DISCUSSION
There are 4 main dermoscopic structures in dermatofibromas, mainly
peripheral delicate pigment network, central white scar-like patch,
white network and homogeneous pigmentation. (2). Delicate pigment
network associated with dermatofibromas is usually thin, light and
medium brown, graceful and gradually pale towards the periphery of the
lesion (2).
Although the pigment network in DFs is quite similar to that of
melanocytes, the pigment network in DF is caused by hyperpigmentation in
rete ridges rather than melanocytic proliferation in the basal layer
(2,5). Arpaia et al., in their study evaluating 32 DF, detected a
peripheral delicate pigment network in 31% of the cases. According to
the study, the pigment network was darker in the center, but gradually
pale towards the periphery and became brownish thin streaked (6).
In 2006, Ferrari et al. detected a pigment network in 83.3% of the
cases (5). The same author detected the peripheral pigment network in
50.8% of the cases in 2013. They also found that the peripheral pigment
network and central white scar-like patches are associated with female
sex and the classic histopathology variant of DFs (7). Kelati et al.
found a pigment network in 79% of the lesions in their study, in which
they evaluated 100 DF (3). Agero et al. found that the pigment network
rate in DF was 72% in nonpolarized dermoscopy, 72% in polarized
contact dermoscopy, and 54% in polarized noncontact dermoscopy (1).
Karaarslan et al., on the other hand, observed the pigment network,
which they defined as lentigo-like reticular pigmentation, in 23% of
DFs (8). Zaballos et al, in their study evaluating 412 DF, detected
pigment network in 71.8% of DFs (2). In our study, we found pigment
network in 81 (57%) of DFs. We think that the very different pigment
network ratios among previous studies may be due to factors such as the
quality of the dermoscopes, contact or non-contact, being polarized or
nonpolarized, and the difference between genders in the studies. We also
observed that the pigment network was present in the holes in the white
network in some lesions.
Another dermoscopic structure frequently observed in dermatofibromas is
the central white scar-like patch. These structures are irregular and
sharply demarcated white areas. It is characterized histopathologically
by distinct fibrosis in the papillary dermis (8). Zaballos et al., in
their study evaluating 412 dermatofibroma, detected white scar-like
patch mostly localized in the center of the lesion in 57% of patients
(2). In the study of Ferrari et al, this rate was 91.6% (5). The most
common structure that Agero et al. encountered in DFs with non-polarized
and polarized contact dermoscopy was the white patch. In this study,
they found central white patch in 84% of dermatofibromas by polarized
contact dermoscopy (1). These structures were found to be mostly well
and irregularly circumscribed and sometimes in the form of a star. They
observed that the central white patch seen in 33 lesions with polarized
contact dermoscopy was characterized by shiny white streaks (1). In the
study of Arpaia et al, the most common dermoscopic structure was the
central white patch with 91.6% (6). Ferrari et al detected white area
in 50% of DFs (7). In another study, a central white patch was found in
70% of DFs (3). This structure is thought to be the most stereotypic
and widespread manifestation of diffuse fibrous dermatofibroma with
peripheral delicate pigment network (3,5,8). In our study, a central
white scar-like patch was detected in 37.3% of DFs. In the periphery of
this central structure, the scar structure sometimes turned into a white
radial streaks appearance. Some white radial streaks transformed the
appearance of the central scar-like patch into a spitzoid pattern
(Fig. 5 ).
Zaballos et al. identified a new dermoscopic structure, which they
identified mostly in large DFs in 2006 and named as white network (9).
This network of white lines and brown holes, seen in 17.7% of
dermatofibromas in the study at 2008, is considered a variation of the
classic white scar-like patch with a similar histopathological
correlation. It is important to distinguish this structure from the
Spitz nevi, dysplastic nevi and negative pigment network in melanomas.
In suspected cases or cases with an irregularly distributed white
network, excision is required. (2). Ferrari et al. Observed the white
network structure in 30% of DFs in their study in 2013 (7). In our
study, white network was detected in 25.3% of the cases.
The term central white scar-like patch, white network and globule-like
structures have been used intertwined in the past. For this reason, we
think that the information about globule-like structures from past to
present is controversial. Agero et al. did not use the term white
network in their study in 2006. Instead, the term globule-like
structures was used and this structure was detected 44% in ratio when
polarized contact dermoscopy was used (1). In fact, the globule-like
structure appears to be ’holes’ in the white network. Ferrari et al.
stated in their study in 2000 that there were light and dark brown
globules and dots within the central white scar-like patch (5).
Karaarslan et al. reported that the main feature of the lesion, which
they defined as a type 1 lesion, is globules in the scar-like area. They
observed this structure at a rate of 38.5% in their study (8). In our
opinion, this term is the name given to the combination of white
scar-like patch and a white network. This term has been used before
(5,6,10). In fact, it has been stated that these globules are not real
globules but ’holes’ in the central white fibrotic network (8). Later
Zaballos et al. proposed the term central white network for globules
within the scar-like area (9). In our study, the ”holes” seen in the
white network were not evaluated as a globul-like structure and it was
observed at a rate of 11.3% in DFs. Ferrari et al found globul-like
structures and ring-like structures, which they consider to be its
(globul-like structures) variant, in 27.7% of the cases. They stated
that these structures are not caused by melanocyte nests, but by flat,
confluent and hyperpigmented rete ridges. In addition, in this study,
unlike other studies, the white network structure and globule-like
structures within the central white patch were evaluated separately (7).
Flattened and dilated rete ridges are also responsible for small
ring-like structures, which are globules with darker peripheral rims
(2). Zaballos et al. detected globule-like structures in 41.6% of the
lesions and ring-like structures in 25% of the lesions. However,
according to our observation, globule-like structures close to the
center are brown holes in the white network structure. We also think
that ring-like structures are a more regular, darker, rounder and
thicker lined variant of the peripheral delicate pigment network.
Ring-like structures were found 5.6% in our study.
Another basic dermoscopic finding of dermatofibromas is homogeneous
pigmentation. Agero et al observed that brown homogeneous areas, which
they classified as the peripheral halo of brown homogeneous
pigmentation, were 4% in nonpolarized dermoscopy and 12% in polarized
noncontact dermoscopy (1). Ferrari et al. observed homogeneous area at a
rate of 6.9 in their study in 2013. They found total homogeneous
pigmentation most frequently in women and DFs with sebaceous
hyperplasia, and peripheral homogeneous pigmentation most frequently in
men. (7). Kelati et al. found homogeneous pigmentation at a rate of 36%
in their studies (3). Karaarslan et al. detected homogeneous bluish
pigmentation at a rate of 5.9% and found that this appearance was
associated with the hemosiderotic variant of the dermatofibroma (8). On
the other hand, Zaballos et al. detected the homogeneous areas at a rate
of 24.8%. They observed that 4.4% of the lesions had yellowish
homogeneous areas. (2). In our study, brown homogeneous areas were found
at a rate of 25.4%, yellowish homogeneous areas at a rate of 12.6%,
multiple colored homogeneous areas at a rate of 4.9%, hypopigmented
homogenous areas at a rate of 20.4% and perifollicular hypopigmentation
at a rate of 5.5%.
Vascular structures are one of the criteria used in dermoscopic
diagnosis of melanoma and other pigmented and vascular tumor lesions
that can mimic melanoma. Ferrari et al. found a reddish coloration
around the central white scar-like patch in 29.2% of the 24 series with
dermatofibromas (5). Agero et al. have observed blood vessels in 44% of
dermatofibromas when using polarized contact dermoscopy (1). These
authors also found a central pink shade in 10% of cases and a
peripheral diffuse pinkish or reddish area in 28% (1). Zaballos et al.
found vascular structures in 49.5% of dermatofibromas. The most common
vascular structure in this study was erythema (31.5%) followed by
dotted vessels (30.6%) (2). Ferrari et al. have described 2
dermatofibromas with dotted vessel pattern (11). In 2013, Ferrari et al.
detected vascular structure in 50% of DFs (7). Kelati et al. observed
vascular structure in 30% of the cases. They stated that the most
common vascular structures were vessels in the form of dots and commas
(3). In our study, the most common vascular structures we found were
erythema (48.5%) and dotted vessels (23.9%).
It was evaluated by Zaballos et al. that dermoscopic structures such as
scales (12.4%) and ulceration (4.4%) detected in dermatofibromas were
due to external injuries. It was stated that the yellowish homogeneous
areas observed in 4.4% of the cases corresponded histopathologically to
lipophage and Touton giant cells, and these structures were detected in
cholestatic and lipidized DF or some early lesions. In addition, fissure
and ridges (2.7%) and exophytic papillary structures (1.5%) were
observed in DFs with significant epidermal hyperplasia (2). These
dermoscopic structures are seen in seborrheic keratoses and less
frequently compound and dermal nevi, and represent keratin-filled
epidermis invaginations (12-15).
In their dermoscopic study performed by Zaballos et al on 412 DF, they
divided DFs into 11 patterns, one of which is subtype, according to
their dermoscopic appearance. The most common pattern they observed was
the combination of the central white scar-like patch and peripheral
delicate network, which they classified as pattern 2 (34.7%). The
second most frequent was pattern 1 (14.6%) with only the total delicate
pigment network (2). Ferrari et al. determined pattern 2 (17.7%) and
pattern 6 (total homogeneous pigmentation, 11.5%) in their study (when
evaluated according to the patterns of Zaballos et al.). (7). In the
studies of Kelati and Juliandri, it is seen that the most common pattern
is the second pattern (3,16). In our study, the most common patterns we
found were pattern 1 (21.9%), pattern 8 (14.8%), pattern 2 (13.4%),
respectively. We think that this difference may be due to our
reclassification of DFs from a different point of view and racial
difference.
As a result, with this study, the patterns of DF’s were reclassified by
preserving the basic patterns. We think that the new sub-patterns and
schematization with this study can contribute to better understanding of
DFs.
Conflicts of interest: There are no conflicts of interest.
Financial support: There is no financial support.
Data availability statement: No data are available.