by the mite
easily resolved because tinea nigra has no parallel Sarcoptes scabiei. The mite lives in burrows in the
lines but very thin brown or gray branched lines over stratum corneum and causes skin lesions and a some-
the entire lesion (37). times intractable itch. To make the diagnosis without
dermatoscopy the mite, which is barely visible with the
Bacteria unaided eye, has to be located microscopically in skin
Trichomycosis palmellina is probably the only bacterial scrapings. However, the mite is easily demonstrated on
infection whose diagnosis is facilitated even slightly dermatoscopy with no need for skin scrapings (8.51).
by dermatoscopy. This disease is caused by Coryne- On dermatoscopy one sees a curved or serpentine
bacterium tenuis and is seen on dermatoscopy as a line, which corresponds to the burrow (8.52). At the
yellow, black or red structureless area that surrounds end of the line the mite is seen as a small, dark triangle
292 Special situations
(anterior portion of the parasite). This manifestation has A rather rare disease in high latitudes is caused by the
been compared to a jet with a contrail. Occasionally sand flea Tunga penetrans. The main breeding areas
mite excrement is seen as black dots within the thick are South-Central America, Africa or Asia. The sand
curved line (38). flea bores its way through the epidermis, where it lays
Lice (pediculosis capitis/phthiriasis pubis) can also be its eggs. The disease is manifested as dark papules that
diagnosed in vivo with the dermatoscope. Nits may be usually occur on acral locations. On dermatoscopy one
distinguished from other structures such as dandruff. finds a round, black or red structureless area with a
When the nit contains a nymph one sees brown oval central black clod or a dot (8.53). This feature correlates
clods on dermatoscopy. If they are empty (after the with the parasite’s rump protruding out of the skin (40).
nymphs have hatched) there will be transparent clods
with a flat end. These should be distinguished from
debris of hair gel or dandruff, which will be seen as
polygonal clods with white dots and lines (39).
294 Special situations
References
1 Braun RP, Baran R, Le Gal FA, Dalle S, Ronger S, Pan- 15 Stolz W, Schiffner R, Burgdorf WH. Dermatoscopy for
dolfi R, et al. Diagnosis and management of nail pig- facial pigmented skin lesions. Clin Dermatol. 2002; 20;
mentations. J Am Acad Dermatol. 2007; 56; 835–847. 276–278.
2 Thomas L, Dalle S. Dermoscopy provides useful infor- 16 Tschandl P, Rosendahl C, Kittler H. Dermatoscopy of flat
mation for the management of melanonychia striata. pigmented facial lesions. J Eur Acad Dermatol Venere-
Dermatol Ther. 2007; 20; 3–10. ol. 2015; 29; 120–127.
3 Jellinek N. Nail matrix biopsy of longitudinal melanon- 17 de Carvalho N, Farnetani F, Ciardo S, Ruini C, Witkow-
ychia: diagnostic algorithm including the matrix shave ski AM, Longo C, et al. Reflectance confocal microsco-
biopsy. J Am Acad Dermatol. 2007; 56; 803–810. py correlates of dermoscopic patterns of facial lesions
4 Ronger S, Touzet S, Ligeron C, Balme B, Viallard AM, help to discriminate lentigo maligna from pigmented
Barrut D, et al. Dermoscopic examination of nail pig- nonmelanocytic macules. Br J Dermatol. 2015; 173;
mentation. Arch Dermatol. 2002; 138; 1327–1333. 128–133.
5 Iorizzo M, Tosti A, Di Chiacchio N, Hirata SH, Misciali 18 Guitera P, Pellacani G, Crotty KA, Scolyer RA, Li LX,
C, Michalany N, et al. Nail melanoma in children: dif- Bassoli S, et al. The impact of in vivo reflectance con-
ferential diagnosis and management. Dermatol Surg. focal microscopy on the diagnostic accuracy of lentigo
2008; 34; 974–978. maligna and equivocal pigmented and nonpigmented
6 Braun RP, Baran R, Saurat JH, Thomas L. Surgical Pearl: macules of the face. J Invest Dermatol. 2010; 130;
Dermoscopy of the free edge of the nail to determine 2080–2091.
the level of nail plate pigmentation and the location of 19 Blum A, Simionescu O, Argenziano G, Braun R, Cabo
its probable origin in the proximal or distal nail matrix. H, Eichhorn A, et al. Dermoscopy of pigmented lesions
J Am Acad Dermatol. 2006; 55; 512–513. of the mucosa and the mucocutaneous junction: results
7 Saida T, Koga H. Dermoscopic patterns of acral me- of a multicenter study by the International Dermoscopy
lanocytic nevi: their variations, changes, and signifi- Society (IDS). Arch Dermatol. 2011; 147; 1181–1187.
cance. Arch Dermatol. 2007; 143; 1423–1426. 20 Ferrari A, Zalaudek I, Argenziano G, Buccini P, De Sim-
8 Miyazaki A, Saida T, Koga H, Oguchi S, Suzuki T, one P, Silipo V, et al. Dermoscopy of pigmented lesions
Tsuchida T. Anatomical and histopathological correlates of the vulva: a retrospective morphological study. Der-
of the dermoscopic patterns seen in melanocytic nevi matology. 2011; 222; 157–166.
on the sole: a retrospective study. J Am Acad Dermatol. 21 Lin J, Koga H, Takata M, Saida T. Dermoscopy of pig-
2005; 53; 230–236. mented lesions on mucocutaneous junction and mucous
9 Malvehy J, Puig S. Dermoscopic patterns of benign vo- membrane. Br J Dermatol. 2009; 161; 1255–1261.
lar melanocytic lesions in patients with atypical mole 22 Blum A, Hofmann-Wellenhof R, Marghoob AA, Argen-
syndrome. Arch Dermatol. 2004; 140; 538–544. ziano G, Cabo H, Carrera C, et al. Recurrent melano-
10 Braun RP, Thomas L, Kolm I, French LE, Marghoob AA. cytic nevi and melanomas in dermoscopy: results of a
The furrow ink test: a clue for the dermoscopic diagno- multicenter study of the International Dermoscopy Soci-
sis of acral melanoma vs nevus. Arch Dermatol. 2008; ety. JAMA Dermatol. 2014; 150; 138–145.
144; 1618–1620. 23 Menzies SW, Kreusch J, Byth K, Pizzichetta MA, Mar-
11 Saida T, Miyazaki A, Oguchi S, Ishihara Y, Yamazaki ghoob A, Braun R, et al. Dermoscopic evaluation of
Y, Murase S, et al. Significance of dermoscopic patterns amelanotic and hypomelanotic melanoma. Arch Der-
in detecting malignant melanoma on acral volar skin: matol. 2008; 144; 1120–1127.
results of a multicenter study in Japan. Arch Dermatol. 24 Kittler H, Riedl E, Rosendahl C, Cameron A. Dermatos-
2004; 140; 1233–1238. copy of unpigmented lesions of the skin: a new classifi-
12 Ishihara Y, Saida T, Miyazaki A, Koga H, Taniguchi A, cation of vessel morphology based on pattern analysis.
Tsuchida T, et al. Early acral melanoma in situ: correla- Dermatopathol: Pract & Conc. 2008; 14.
tion between the parallel ridge pattern on dermoscopy 25 Ferrara G, Zalaudek I, Cabo H, Soyer HP, Argenziano
and microscopic features. Am J Dermatopathol. 2006; G. Collision of basal cell carcinoma with seborrhoeic
28; 21–27. keratosis: a dermoscopic aid to histopathology? Clin
13 Zalaudek I, Argenziano G, Soyer HP, Saurat JH, Braun Exp Dermatol. 2005; 30; 586–587.
RP. Dermoscopy of subcorneal hematoma. Dermatol 26 Zaballos P, Llambrich A, Puig S, Malvehy J. Dermosco-
Surg. 2004; 30; 1229–1232. py is useful for the recognition of benign-malignant com-
14 Sahin MT, Ozturkcan S, Ermertcan AT, Gunes AT. A pound tumours. Br J Dermatol. 2005; 153; 653–656.
comparison of dermoscopic features among lentigo 27 de Giorgi V, Massi D, Sestini S, Alfaioli B, Carelli G,
senilis/initial seborrheic keratosis, seborrheic keratosis, Carli P. Cutaneous collision tumour (melanocytic nae-
lentigo maligna and lentigo maligna melanoma on the vus, basal cell carcinoma, seborrhoeic keratosis): a
face. J Dermatol. 2004; 31; 884–889. clinical, dermoscopic and pathological case report. Br
J Dermatol. 2005; 152; 787–790.
Special situations 295
297
Figure 9.3: The morphology of a nevus is not predictive of the risk that a melanoma will develop in this nevus.
The clinical image on the left shows a melanoma that developed in a pre-existing nevus. On the right one can see digital dermatoscopic
images of four nevi of the same patient. Can you predict by morphology which nevus is the precursor lesion of the melanoma on the left?
Answer see Figure 9.4.
Figure 9.4: The morphology of a nevus is not predictive of the risk that a melanoma will develop in this nevus.
Sequence of images which demonstrates the development of the melanoma in one of the four nevi shown in Figure 9.3. By morphology it
cannot be predicted which nevus will be the precursor lesion. The concept of the “dysplastic” or “atypical” nevus is flawed fundamentally
because the morphology of a nevus is not predictive of its biological fate.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
examination of lesions without clear clues to melanoma. Short term monitoring is more efficient – one needs to
A second type of monitoring may be used for high risk monitor fewer lesions to detect one melanoma. Short
patients with multiple nevi (9.5). Lesions to be monitored term monitoring, however, will only increase the spec-
are selected randomly, with as many lesions as possible ificity by reducing the number of excisions. It sorts
documented at the initial examination and re-imaged out false positives but does not find additional false
at each consultation (4). There are no criteria for the negatives. To find additional false negatives one must
selection of lesions because one wishes to identify include inconspicuous lesions (long term monitoring).
melanomas that demonstrate no clues – at least on Because inconspicuous lesions are monitored one needs
initial inspection. One therefore relies on chance and to monitor a lot of lesions to detect one melanoma,
increases the likelihood of identifying a melanoma by more than 1000 in some settings. Long term monitor-
documenting as many lesions as possible and confin- ing, however, will increase specificity and sensitivity
ing the investigation to high-risk patients. The intervals because it will detect melanomas that do not have any
between examinations are longer – typically six to clues and cannot be diagnosed otherwise.
twelve months. This type of monitoring is an alterna- A third possible approach is to identify new or changed
tive to indiscriminate excision of nevi in patients with pigmented lesions by comparison of serial clinical total
multiple nevi. body photographs, and then subjecting these lesions
A meta-analysis by Salerni et al. (5) shows the different to digital monitoring. Another study by Salerni et al
outcomes of these two monitoring strategies (Table 9.1). showed this to be an effective strategy, detecting more
Table 9.1
Authors Patients (n) Lesions (n) Follow-up Melanomas In-situ Lesions Relative Malignant/ Frequency
interval detected melanomas excised (n) frequency benign ratio of melano-
(months) during fol- (%) of lesions mas among
low-up (n) excised (%) excised
lesions (%)
Kittler 2000,
202 1,862 3, 6, 12 8 62.5 % 75 4 1: 8.4 10.7
Austria (4)
Menzies 2001,
245 318 3 7 71.4 % 53 16.6 1:6.5 13.2
Australia (3)
Malvehy 2002,
290 3,170 3, 6 8 25 % 42 1.3 1: 4.2 19
Spain (14)
Schiffner 2003,
145 272 3, 6, 12 0 – 7 2.6 – –
Germany (15)
Robinson 2004,
100 3,482 12 4 100 % 193 5.5 1:47.3 2.1
USA (16)
Haenssle 2004,
212 2,939 3, 6, 12 17 52.9 % 112 3.8 1:5.5 15.2
Germany (17)
Bauer 2005,
196 2,015 3, 6, 12 2 100 % 33 1.6 1:15.5 6.1
Germany (18)
Haenssle 2006,
530 7,001 3, 6, 12 53 52.8 % 637 9.1 1:12 8.3
Germany (19)
Fuller 2007,
297 5,945 6, 12 6 33.3 % 324 5.4 1:53 1.9
USA (20)
Altamura 2008,
1,859 2,602 1.5, 3 81 67.9 % 487 18.7 1:5 16.6
Australia (21)
Argenziano
405 600 3, 6, 12 12 50 % 54 9 1: 3.4 22.2
2008, Italy (10)
Haenssle 2010,
688 11,137 3, 6, 12 87 43.6 % 1,219 10.9 1:8.5 10.4
Germany (22)
Salerni 2011,
618 11,396 3, 6, 12 98 54 % 1,152 10.1 1:10.7 8.5
Spain (6)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
melanomas than by short-term monitoring, but requiring a lesion is a melanoma, it will nearly always be slow
far fewer lesions to be monitored than random long-term growing at this stage of its growth. More than 50 % of
monitoring. In practice, a combination of strategies melanomas discovered only by the use of monitoring
seems to be most useful (6, 7). are in situ melanomas, and nearly all the invasive mela-
nomas are thinner than 1 mm, meaning that the risk of
Regardless of how one uses monitoring, knowing which metastasis is either nil (in cases of in situ melanomas) or
lesions are unsafe to monitor is essential. By definition, very low (in cases of thin invasive melanomas).
monitoring is performed on potential melanomas. Moni- Nodular lesions that cannot be diagnosed as benign
toring must therefore be confined to lesions which, should with absolute certainty must be excised at the initial
melanoma be confirmed, will still have an excellent consultation. If one monitors a nodule, one may be
prognosis after the monitoring period. In practice this monitoring a rapidly growing nodular melanoma that
means flat and small melanocytic lesions, preferably could conceivably have already reached a Breslow
with a reticular pattern on dermatoscopy. Even if such thickness of several millimeters by the review. To a
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Table 9.2
Nevus Melanoma
Change in size No or symmetrical increase in size Sometimes an asymmetrical increase in size
Change in color No or lighter/darker brown or more/less erythema Sometimes a new color
Changes in No or non-essential changes in structure (the pre-existing structure Occasionally a new structure
structure becomes more prominent/less prominent) or pre-existing basic (includes clues to melanoma)
elements (e.g. clods or dots) are more/less numerous
lesser extent this is true for larger (over 1 cm diameter) When longer intervals are used, a distinction must be
flat lesions as well. made between significant and insignificant changes.
When examining patients with multiple nevi, the question The latter are much more common. For instance, the
obviously arises as to how many lesions one should transient stimulation of melanin production provoked
document. An unequivocal answer cannot be provided by UV exposure may cause nevi to initially darken and
for this question. In clinical practice, the availability then become lighter after a few weeks.
of time and personnel resources limits the number of Distinctions are made on the basis of changes in size,
dermatoscopic images one can obtain and record. For color, and structure (9.6). Changes in size may be sym-
practical reasons it would be advisable to dispense metrical (the shape of the lesion remains the same) or
with the documentation of melanocytic lesions that asymmetrical (the shape changes). The color of a lesion
are smaller than 3 mm in diameter; otherwise one will may change in that a pre-existing color (usually brown)
have to record far too many images for each patient. may appear lighter or darker, or that a new color is
seen. With regard to structure, an existing pattern (for
instance reticular lines) may disappear, a new pattern
9.2 Interpretation of changes may appear, or basic elements of a pattern such as dots
The interpretation of changes seen in monitored lesions or clods may be present in greater numbers or reduced
depends on the type of monitoring being performed, numbers. A change in the number of brown clods is
and the monitoring interval chosen. What is an appro- relatively common and is not relevant. However, any
priate interval between examinations? If the observation change in size, any new color, and any new structure
period is too short, even changes in melanomas will should be regarded as a significant change.
be difficult to notice. Initially one tends to over-interpret changes, and in par-
For adults, reasonable monitoring intervals range ticular fail to realize that “changes” in lesions are often
between 3 and 12 months. Three months is a gen- produced by variations in the imaging process, rather
erally accepted period, over which most melanomas than changes in the lesion itself. Different exposure or
are expected to change (although this is too short an color balance is common if the digital dermatoscope
observation period for many lentiginous melanomas). is not calibrated. Stretching of the skin during imag-
Over longer observation periods, a significant number ing distorts lesions, and rotation of the dermatoscope
of benign nevi will change, and so criteria must be produces different lesion orientation. Of course, if the
established for “acceptable” and “unacceptable” change. wrong lesion is imaged at the follow-up examination,
As many nevi change even in the short term in children, “change” is inevitably produced. When patients have
and melanomas are very rare, monitoring of pigmented multiple nevi, this is not at all uncommon.
lesions in pre-pubertal children is best avoided.
Short term monitoring (three months or less) is usually
performed for lesions that, because of history or appear- 9.3 Growing nevus or melanoma?
ance, cannot be deemed benign with absolute certainty An important criticism of digital monitoring is the fact
at the initial examination. For short-term monitoring, that nevi also change. This is basically true, but nevi
even small changes are regarded as significant, and stop growing at some time whereas melanomas do not.
all lesions showing changes (even to a “more benign” Melanomas and nevi also grow in different ways —
appearance) over 3 months or less should be submitted nevi tend to grow symmetrically, whereas melanomas
for histopathology. grow asymmetrically, i. e. their shape changes (9.6,
Longer monitoring intervals – six months to one year – 9.8, Table 9.2). The rate of change for nevi varies,
are mainly used for lesions with no features to suggest mainly depending on the type of nevus and the patient’s
malignancy, typically in patients with multiple nevi. age. Spitz or Reed nevi, for instance, grow faster than
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
time period. For these slowly growing melanomas, a case of two separate sequential examinations with no
monitoring period longer than 3 months is required comparison of images. In other words, the increase in
(8–10). sensitivity at the follow-up examination outweighs the
In adults the proportion of nevi that show significant losses at the initial examination. The specificity of the
changes in the course of a year is less than 5 %. This follow-up examination is not significantly influenced by
remains true for patients with multiple nevi and patients the use of monitoring, so the overall benefit remains.
with the so-called “dysplastic nevus syndrome”. The It is this dependence on the subsequent examination
number increases with longer observation periods and that is the greatest weakness of sequential derma-
reduces with advancing age. As mentioned earlier, toscopy. When the patient does not report for the
one finds a large number of growing nevi in children follow-up examination, the loss in sensitivity at the
and adolescents. initial examination is not counterbalanced by increased
Critics of digital monitoring assert that the procedure melanoma detection at the (missed) second visit. As a
may document features which could have diagnosed a result the excision of some melanomas is delayed. In
melanoma at the initial examination. The contention has other words, the benefits of monitoring depend on the
been examined and refuted. In a study in which expe- patient’s compliance. Therefore, the use of sequential
rienced dermatologists were shown the initial pictures dermatoscopic monitoring should only be offered as an
of melanomas diagnosed by monitoring, but were not alternative to biopsy, after patients are appropriately
shown the follow-up sequences, the best dermatologists informed about the procedure, and preferably provide
achieved a sensitivity of just 27 % (11). Lacking the addi- written consent.
tional information about the change, it was impossible To summarize, the successful application of digital mon-
to identify these melanomas. Without explicitly saying itoring requires correct selection of lesions, an informed
so, these critics are arguing that melanomas should be and motivated patient, and a fail-safe patient recall
allowed to grow until they demonstrate conventional system. Provided these principles are strictly adhered to,
clinical or dermatoscopic clues. monitoring is a safe and useful method of investigation
Although the majority of melanomas occur de novo, that both improves the early detection of melanomas
a melanoma may develop in a pre-existing nevus. A and reduces the number of excised nevi (13).
melanoma may develop in any nevus, but most often
they arise in “superficial and deep” congenital nevi or
in Clark nevi. However, this is very rarely the case from
a prospective point of view, making it unlikely that one
will document a melanoma developing in a pre-existing
nevus – unless one observes tens of thousands of nevi
per year. Figure 9.11 shows a melanoma that developed
in a pre-existing Clark nevus.
References
1 Skvara H, Teban L, Fiebiger M, Binder M, Kittler H. Lim- 14 Malvehy J, Puig S. Follow-up of melanocytic skin lesions
itations of dermoscopy in the recognition of melanoma. with digital total-body photography and digital dermos-
Arch Dermatol. 2005; 141(2): 155–160. copy: a two-step method. Clin Dermatol. 2002; 20(3):
2 Kittler H, Guitera P, Riedl E, et al. Identification of clin- 297–304.
ically featureless incipient melanoma using sequential 15 Schiffner R, Schiffner-Rohe J, Landthaler M, Stolz W.
dermoscopy imaging. Arch Dermatol. 2006; 142(9): Long-term dermoscopic follow-up of melanocytic naevi:
1113–1119. clinical outcome and patient compliance. Br J Dermatol.
3 Menzies SW, Gutenev A, Avramidis M, Batrac A, McCa- 2003; 149(1): 79–86.
rthy WH. Short-term digital surface microscopic monitor- 16 Robinson JK, Nickoloff BJ. Digital epiluminescence micros-
ing of atypical or changing melanocytic lesions. Arch copy monitoring of high-risk patients. Arch Dermatol.
Dermatol. 2001; 137(12): 1583–1589. 2004; 140(1): 49–56.
4 Kittler H, Pehamberger H, Wolff K, Binder M. Follow-up 17 Haenssle HA, Vente C, Bertsch HP, et al. Results of
of melanocytic skin lesions with digital epiluminescence a surveillance programme for patients at high risk of
microscopy: patterns of modifications observed in early malignant melanoma using digital and conventional
melanoma, atypical nevi, and common nevi. J Am Acad dermoscopy. Eur J Cancer Prev. 2004; 13(2): 133–138.
Dermatol. 2000; 43(3): 467–476. 18 Bauer J, Blum A, Strohhacker U, Garbe C. Surveillance of
5 Salerni G, Teran T, Puig S, et al. Meta-analysis of digital patients at high risk for cutaneous malignant melanoma
dermoscopy follow-up of melanocytic skin lesions: a study using digital dermoscopy. Br J Dermatol. 2005; 152(1):
on behalf of the International Dermoscopy Society. J Eur 87–92.
Acad Dermatol Venereol. 2013; 27(7): 805–814. 19 Haenssle HA, Krueger U, Vente C, et al. Results from an
6 Salerni G, Carrera C, Lovatto L, et al. Benefits of total observational trial: digital epiluminescence microscopy
body photography and digital dermatoscopy (“two-step follow-up of atypical nevi increases the sensitivity and
method of digital follow-up”) in the early diagnosis of the chance of success of conventional dermoscopy in
melanoma in patients at high risk for melanoma. J Am detecting melanoma. J Invest Dermatol. 2006; 126(5):
Acad Dermatol. 2012; 67(1): e17–27. 980–985.
7 Salerni G, Carrera C, Lovatto L, et al. Characterization 20 Fuller SR, Bowen GM, Tanner B, Florell SR, Grossman
of 1152 lesions excised over 10 years using total-body D. Digital dermoscopic monitoring of atypical nevi in
photography and digital dermatoscopy in the surveil- patients at risk for melanoma. Dermatol Surg. 2007;
lance of patients at high risk for melanoma. J Am Acad 33(10): 1198–1206; discussion 1205–1196.
Dermatol. 2012; 67(5): 836–845. 21 Altamura D, Avramidis M, Menzies SW. Assessment
8 Terushkin V, Dusza SW, Scope A, et al. Changes observed of the optimal interval for and sensitivity of short-term
in slow-growing melanomas during long-term dermoscopic sequential digital dermoscopy monitoring for the diagnosis
monitoring. Br J Dermatol. 2012; 166(6): 1213–1220. of melanoma. Arch Dermatol. 2008; 144(4): 502–506.
9 Argenziano G, Kittler H, Ferrara G, et al. Slow-growing 22 Haenssle HA, Korpas B, Hansen-Hagge C, et al. Selec-
melanoma: a dermoscopy follow-up study. Br J Dermatol. tion of patients for long-term surveillance with digital
2010; 162(2): 267–273. dermoscopy by assessment of melanoma risk factors.
10 Argenziano G, Mordente I, Ferrara G, Sgambato A, Arch Dermatol. 2010; 146(3): 257–264.
Annese P, Zalaudek I. Dermoscopic monitoring of melano-
cytic skin lesions: clinical outcome and patient compliance
vary according to follow-up protocols. Br J Dermatol.
2008; 159(2): 331–336.
11 Kittler H, Binder M. Follow-up of melanocytic skin lesions
with digital dermoscopy: risks and benefits. Arch Der-
matol. 2002; 138(10): 1379.
12 Kittler H. Use of digital dermoscopy to monitor melano-
cytic lesions: risks and benefits. J Drugs Dermatol. 2003;
2(3): 309–311.
13 Moloney FJ, Guitera P, Coates E, et al. Detection of
primary melanoma in individuals at extreme high risk:
a prospective 5-year follow-up study. JAMA Dermatol.
2014; 150(8): 819–827.
309
10 Cases
The following cases are best used to apply and test your not always be exactly the same. In these cases you
knowledge of pattern analysis. Each right-hand page should ask yourself whether the differing descriptions
presents paired clinical and dermatoscopy images. Each have led to different diagnoses. As mentioned earlier,
left-hand page shows our description of each lesion the algorithm is redundant so that for most lesions,
using the method of pattern analysis, a dermatoscopic most plausible descriptions lead to the same diagnosis.
diagnosis or differential diagnosis derived from this Most of the lesions shown here were excised, so con-
description, and, where obtained, the histopathologic spicuous, equivocal and malignant lesions are over-rep-
diagnosis. resented. In order to restore some balance, we have
In order to maximize the learning effect, we recommend also shown some lesions that were not excised because
you do not look at our descriptions or diagnoses until dermatoscopy could confidently confirm a benign
after you have attempted to describe the dermatoscopic diagnosis. It is still important to use pattern analysis
image yourself using pattern analysis, and reached a for these lesions so that one gets a feeling for the full
reasoned diagnosis or differential diagnosis on the range of appearances of common and – if one may
basis of this description. Alternatively, you may use say so – banal lesions. A strong grounding in the
the simplified algorithm “Chaos and Clues” described appearance of common benign lesions is essential in
in chapter 5 to determine whether a lesion should becoming an expert, as variation from these benign
be excised or not. Do not forget that the assessment patterns is in itself a clue to malignancy.
of pattern and color is based on overall impression. Like all morphological methods, pattern analysis occa-
Pattern and color indicate the direction in which one sionally leads to an incorrect diagnosis. We do however
should proceed; differential diagnoses are resolved by firmly believe that adherence to the principles of pattern
the use of clues. The principle of pattern analysis is: analysis will lead more often to the correct diagnosis
than any other method (including no method!). We have
Pattern + Color + Clues = Diagnosis provided some examples of such incorrect diagnoses,
and have attempted not to alter our descriptions of
Do not allow “at-a-glance” diagnoses to tempt you these lesions to better fit with the histological diagnosis.
into making a short description or into not making a The algorithm is designed so the most common type
description at all. Part of the learning process is to pro- of misdiagnosis is the false positive for melanoma,
ceed on the basis of pattern analysis even in cases of i.e. those cases in which dermatoscopy suggests the
obvious diagnoses. Experience is needed to establish diagnosis of a melanoma while the histopathological
and perfect your own version of pattern analysis, and investigation yields a benign condition. As overlooking
this includes banal lesions. a melanoma is a far more serious matter than excising
a benign lesion, maximizing true positive findings for
As in actual clinical practice, not all of these lesions melanoma (sensitivity) must take priority over maximiz-
permit an unequivocal diagnosis. In these cases you ing true positive findings for benign lesions (specificity).
should weigh the clues and select the most likely diagno-
sis. Sometimes, even after careful consideration several
equally likely diagnoses may remain. You should take the
process of diagnosis as far as possible, but no further.
If you are uncertain as to how to proceed from descrip-
tion to diagnosis, refer to chapter 5 and follow the
algorithm. A standardized description that follows
this principle, and with which you can compare your
description is shown on the left-hand page opposite
the images. Your description and the one given may
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
310 Cases
Table 10.1
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern More than one color Clue to melanoma: Melanoma must be "Superficial and deep"
(structureless, reticular, (light brown and dark Thick reticular lines excluded congenital nevus
and clods), arranged brown) (between 9 and 11
asymmetrically o’clock)
(chaotic)
Middle One pattern (reticular) Eccentric None Clark nevus Clark nevus
hyperpigmentation
Bottom More than one More than one color Clues to melanoma: Melanoma Melanoma
pattern (structureless, (brown and pink) Polymorphous vascular
and branched pattern, segmental
lines), arranged radial lines
asymmetrically
(chaotic)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 311
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
312 Cases
Table 10.2
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern (reticular) Variegate None Clark nevus or Clark nevus
"superficial"
congenital nevus
Middle One pattern (reticular) One color None Clark nevus Clark nevus
(light brown)
Bottom One pattern More than one color Clues to melanoma: Melanoma Melanoma
(structureless) (melanin) Gray lines and black
dots at the periphery
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 313
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
314 Cases
Table 10.3
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern Dark-brown in the Maybe some Most likely Spitz nevus Spitz nevus
(structureless in the center, light brown segmental pseudopods (symmetry of pattern
center, clods in the in the periphery, no but difficult to and color)
periphery), no chaos chaos differentiate from clods
Middle One pattern (clods) Light brown and gray Clue to melanoma: Unna nevus (the typical Unna nevus
Gray clods pattern of clods and
the symmetry is more
important than a single
clue)
Bottom One pattern (clods) Multiple colors Clues to melanoma: Melanoma Melanoma
arranged Black dots and clods
asymmetrically (chaos), in the periphery,
melanin dominates perpendicular white
lines
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 315
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
316 Cases
Table 10.4
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern More than one color Clues to melanoma: Melanoma Melanoma
(structureless and (melanin dominates) Gray lines;
clods), arranged structureless eccentric
asymmetrically area
(chaotic)
Middle More than one More than one color Clues to melanoma: Melanoma Melanoma
pattern (reticular Black dots in the
and dots), arranged periphery, white lines
asymmetrically
(chaotic)
Bottom One pattern (reticular) Eccentric None Clark nevus Clark nevus
hyperpigmentation
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 317
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
318 Cases
Table 10.5
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern One color (blue) None Blue nevus Blue nevus
(structureless). The (By definition if there
single peripheral is one pattern and one
orange clod is an color there is no chaos)
erosion due to trauma
and has been ignored.
Middle One pattern (reticular) Central None Clark nevus Clark nevus
hyperpigmentation
Bottom One pattern (reticular) Variegate None Clark nevus or Clark nevus
"superficial"
congenital nevus
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 319
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
320 Cases
Table 10.6
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one More than one None Clark nevus or Clark nevus
pattern (structureless color, central "superficial"
in the center, reticular hyperpigmentation congenital nevus
and dots in the
periphery). The
patterns are arranged
symmetrically (no
chaos).
Middle More than one More than one None. The dots in the Clark nevus Clark nevus
pattern (reticular color, eccentric periphery are brown
and dots), arranged hyperpigmentation and not black.
asymmetrically
(chaotic)
Bottom One pattern (reticular) One color (brown) None Clark nevus Clark nevus
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 321
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
322 Cases
Table 10.7
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern More than one color, Thin reticular lines Dermatofibroma Not excised
(structureless in the white in the center in the periphery and
center, reticular in a white structureless
the periphery). The center are a clue to
patterns are arranged dermatofibroma.
symmetrically (no
chaos).
Middle More than one More than one Clue to melanoma: Melanoma or Combined congenital
pattern (clods and color, eccentric Eccentric structureless melanoma in nevus
structureless), arranged hyperpigmentation area association with
asymmetrically a nevus
(chaotic)
Bottom One pattern (circles) Brown (not gray or Clue to seborrheic Seborrheic keratosis or Seborrheic keratosis
black) keratosis: Some circles solar lentigo
are "distorted", i.e.
they appear oval or
even polygonal (see
also figure 7.4 for
higher magnification)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 323
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
324 Cases
Table 10.8
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern Brown Discrete circles and Seborrheic keratosis Seborrheic keratosis
(clods and curved curved lines are clues
lines). The few circles to seborrheic keratosis
were not counted as a
pattern. If the circles
are regarded as an
additional pattern the
diagnosis would be the
same.
Middle One pattern (reticular) One color (brown) None Clark nevus Not excised but no
change during follow-
up therefore Clark
nevus
Bottom One pattern More than one Polymorphous vascular Melanoma (a basal Melanoma (invasion
(structureless) color, melanin (blue pattern, gray and blue cell carcinoma does thickness > 1 mm)
and gray), the structures not occur on acral
colors are arranged skin)
asymmetrically (chaos)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 325
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
326 Cases
Table 10.9
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern (reticular) One color (brown) None Clark nevus or Not excised but no
"superficial" change during follow-
congenital nevus up
Middle One pattern (reticular) Variegate Terminal hairs are a "Superficial" or "Superficial and deep"
clue to a congenital "superficial and deep" congenital nevus
nevus. No clue to congenital nevus
melanoma.
Bottom More than one Structureless = brown White clods are a clue Seborrheic keratosis Seborrheic keratosis
pattern (clods and Clods = white to seborrheic keratosis
structureless)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 327
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
328 Cases
Table 10.10
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one Brown and black Clues to melanoma: Melanoma Melanoma
pattern (clods and Pseudopods in the
structureless). The periphery, black clods
patterns are arranged in the periphery
asymmetrically
(chaos).
Middle One pattern (clods) Brown and gray Discrete orange and Seborrheic keratosis Seborrheic keratosis,
yellow clods and sharp with unconventional clonal type
circumscription point to criteria (gray clods
a seborrheic keratosis are rarely found in a
seborrheic keratosis)
Bottom One pattern (large Skin colored and light Curved vessels in the Unna nevus (in Unna nevus (in
polygonal clods) brown center of skin colored combination with a combination with a
clods. Hyperkeratotic viral wart) viral wart)
zone at 3 o’clock.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 329
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
330 Cases
Table 10.11
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern More than one Clues to basal cell Basal cell carcinoma Basal cell carcinoma
(structureless) color including carcinoma: serpentine
brown and gray vessels, some of which
(melanin), arranged are branched
asymmetrically (chaos)
Middle One pattern (clods) Orange and white Clues to seborrheic Seborrheic keratosis Seborrheic keratosis
clods keratosis: White clods
and orange clods,
looped vessels
Bottom One pattern Skin colored, None Most likely combined "Superficial and deep"
(structureless) light brown, blue congenital nevus congenital nevus
(melanin) arranged
symmetrically (no
chaos)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 331
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
332 Cases
Table 10.12
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern More than one color Clues to seborrheic Seborrheic keratosis Seborrheic keratosis
(structureless) including brown keratosis: discrete
and gray, arranged circles, sharp
asymmetrically (chaos) circumscription, blue
structureless area,
white clods
Middle More than one Central Clues to melanoma: Clark nevus or, less Melanoma in situ
pattern (reticular hyperpigmentation thick reticular lines likely, melanoma. The
in the periphery, symmetry favors a
structureless in the Clark nevus but thick
center), arranged reticular lines are a
symmetrically clue to melanoma.
Bottom More than one Skin colored, light Clues to basal cell Basal cell carcinoma Basal cell carcinoma
pattern (clods and brown, gray (melanin) carcinoma: Serpentine
structureless), arranged vessels, adherent fiber
asymmetrically (chaos)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 333
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
334 Cases
Table 10.13
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern Brown in the None Combined congenital "Superficial and deep"
(clods in the periphery, periphery, gray in the nevus or "superficial congenital nevus
structureless in the center (no chaos) and deep" congenital
center), arranged nevus
symmetrically (no
chaos)
Middle More than one Brown None Clark nevus or Clark nevus
pattern (circles and "superficial and deep"
structureless), arranged congenital nevus
asymmetrically (chaos)
Bottom More than one pattern Brown A peripheral rim of Growing "superficial "Superficial and deep"
(clods in the periphery, clods or dots is a clue and deep" congenital congenital nevus
structureless in the to a growing nevus nevus or Spitz nevus.
center), arranged In a growing Clark
symmetrically (no nevus one expects the
chaos) reticular pattern in the
center.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 335
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
336 Cases
Table 10.14
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern (reticular) Brown None. Some brown Clark nevus Not excised
dots in the periphery
are not specific.
Middle One pattern Skin colored, pink, and Clues to melanoma: Melanoma Melanoma in a pre-
(structureless) brown polymorphous vascular existing Clark nevus
pattern, the presence
of vessels as dots are
in favor of melanoma
and exclude other
diagnoses such as
basal cell carcinoma
Bottom More than one pattern Brown A peripheral rim of Growing "superficial "Superficial and deep"
(clods in the periphery, clods or dots is a clue and deep" congenital congenital nevus
structureless in the to a growing nevus nevus or Spitz nevus.
center), arranged In a growing Clark
symmetrically (no nevus one expects the
chaos) reticular pattern in the
center.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 337
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
338 Cases
Table 10.15
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern Brown A peripheral rim of Growing "superficial Not excised
(clods in the periphery, clods or dots is a clue and deep" congenital
structureless in the to a growing nevus nevus or Spitz nevus.
center), arranged In a growing Clark
symmetrically (no nevus one expects the
chaos) reticular pattern in the
center.
Middle One pattern Black in the center, None Clark nevus or Reed Clark nevus
(structureless) brown at the periphery nevus
Bottom More than one Central Clue to melanoma: Melanoma Clark nevus
pattern (radial lines hyperpigmentation Segmental radial lines
in the periphery,
structureless in the
center), arranged
asymmetrically (chaos)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 339
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
340 Cases
Table 10.16
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one Brown, gray, and skin Serpentine vessels are Basal cell carcinoma Basal cell carcinoma
pattern (dots and colored a clue to basal cell
structureless) arranged carcinoma, one or two
asymmetrically (chaos) blue clods (too few to
form a pattern) support
this diagnosis
Middle More than one Clods are mostly black Clues to seborrheic Seborrheic keratosis Seborrheic keratosis
pattern (clods, dots, and dots are gray keratosis: in regression (lichen
structureless) arranged White and yellow planus-like keratosis)
asymmetrically clods, looped vessels,
sharp demarcation
Bottom One pattern (reticular) Eccentric Clues to melanoma: Melanoma Clark nevus
hyperpigmentation thick reticular lines
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 341
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
342 Cases
Table 10.17
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern (dots). Gray (dots) and red Clues to melanoma: Melanoma Melanoma
If one prefers Gray dots and white
structureless it would lines
lead to the same
diagnosis.
Middle One pattern (clods) Clods are grey, yellow Clues to seborrheic Seborrheic keratosis Seborrheic keratosis
and white keratosis:
Yellow and white clods
Bottom One pattern (reticular) Eccentric None Clark nevus, Clark nevus
hyperpigmentation "superficial"
congenital nevus, or
"superficial and deep"
congenital nevus
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 343
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
344 Cases
Table 10.18
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern Eccentric None Clark nevus, Clark nevus
(reticular and clods). hyperpigmentation "superficial"
If one prefers only congenital nevus, or
reticular it would lead "superficial and deep"
to the same differential congenital nevus
diagnosis.
Middle One pattern (reticular) Eccentric None Clark nevus Clark nevus
hyperpigmentation
Bottom One pattern (reticular) Central None Clark nevus Not excised
hyperpigmentation
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 345
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
346 Cases
Table 10.19
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one Central Clues to melanoma: "Superficial and "Superficial and deep"
pattern (reticular hypopigmentation Gray dots deep" congenital congenital nevus
in the periphery, nevus. Symmetry of
structureless in the pattern and color takes
center), arranged precedence over a
symmetrically single clue (gray dots).
Middle One pattern (reticular) Central Clue to melanoma: As a general Clark nevus
hyperpigmentation Thick reticular lines principle symmetry
of pattern and color
takes precedence
over a single clue.
Therefore the most
likely diagnosis is
Clark nevus. However,
a melanoma in situ
cannot be excluded
with certainty.
Bottom More than one pattern Central Clue to melanoma: As a general Clark nevus
(reticular in the hyperpigmentation Gray clods principle symmetry
periphery, clods in of pattern and color
the center), arranged takes precedence
symmetrically. over a single clue.
Note that the two Therefore the most
eccentric clods are too likely diagnosis is
insignificant to disturb Clark nevus. However,
the overall symmetry. a melanoma in situ
cannot be excluded
with certainty.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 347
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
348 Cases
Table 10.20
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one Brown, orange, yellow, Clues to seborrheic Seborrheic keratosis Seborrheic keratosis
pattern (clods and white keratosis:
structureless) White and yellow
clods, looped vessels
Middle More than one Brown and skin- Clue to melanoma: Most likely "superficial "Superficial and deep"
pattern (dots and colored, central None, the brown dots and deep" congenital congenital nevus
structureless), arranged hypopigmentation at the periphery are nevus
asymmetrically (chaos) not specific
Bottom More than one More than one color Clues to melanoma: Melanoma Melanoma
pattern (reticular and Gray clods and white
structureless), arranged lines
asymmetrically (chaos)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 349
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
350 Cases
Table 10.21
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern Brown Clues to seborrheic Seborrheic keratosis Seborrheic keratosis
(thick curved lines and keratosis:
clods) Discrete circles and
thick curved lines
Middle One pattern Blue and brown None Blue nevus. The Blue nevus
(structureless) (arranged discrete brown
symmetrically) pigmentation in the
periphery is unusual
but in keeping with the
diagnosis of a blue
nevus.
Bottom One pattern Gray and brown Clues to melanoma: Most likely melanoma Melanoma in situ
(angulated lines) Gray dots in situ. Pigmented (lentigo maligna)
actinic keratosis or
lichen planus-like
keratosis (solar lentigo
with regression) are
less likely but not
impossible.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 351
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
352 Cases
Table 10.22
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern (reticular) Brown None Clark nevus Not excised
Middle More than one Brown None Clark nevus, Not excised
pattern (reticular "superficial"
in the periphery, congenital nevus, or
structureless in the "superficial and deep"
center), arranged congenital nevus
symmetrically
Bottom One pattern Brown Terminal hairs are a "Superficial and deep" Not excised
(structureless) clue to a congenital congenital nevus
nevus
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 353
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
354 Cases
Table 10.23
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern (clods) Yellow, orange, brown, Yellow, orange, and Seborrheic keratosis Seborrheic keratosis
and white white clods are a clues
to seborrheic keratosis
Middle More than one pattern Central Clods in the periphery Growing Clark nevus Clark nevus
(clods in the p
eriphery, hyperpigmentation are a clue to a or Spitz nevus
structureless in the (dark brown) growing nevus
center), arranged
symmetrically
Bottom One pattern (reticular) Eccentric Clue to melanoma: Melanoma "Superficial"
hyperpigmentation Thick reticular lines congenital nevus
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 355
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
356 Cases
Table 10.24
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top Non-pigmented lesion, Pink Serpentine vessels, Most likely basal Basal cell carcinoma
nodule branched cell carcinoma but
any neoplasm that
grows underneath
the superficial
vascular plexus may
have serpentine
branched vessels on
dermatoscopy
Middle One pattern Brown Curved lines and Solar lentigo Not excised
(curved lines) a well-demarcated
scalloped border are
clues to solar lentigo
Bottom More than one Gray (dots) and Clue to basal cell Basal cell carcinoma Basal cell carcinoma
pattern (dots and orange and pink carcinoma:
structureless) arranged Gray clods, orange
asymmetrically structureless zone
(chaos). The alternative (ulceration), and
interpretation of one serpentine vessels
pattern (structureless)
would lead to the same
diagnosis.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 357
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
358 Cases
Table 10.25
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top Non-pigmented lesion, Pink White lines (clue Most likely basal Metastasis of a
nodule to malignancy), cell carcinoma but malignant peripheral
serpentine vessels, any neoplasm that nerve sheath tumor
branched grows underneath
the superficial
vascular plexus may
have serpentine
branched vessels on
dermatoscopy
Middle More than one pattern Central A structureless white Dermatofibroma Not excised
(reticular in the hypopigmentation center in the context
periphery, structureless of reticular lines in the
in the center) periphery is a clue to
dermatofibroma
Bottom More than one pattern More than one color Clue to melanoma: Melanoma Melanoma
(reticular, clods, (brown, gray, skin) Gray clods
structureless) arranged
asymmetrically (chaos)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 359
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
360 Cases
Table 10.26
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one pattern Brown Terminal hairs are "Superficial and deep" Not excised
(reticular in the clues to a congenital congenital nevus
periphery, structureless nevus
in the center) arranged
symmetrically
Middle More than one pattern Central None "Superficial and deep" Not excised
(reticular in the hypopigmentation congenital nevus
periphery, structureless
in the center) arranged
symmetrically
Bottom More than one pattern Brown Terminal hairs are "Superficial and deep" Not excised
(reticular in the clues to a congenital congenital nevus
periphery, structureless nevus
in the center) arranged
symmetrically
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 361
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
362 Cases
Table 10.27
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one Brown and gray Clues to melanoma: Melanoma Melanoma
pattern (reticular Gray dots, peripheral
and dots) arranged black dots
asymmetrically (chaos)
Middle One pattern (clods) Brown None "Superficial and deep" "Superficial and deep"
congenital nevus congenital nevus
Bottom More than one Brown, orange, gray, White and orange Most likely seborrheic Seborrheic keratosis
pattern (clods, dots, and white clods are clues to a keratosis
structureless) arranged seborrheic keratosis.
asymmetrically (chaos) Gray clods are clues to
basal cell carcinoma.
However, the sharp
circumscription and
coiled vessels favor the
diagnosis of seborrheic
keratosis over basal
cell carcinoma.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 363
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
364 Cases
Table 10.28
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one Clods are blue and Clues to basal cell Basal cell carcinoma Basal cell carcinoma
pattern (clods and gray carcinoma:
structureless) arranged Blue and gray clods;
asymmetrically (chaos) serpentine, branched
vessels
Middle One pattern (large Brown None Unna nevus (a Not excised
polygonal clods) congenital nevus of
one type)
Bottom More than one More than one color Clues to melanoma: Melanoma Melanoma
pattern (clods, and (melanin dominant) White lines and gray
structureless) arranged clods
asymmetrically (chaos)
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 365
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
366 Cases
Table 10.29
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top More than one Black, gray and brown Some seborrheic Seborrheic keratosis Seborrheic keratosis
pattern (clods and (melanin dominates) keratosis are so
structureless) arranged heavily pigmented that
asymmetrically (chaos) the clods appear black
or dark brown and
not yellow, orange, or
white
Middle One pattern (large Brown and skin- Large polygonal, skin- Unna nevus (a Unna nevus
polygonal clods) colored colored or light brown congenital nevus of
clods are clues to a one type)
mainly intradermal
congenital nevus
(Unna nevus)
Bottom More than one More than color Clues to melanoma: Melanoma Melanoma
pattern (clods, and (melanin dominant) Gray clods, white
structureless) arranged lines, pseudopods,
asymmetrically (chaos) eccentric structureless
area
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 367
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
368 Cases
Table 10.30
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern (clods). Brown None "Superficial and deep" "Superficial and deep"
The clods are arranged congenital nevus congenital nevus
in the furrows but this
does not make them
parallel lines.
Middle More than one Clods are gray and Clues to malignancy: Basal cell carcinoma Basal cell carcinoma
pattern (clods, and brown Gray clods and white
structureless) arranged lines. The small zone
asymmetrically (chaos) of ulceration (small red
clod) favors basal cell
carcinoma.
Bottom More than one More than one color Clues to melanoma: Melanoma Melanoma
pattern (clods and (melanin dominant) Black dots in the
structureless) arranged periphery, segmental
asymmetrically (chaos) radial lines, eccentric,
structureless area
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 369
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
370 Cases
Table 10.31
Pattern Color Clues Dermatoscopic Histopathologic
diagnosis diagnosis
Top One pattern (reticular) Brown None Clark nevus Clark nevus
Middle One pattern (parallel Brown Parallel lines on the Acral melanoma, Intracorneal
lines on the ridges) ridges are found in hemorrhage, hemorrhage
acral melanoma, or exogenous
hemorrhage, pigmentation
and exogenous
pigmentation
Bottom One pattern (parallel Brown Parallel lines on the Acral melanoma, Intracorneal
lines on the ridges) ridges are found in hemorrhage, hemorrhage
acral melanoma, or exogenous
hemorrhage, pigmentation
and exogenous
pigmentation
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Cases 371
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
373
11 Dermatoscopic-dermatopathologic correlation
A B
one finds just a thin layer of melanin at this site as In melanocytic lesions the melanocytes are increased in
well, and occasionally one does actually see double number and at least partly arranged in nests. When a
lines (11.2). nevus is situated in the epidermis the nests are mainly
However, compared to the dermal papilla this zone is at the base of the rete ridges. However, if the melano-
quite narrow and therefore usually creates the impression cytes that constitute the nests are not filled with melanin
of a single line. Very broad rete ridges create a pattern they are invisible on dermatoscopy. Even so, a reticular
of circles rather than reticular lines (11.3). pattern may still be seen, due entirely to hyperpigmenta-
For this reason, a pattern of closely adjacent circles tion of basal keratinocytes (11.1B). Nests of pigmented
in lesions on the trunk or limbs should be interpreted melanocytes are seen as brown dots or small brown
identically to a pattern of reticular lines. clods. However, pigmented melanocytes at the base of
As a proliferation of melanocytes is not necessary to rete ridges may not be arranged in nests, but may be
produce reticular lines, it is not surprising that some spread out as a confluent proliferation of single cells.
non-melanocytic lesions have reticular lines on der- To use an analogy: dermal papillae extend vertically
matoscopy. Typical examples are solar lentigines, flat like mountain tops while the rete ridges form the inter-
seborrheic keratoses and dermatofibromas. vening valleys. The valleys (base of rete ridges) are
Therefore it is impossible to establish, on the basis coated with a confluent proliferation of melanocytes
of reticular lines, whether a lesion is melanocytic or that, when viewed from above, appear as reticular
non-melanocytic (11.4). lines. However, on histological (cross-) sections the
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
A B
C D
E F
G H
I J
K L
of melanocytes within the flanking rete ridges causes In histological cross-sections cut across ridges and
pigmentation of the epidermal furrows while prolifera- furrows, a contiguous proliferation of melanocytes
tion of melanocytes in the broader middle rete ridges may look like nests. However, when one looks at the
causes pigmentation of the epidermal ridge (11.7). The dermatoscopy, pigmentation on the ridges or in the
ridges are usually broader than the furrows, and the furrows is nearly always seen as lines and not as dots
eccrine duct openings are seen as a row of white dots or clods arranged as lines, suggesting that the mela-
along the middle of the ridge. nocytes are indeed arranged as a contiguous strand.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Figure 11.16: Melanoma in association with a “superficial and deep” congenital nevus.
Most melanomas, i.e. about 80–85 % of them in “Caucasians”, develop de novo. When, however, melanoma develops in a pre-existing
nevus, that nevus is likely to be congenital, usually one “superficial and deep” and not a Clark nevus. Clinically this lesion is multicolored
and asymmetric. One can see an increased number of terminal hairs. On dermatoscopy the lesion is chaotic with white reticular lines as a
clue to melanoma. At scanning power magnification, two completely different architectural patterns are evident, namely, that of a “superfi-
cial and deep” congenital nevus on the left and a melanoma on the right. The patterns contrast sharply, the congenital nevus being present
in abundance in the upper part of the reticular dermis, thereby qualifying it as “superficial and deep”, with no nests at the dermoepidermal
junction. The melanoma is housed mostly in the epidermis, but also in the papillary dermis, and shows a dense infiltrate of lymphocytes.
The reticular lines correspond to pigmented keratinocytes in the epidermis, especially the basal layer. White reticular lines correspond to
fibroplasia of the papillary dermis beneath an epidermis that displays distinct rete ridges. The white lines are thick because the dermal
papillae are broad.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
387
Supplement
“One of the symptoms of an approaching nervous breakdown is the belief that one’s work is terribly important.”
Bertrand Russell
“Los globulos marrones” with Harald Kittler (guitar), Philipp Tschandl (keyboard), Giuseppe Argenziano (drums), Luc Thomas (guitar), Peter
Bourne (guitar) live in Vienna, Austria during the World Congress of Dermatoscopy 2015.
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
389
Index
3-point checklist 17 C
4-dot clod 210 calcinosis cutis 210, 221
7-point check-list (Argenziano) 17 Carney complex 49
“atypical” Spitz nevus 35 CASH algorithm 17
central white patch 129
A cerebriform pattern 129
ABCD rule (Stolz) 17 chaos 191f
accessory nipple 285 chaos and clues 18, 190
Ackerman nevus 31 cherry angioma (tardive angioma) 42
acral lesion 260 chrysalids 130
acral nevus 28 chrysalis 130
actinic keratosis 50, 215 circles 55
adherent fiber 235f Clark nevus 29f, 97
adnexal neoplasm 50 clear cell acanthoma 221, 223
age spots 49 cliché 243ff
amelanotic melanoma 229 clods 55
angiokeratoma 42ff clues 66f, 193ff
angulated lines 55, 125f, 242 clues to malignancy 192f
annular-granular pattern 126 cobblestone pattern 130
apocrine cysts 51 combined congenital nevus 32f, 104
atypical fibroxanthoma 230 combined nevus 33
atypical nevus 37 comedo-like openings 130
atypical pigment network 127 compound nevus 39
atypical Spitz nevus 35 congenital nevus 31f
crown vessels 132
B crypts 132
bacteria 291 crystalline 130
Bannayan-Riley-Ruvalcaba syndrome 49 curved lines 55
BAPomas 40 cutaneous amyloidosis 290
basal cell carcinoma 50, 89, 215 cutaneous Lupus erythematosus 287f
biopsy of the nail matrix 257
black heel 45 D
blotch 127, 140 dermal nevus 35, 39, 228
blue nevus 29, 106 dermatofibroma 51, 81
blue veil 128 dermatoscopy of nail pigmentation 253
blue-gray ovoid nests 128 digital dermatoscopic monitoring 297ff
blue-white veil 128 digital mucoid (myxoid) cyst 225
Bowen’s disease 50, 93, 215 dots 55
brain-like pattern 129 double reticular lines 238
branched lines 55 dysplastic nevus 37
branched streaks 129
broadened pigment network 129
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
390 Index
E K
eccrine poroma 221, 224, 232 Kaposi sarcoma 43f, 284
exogenous pigmentation 266 Kaposi’s disease (Kaposi sarcoma) 44f
keratin 204
F keratinocyte cancer 50
fat fingers 132 keratoacanthoma 224, 226
fibrillar pattern 132
fibropithelioma of Pinkus 231 L
fingerprinting 132 labial lentigo 89, 279
fissures and ridges 133 large cell anaplastic T-cell lymphoma 232
folliculitis 207, 209 lattice-like pattern 134
four dots in a square (four-dot clod) 238 Laugier-Hunziker syndrome 47, 49
four-dot clod 210 leaf-like structure 135
freckle 49 leishmaniasis 210, 221
fungal infection of the nail plate 259 lentigines (melanotic macules) 280
fungi 290f lentiginosis of the lip 47
furuncle 207 lentigo 45
lentigo solaris 49
G LEOPARD syndrome 47, 49
genital lentiginosis 47, 280 lichen planus 215, 286, 288f
genital lentigo 89 lichen planus-like keratosis 50, 81, 278
globules (globuli) 133 liver spots 49
granulomatous inflammation 210 Lupus erythematosus 215, 286
Grover’s disease 286, 290 lymphangioma 210
gyri 129
M
H malignant peripheral nerve sheath tumor 230
hemangioma 42, 74f maple leaf-like areas 135
hemoglobin 64 melanin 62
hemorrhage 45 melanocytoma 40
high-grade dysplastic nevus 37 melanoma 42, 113
homogeneous pattern 133 melanoma of the nail plate 255
Hortaea werneckii 51 melanotic macule 45, 81
Hutchinson’s sign 255 MELTUMP 35
hypermelanotic nevus 40 Menzies’ method 17
Merkel cell carcinoma 221, 227
I metastases of melanoma 121
ink-spot lentigo 47f, 81 metastasis 221
intracorneal hemorrhage 77 Meyerson nevus 40
intraepidermal carcinoma 50 micro-Hutchinson’s sign 255
inverse (negative) pigment network 134 Miescher nevus 35, 113
irregular peripheral extensions 136 milia-like cysts 135
irregular pigment network 127 milky red areas 136
Molluscum contagiosum 221, 286, 290
J monomorphous 215
junctional nevus 39 moth-eaten border 136
mucosal lentiginosis 280
mucosal lesions 280
mucosal melanoma 280
myiasis 207
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
Index 391
N S
nails 253f. sarcoidal rosacea 286
nevus araneus (“spider nevus”) 42, 44 sarcoidosis 286
nevus flammeus (“port-wine stain”) 44 scabies 286, 291
nevus sebaceous 211 scale 203
nevus spilus 35 scar-like depigmentation 139
non-pigmented (amelanotic) lesions 203 sebaceous gland hyperplasia 207, 209, 221
nummular dermatitis 215 seborrheic keratosis 49, 80
shiny white lines 130
P shiny white streaks 130, 140
parallel lines 55 solar lentigines with regression 278
parasites 291 solar lentigo 77
pattern analysis 16, 18 solitary angiokeratoma 76
pediculosis 286 spider nevus (nevus araneus) 42, 215
penile lentiginosis 279 Spitz nevus 35, 106
peppering 136 spoke-wheel areas 140
peripheral streaks 136 spongiotic dermatitis 215, 286
periungual pigmented Bowen’s disease 267 squamous cell carcinoma 50, 89
Peutz-Jeghers syndrome 47, 49 starburst pattern 141
picker’s nodule 221 stasis purpura 51
pigment network 136f strands 140
pigmented actinic keratosis 89, 278 strawberry pattern 141
pigmented purpura 51 string of pearls 142
pilomatrixoma 207, 210, 221 structureless pattern 55
pityriasis rosea 215, 286 subungual bleeding 258
polygons 125f, 242 sulci 129
porokeratosis 215, 286, 288 superficial and deep congenital nevus 31, 97
prurigo 221 superficial congenital nevus 31, 97
pseudo-Hutchinson sign 256 surface scale 203
Pseudomonas aeruginosa 259 Sutton nevus (Halo nevus) 35
pseudo-network 137, 269
pseudopods 55 T
psoriasis 215, 286, 287 tardive angioma (cherry angioma) 42
PUVA lentigines 47f targetoid dots 142
pyogenic granuloma 42, 76, 207 targetoid vessels 142
tattoo 51, 284
R telangiectasia macularis perstans 215
radial lines 55 Tinea nigra 51, 286, 290, 292
radial streaming 137 trichoblastoma 50
rainbow pattern 138 trichoepithelioma 224
recurrent melanocytic lesions 280 Trichomycosis palmellina 286, 291
recurrent nevus 35, 106 tungiasis 286
red lacunes 138
Reed nevus 35, 106 U
reticular depigmentation 134, 138 ulceration 203
reticular lines 53 Unna nevus 35, 113, 228
rhomboids 126, 139
rosettes 139, 238
© Dies ist urheberrechtlich geschütztes Material. Bereitgestellt von: TH Mittelhessen Mo, Okt 5th 2020, 09:05
392 Index
V
Verruca genitalis 286
Verruca palmaris 286
Verruca plana 286
Verruca vulgaris 286
vessels 70ff
viral warts 221, 290
vulvar melanoma 279
W
white circles 208, 242
white clues 205
white lines 241
white scale 215
Wiesner nevus 40
X
xanthelasma 210
xanthogranuloma 210f
Z
zig-zag pattern 126, 142
Zitelli nevus 31f