Report

The role of histological presentation in erythroderma

Matteo Megna1, MD, Akmal A. Sidikov2, MD, Denis V. Zaslavsky2, MD, PhD,
Igor N. Chuprov3, MD, Elena A. Timoshchuk2, MD, Uliya Egorova2, MD,
Joerg Wenzel4, MD, and Ruslan A. Nasyrov2, MD, PhD

1
Department of Dermatology, University of
Naples Federico II, Napoli, Italy,
2
St. Petersburg State Pediatric Medical
University, Saint Petersburg, Russia,
3
North-Western State Medical University by
the name I.I. Mechnikov, Saint Petersburg,
Russia, and 4Department of Dermatology,
University Hospital of Bonn, Germany
Correspondence
Matteo Megna, MD
Department of Dermatology
University of Naples Federico II
Via Pansini, 5
80131 Napoli
Italy
E-mail: mat24@libero.it
Funding: None declared.
Conflicts of interest: None declared.
doi: 10.1111/ijd.13488
Abstract
Background Erythroderma is a serious medical condition characterized by inflamed red
skin involving over 90% of the body. It can be the common presentation of different
diseases, therefore clinical diagnosis can be problematic. Controversial data are reported
regarding the diagnostic value of histological examination in erythroderma subjects.
Methods A retrospective study was performed, investigating histological skin specimens of
patients with a clinical diagnosis of erythroderma admitted to the Department of
Dermatology of State Pediatric Medical University, Saint Petersburg, from 2001 to 2014.
Histopathology examination was performed in each case by a pathologist with a special
interest in the skin disease who was blind to any clinical information as well as to final
diagnosis.
Results Blinded histopathology examination alone was able to give the correct diagnosis in
61% (n = 50/82) of cases when compared to final diagnosis. A diagnosis of psoriasis was
made in 23.2% (n = 19/82) of subjects, spongiotic dermatitis/eczema in 20.7% (n = 17/82),
mycosis fungoides in 8.5% (n = 7/82), and drug eruption in 8.5%; histological diagnosis
was inconclusive or not matching the final diagnosis when available in the remaining
39.1% of cases (n = 32/82).
Conclusion Erythroderma remains a condition difficult to study and treat. We showed that
a correct judgment about its cause can be based on objective histopathological criteria in
up to 60% of cases. More studies are needed to try to find out further histological and/or
immunohistochemical markers that could help the clinician with the erythroderma etiology
diagnostic process.

nonspecific, only showing hyperkeratosis, parakeratosis, acan-

Introduction
Erythroderma is defined as a diffuse redness of the skin accom-
panied by a variable degree of scaling involving more than 90%
of the body surface. Although it is potentially a life-threatening
condition with a high rate of mortality because of both the pri-
mary disease causing erythroderma as well as the pattern of
the resulting metabolic alterations, it is still poorly studied in lit-
erature. Erythroderma does not represent a defined entity, pos-
sibly being the clinical presentation of numerous different
diseases.1,2 Indeed, its etiology is very diverse including a vari-
ety of inflammatory dermatoses (atopic dermatitis, eczema,
pityriasis rubra pilaris, psoriasis, etc.), cutaneous T-cell lym-
phoma (mycosis fungoides and Sezary Syndrome), as well as
drug-induced forms, hematological or internal malignancies, and
other diseases.1,2 Therefore, histological investigation repre-
sents a fundamental aid in the differential diagnosis of erythro-
derma, often being mandatory. However, in erythroderma
subjects, the results of histological examination can be
thosis, and chronic inflammatory infiltrate with or without
eosinophils.3–9 Conflicting views exist about the diagnostic value
of skin biopsy in the investigation of erythrodermic patients.9–11
Literature has previously emphasized the possibility of having
nondiagnostic biopsies even in erythrodermic patients with a
definite history of pre-existent dermatosis,10 and clinical and
pathological correlation in erythroderma may be a very compli-
cated task.4 On the other hand, Rothe et al. highlighted the
necessity to combine clinical data with the histopathological fea-
tures of multiple biopsies over time,6 whereas in two different
studies involving 40 patients with erythroderma and a total of 56
skin biopsies, a group of dermatopathologists achieved a high
mean of diagnostic accuracy (53–66%) only relying on histologi-
cal examination.9,11
Since the data regarding the value of histological examination
in erythroderma subjects are opposing, the aim of this retro-
spective study was to evaluate the accuracy of histopathologic
examination in erythroderma diagnosis and to highlight the 1

ª 2017 The International Society of Dermatology International Journal of Dermatology 2017

Report Histology in erythroderma Megna et al.

diverse microscopic features which characterize different ery-

throderma etiologic subtypes.
Material and methods
A retrospective study was performed, investigating histological
skin specimens of adult patients with a clinical diagnosis of
erythroderma admitted to the Department of Dermatology of
State Pediatric Medical University of Saint Petersburg from
2001 to 2014. The study population involved 82 erythrodermic
patients whose skin samples were taken from the archives of
the Department of Dermatology of State Pediatric Medical
University of Saint Petersburg. Skin biopsies were taken within
the second day of hospitalization as registered in patient’s case
files. Histopathology examination (through standard hematoxylin
and eosin and periodic acid–Schiff staining) was performed in
each case by a pathologist with a special interest in the skin
disease who was blind to any clinical information as well as to
final diagnosis (established on the basis of combined clinical
and laboratory data and response to therapy). Therefore, a
diagnosis was only based on objective microscopic data
through a detailed checklist of dermal and epidermal
abnormalities (Table 1), which were registered for each case.
Inflammatory cell infiltration in the epidermis or dermis (number
of cells in the four fields of view) was evaluated following these
criteria: 0–20: poorly expressed infiltrate, 21–49: moderately
expressed infiltrate, and >50 highly/severely expressed infiltrate.
The blinded histological diagnosis was compared with final
nonblinded diagnosis in each case.
Results
The study population consisted of 82 patients (55 men and 27
women) with a mean age of 73 years (range 25–95). A total of
95 skin biopsies have been analyzed since 11/83 patients had
yielded more than one specimen. Blinded histopathology exami-
nation alone was able to give the correct diagnosis in 61%
(n = 50/82) of cases when compared to final diagnosis made on
the basis of combined clinical and laboratory data and response
to therapy. In particular, the diagnosis of psoriasis was made in
23.2% (n = 19/82) of subjects, spongiotic dermatitis/eczema in
20.7% (n = 17/82), mycosis fungoides in 8.5% (n = 7/82), and
drug eruption in 8.5%; histological diagnosis was inconclusive
or not matching the final diagnosis when available in the
remaining 39.1% of cases (n = 32/82). With regard to a psori-
atic subgroup of patients (n = 19, mean age 70.2 years, 18
male 1 female), the most common observed histological feature,
apart from the nonspecific presence of inflammatory lympho-
cytes in dermis, was psoriasiform acanthosis (n = 17/19; 89%)
followed by the presence of neutrophils >50 cells in both the
epidermis and dermis (n = 16/19; 84%), diffuse parakeratosis
(n = 11/19; 58%), and diffuse hypogranulosis (n = 11/19; 58%).
Other histological features observed in this group of patients
are shown in Table 2.
The second most common diagnosis was spongiotic dermati-
tis/eczema (n = 17, mean age 74.2 years, 10 male and 7
female). The nonspecific presence of inflammatory lymphocytes
in the dermis, mainly as superficial infiltrate, and exocytosis
were registered in all patients (n = 17/17, 100%). Other frequent

Table 1 Checklist of microscopic details documented in each case

Inflammatory infiltrate Inflammatory infiltrate

Epidermis Derma in epidermis in derma

Orthokeratosis Red blood cell extravasation Neutrophils Neutrophils

Hyperkeratosis – local in papillary dermis Inflammatory lymphocytes Inflammatory lymphocytes
Hyperkeratosis – diffuse Dilated blood vessels in papillary dermis in basal layer Atypical lymphocytes
Parakeratosis – local Coarse collagen in papillary dermis Atypical lymphocytes Histiocytes?
Parakeratosis – diffuse Edema of papillary layer Lymphocytic microabscesses Eosinophils
Parakeratosis – serum Sclerosis of papillary layer Histiocytes Melanophages
Parakeratosis – neutrophils Eosinophils Plasma cells
Acanthosis – regular Linear arrangement of Exocytosis
Acanthosis – irregular lymphocytes in basal layer Epidermotropism
Acanthosis – psoriasiform Superficial
Atrophy or hypoplasia – local Superficial and deep
Atrophy or hypoplasia – diffuse Lichenoid
Hypergranulosis – local Moderately dense
Hypergranulosis – diffuse
Hypogranulosis – local
Hypogranulosis – diffuse
Spongiosis – local
Spongiosis – diffuse
Necrotic keratinocytes
Apoptotic cells
Hydropic degeneration
of basal layer

International Journal of Dermatology 2017 ª 2017 The International Society of Dermatology

Megna et al. Histology in erythroderma Report 3

Table 2 Detailed histopathological features of erythroderma Table 4 Detailed histopathological features of erythroderma
patients with the diagnosis of psoriasis patients with the diagnosis of mycosis fungoides

Histological N Histological N N Histological N

features (total features (total Histological (total features (total
(epidermis) n = 19) % (dermis) n = 19) % features (epidermis) n = 7) % (dermis) n = 7) %

Parakeratosis 8 42 Neutrophils 16 84 Atypical lymphocytes 7 100 Atypical 7 100

(local) >50 cells (lympocytic lymphocytes
Parakeratosis 11 58 Neutrophils 16 84 microabscesses) >50 cells
(diffuse) <50 cells Linear arrangement 7 100 Atypical 0 0
Hypogranulosis 8 42 Dilated blood 14 77 of lymphocytes lymphocytes
(local) vessels in in basal layer <50 cells
papillary Atypical lymphocytes 7 100 Superficial and 4 57
dermis (in epidermis) deep infiltrate
Hypogranulosis 11 58 Inflammatory 19 100 Epidermotropism 7 100 Moderately 3 28
(diffuse) lymphocytes dense infiltrate
Acanthosis 17 89 Eosinophils 10 53
(psoriasiform)
Acanthosis 2 11
(irregular) between local and diffuse spongiosis, this condition was
Neutrophils 16 84 reported in all cases (Table 3).
>50 cells
With regard to the seven patients with mycosis fungoides
Neutrophils 3 16%
<50 cells (n = 7, mean age 77.8 years, six male and one female), the
diagnosis was strongly suggested by the presence of atypical
lymphocytes in the epidermis (lymphocytic microabcesses), lin-
Table 3 Detailed histopathological features of erythroderma ear arrangement of lymphocytes in the basal layer, atypical lym-
patients with the diagnosis of spongiotic dermatitis/eczema phocytes >50 cells in the dermis, and epidermotropism which
were found in all cases (Table 4). In four of seven skin biopsies
N Histological N (57%), a superficial and deep infiltrate in the derma was noted,
Histological (total features (total whereas in three patients (28%), there was a moderately dense
features (epidermis) n = 17) % (dermis) n = 17) %
infiltrate. A total absence of any signs of histologic eczematous
Parakeratosis (local) 6 35 Inflammatory 17 100 patterns of inflammation was registered.
lymphocytes The fourth group of patients was comprised of seven cases
>50 cells of drug eruption/drug-induced erythroderma (n = 7, mean age
Parakeratosis (diffuse) 3 18 Inflammatory 0 0 70.8, four male and three female). Histological features in this
lymphocytes
group were the ones distinctive of interface (vacuolar type) der-
<50 cells
Parakeratosis (serum) 6 35 Eosinophils 4 24 matitis: hydropic degeneration of the basal layer of the epider-
>50 cells mis, the presence of inflammatory lymphocytes in the basal
Parakeratosis 2 12 Eosinophils 13 76 layer of the epidermis, presence of colloidal cells (apoptotic
(neutrophils) <50 cells cells) in the epidermis, exocytosis, superficial infiltrate in the
Acanthosis 3 18 Exocytosis 17 100
dermis, and infiltration of lymphocytes >50 cells in the dermis
(psoriasiform)
Acanthosis (irregular) 14 82 Superficial 17 100 were observed in all cases (Table 5). Moreover, the presence
infiltrate of apoptotic cells <50 in the epidermis was observed in 5/7,
Spongiosis (local) 8 47 71% of patients.
Spongiosis (diffuse) 9 53 Of the remaining patients where histology alone was incon-
Inflammatory 8 47
clusive, in 13/32 subjects (40.6%) no conclusive diagnosis could
lymphocytes >50 cells
Inflammatory 9 53 be established (cases were considered as idiopathic erythro-
lymphocytes <50 cells derma) whereas combined clinical, laboratory data, and
Eosinophils >50 cells 4 24 response to therapy helped to indicate a diagnosis of psoriasis,
Eosinophils <50 cells 7 41 drug-induced erythroderma, pityriasis rubra pilaris, adult atopic
dermatitis, cutaneous T-cell lymphoma, lichen planus, and Nor-
histological features of this group of patients were represented wegian scabies in 8/32 (25%), 4/32 (12.5%), 2/32 (6.2%),
by irregular acanthosis (n = 14/17; 82%), the presence of eosi- 2 (6.3%), 1 (3.1%), 1 (3.1%), and 1 (3.1%) cases, respectively.
nophils <50 cells in the dermis (n = 13/17; 76%), and diffuse The main histological features of the major subgroups of
spongiosis (n = 9/17; 53%). However, without differentiating patients (idiopathic erythroderma, psoriatic erythroderma, and

ª 2017 The International Society of Dermatology International Journal of Dermatology 2017

Report Histology in erythroderma Megna et al.

Table 5 Detailed histopathological features of erythroderma Table 6 Detailed histopathological features of the main sub-
patients with the diagnosis of drug eruption/drug-induced groups of erythroderma patients where histology alone was
forms insufficient to make a diagnosis

Histological N N Erythroderma
features (total Histological (total subgroup Histological features N %
(epidermis) n = 7) % features (dermis) n = 7) %
Idiopathic Acanthosis (irregular) 10 77
Necrotic 0 0 Inflammatory 7 100 erythroderma Acanthosis (regular) 3 23
keratinocytes lymphocytes (n = 13) Hypogranulosis (local) 3 23
>50 cells >50 cells Parakeratosis (local) 13 100
Necrotic 2 28 Inflammatory 0 0 Exocytosis (local) 7 54
keratinocytes lymphocytes Inflammatory lymphocytes 7 54
<50 cells <50 cells in dermis >50 cells
Apoptotic cells 8 8 Superficial 7 100 Inflammatory lymphocytes 6 46
>50 cells infiltrate in dermis <50 cells
Apoptotic cells 5 71 Eosinophils 3 43 Eosinophils in dermis >50 cells 4 31
<50 cells >50 cells Eosinophils in dermis <50 cells 9 69
Hydropic 7 100 Eosinophils 3 43 Superficial infiltrate 13 100
degeneration <50 cells Psoriasis (n = 8) Acanthosis (irregular) 4 50
of basal layer Acanthosis (regular) 4 50
Inflammatory 7 100 Exocytosis 7 100 Hypogranulosis (local) 5 62
lymphocytes in Parakeratosis (local) 5 62
basal layer Exocytosis (local) 1 12
Inflammatory lymphocytes 3 37
in dermis >50 cells
drug-induced erythroderma) where histology alone was not suf- Inflammatory lymphocytes 5 62
ficient to perform a diagnosis are shown in Table 6. in dermis <50 cells
Neutrophils in dermis <50 cells 8 100
A final diagnosis combining histological, clinical, laboratory,
Superficial infiltrate 13 100
and response to therapy data allowed a final diagnosis in 69/82
Drug-induced Hydropic degeneration of basal layer 1 25
subjects (84.1%). Particularly, histological examination alone erythroderma Orthokeratosis 4 100
allowed the diagnosis in 19 out of 27 (70.4%) erythroderma (n = 4) Necrotic keratinocytes <50 cells 1 25
patients with a final diagnosis of psoriasis, 17 out of 19 (89.5%) Apoptotic cells <50 cells 1 25
Spongiosis (local) 3 75
subjects with a final diagnosis of eczema/spongiotic dermatitis,
Superficial infiltrate 4 100
seven out of eight (87.5%) patients with mycosis fungoides, and
Eosinophils in dermis <50 cells 1 25
seven out of 11 (63.6%) subjects with drug-induced erythro-
derma.
assess the real value and accuracy of histopathologic examina-
tion in erythroderma patients, trying also to highlight the diverse
Discussion
microscopic features which characterize different erythroderma
Erythroderma is a rare but severe condition that may lead to etiologic subtypes. We showed that blinded histological exami-
severe systemic manifestations and may be life-threatening, nation alone was able to elucidate the underlying cause of ery-
therefore needing prompt diagnosis and treatment.12 Since it throderma in the majority of the cases (61%, n = 50/82),
can be the consequence of several conditions, mainly skin dis- confirming histological examination as a first useful and obli-
orders, drug consumption and, more rarely, some malignancies, gated step on the road to erythroderma diagnosis. In this con-
it is of indisputable importance to know the etiology to facilitate text, our findings also showed that inflammatory skin diseases,
its management.1 However, despite these factors, erythroderma including psoriasis and spongiotic dermatitis/eczema, were
is still poorly studied in literature. Erythroderma patients usually responsible for the majority of erythroderma cases followed by
undergo skin biopsies even if conflicting views about the diag- drug reactions and cutaneous T-cell lymphoma, that being in
nostic value and utility of skin biopsy in the investigation of ery- line with literature.11,13,14 In particular, the diagnosis of psoriasis
throdermic patients still exist. Indeed, nondiagnostic biopsies was made in 23.2% (n = 19/82) of subjects, spongiotic dermati-
may be possible in erythroderma subjects,10 and clinical and tis/eczema in 20.7% (n = 17/82), mycosis fungoides in 8.5%
pathological correlation may be a very complicated task.4 Con- (n = 7/82), and drug eruption in 8.5%, whereas histological
versely, other authors highlighted the importance of histopatho- diagnosis was inconclusive or not matching the final diagnosis
logical examination, even with multiple biopsies over time,6 when available in the remaining 39.1% of cases (n = 32/82). All
suggesting a high mean of diagnostic accuracy (53–66%).9,11 these data supported the fact that histological examination
For all these reasons with the current study, we aimed to allowed us to specify erythroderma diagnosis in up to 61% of

International Journal of Dermatology 2017 ª 2017 The International Society of Dermatology

Megna et al.
cases at a relatively early stage, which enabled us to define
adequate medical tactics as fast as possible and consequently
to improve the immediate and remote results of the treatment.
Identification of histologic patterns of erythroderma is also
important as most of the patients arrive in a serious condition,
with signs of systemic metabolic disturbances and other associ-
ated diseases. Therefore, in such patients early administration
of etiologic specific treatment allows to compensate manifesta-
tions of cardiovascular insufficiency, disorders of electrolyte bal-
ance, considerably improving the prognosis. The specific
histological features which characterized the main subgroup of
erythroderma patients are shown in Tables 3–6. When examin-
ing a skin biopsy from an erythroderma patient, histological fea-
tures such as acanthosis, diffuse parakeratosis, diffuse
hypogranulosis, and the presence of neutrophils both in the epi-
dermis and dermis strongly supported the diagnosis of psoriasis
providing a possibility for the early beginning of specific therapy,
and for avoiding complications of eventual systemic glucocorti-
costeroids treatment. Indeed, acanthosis was observed in 89%
of the psoriasis subjects group, the presence of neutrophils in
both the epidermis and dermis in 84%, while diffuse parakerato-
sis and hypogranulosis were present in 58% of the cases.
On the other hand, the diagnosis of spongiotic dermatitis/
eczema was defined by histological signs such as exocytosis
(100% of the cases), superficial lymphocytic infiltrate (100%),
spongiosis (100%, diffuse 53% and local 47%), irregular acan-
thosis (82%), and the presence of eosinophils <50 cells in the
dermis (76%).
At the same time, the identification of features of interface
dermatitis such as hydropic degeneration of a basal layer of the
epidermis, lymphocytes in a basal layer of the epidermis, colloid
bodies (apoptotic cells), pointed out to the diagnosis of drug-
induced erythroderma, being present in 100%, 100%, 71% of
the cases, respectively. Exocytosis and superficial lymphocytic
infiltrate were observed in all cases as in the spongiotic dermati-
tis/eczema group, therefore appearing as generic and nonspeci-
fic findings for both conditions.
Finally, detection of lymphocytic microabscesses (100%), a
linear arrangement of lymphocytes in the basal layer of the epi-
dermis (100%), and atypical lymphocytes in the epidermis and
the derma (100%) led to the diagnosis of cutaneous T-cell lym-
phoma/mycosis fungoides and its diagnostic and therapeutic
assessment.
Erythroderma remains a condition difficult to study and treat.
Histological examinations may be enough to determine the cor-
rect diagnosis and treatment in the majority of cases. However,
the necessity of multiple skin biopsies and of combining clinico-
ª 2017 The International Society of Dermatology
Histology in erythroderma Report 5
pathologic parameters and response to treatment represent a
mainstay of its diagnostic process. Relying on our findings, we
showed that a correct judgment about the cause of erythro-
derma can be based on objective histopathological criteria in up
to 60% of cases. More studies are needed to try to determine
further histological and/or immunohistochemical markers (e.g.
interleukin 36-Y may possibly identify psoriasis),15 which could
help the clinician in erythroderma etiololgy diagnostic process.
References
1 Khaled A, Sellami A, Fazaa B, et al. Acquired erythroderma in
adults: a clinical and prognostic study. J Eur Acad Dermatol
Venereol 2010; 24: 781–788.
2 Cesar A, Cruz M, Mota A, et al. Erythroderma. A clinical and
etiological study of 103 patients. J Dermatol Case Rep 2016;
10: 1–9.
3 Abrahams I, McCarthy JT, Sanders SL. 101 cases of exfoliative
dermatitis. Arch Dermatol 1987; 63: 96–101.
4 Botella-Estrada R, Sanmarin O, Oliver V, et al. Erythroderma: a
clinicopathological study of 56 cases. Arch Dermatol 1994; 130:
1503.
5 King LE Jr, Dufresne RG Jr, Lovett GL, et al. Erythroderma:
review of 82 cases. South Med J 1986; 79: 1210–1215.
6 Rothe MJ, Bernstein ML, Grant-Kels JM. Life-threating
erythroderma: diagnosing and treating the «red man». Clin
Dermatol 2005; 23: 206–217.
7 Sehgal VN, Srivastava G. Exfoliative dermatitis. A prospective
study of 80 patients. Dermatologica 1986; 173: 278–284.
8 Sentis HJ, Willemze R, Scheffer E. Histopathologic studies in
Sezary syndrome and erythrodermic mycosis fungoides. A
comparison with benign forms of erythroderma. J Am Acad
Dermatol 1986; 15: 1217–1226.
9 Walsh NMG, Prokopetz R, Tron VA, et al. Histopathology in
erythroderma. Review of a series of cases by multiple
observers. J Cutan Pathol 1994; 21: 419.
10 Wilson DC, Jester JD, King LE Jr. Erythroderma and exfoliative
dermatitis. Clin Dermatol 1993; 11: 67–72.
11 Zip C, Murray S, Walsh NMG. The specificity of histopathology
in erythroderma. J Cutan Pathol 1993; 20: 393–398.
12 Egeberg A, Thyssen JP, Gislason GH, et al. Prognosis after
hospitalization for erythroderma. Acta Derm Venereol 2016; 96:
959–962.
13 Vasconcellos C, Domingues PP, Aoki V, et al. Erythroderma:
analysis of 247 cases. Rev Saude Publica 1995; 29: 177–182.
14 Banerjee S, Ghosh S, Mandal RK. A study of correlation
between clinical and histopathological findings of erythroderma
in North Bengal population. Indian J Dermatol 2015; 60: 549–
555.
15 Braegelmann J, D Erme AM, Akmal S, et al. Interleukin-36c (IL-
1F9) identifies psoriasis among patients with erythroderma. Acta
Derm Venereol 2016; 96: 386–387.
International Journal of Dermatology 2017