ARTICLE
Auteur(s) : Turna Ilknur1, Uğur
Pabuççuoglu2, Ciler Akin1, Banu
Lebe2, Ali Tahsin Gunes1
1Department of Dermatology, Dokuz Eylül University
Faculty of Medicine, Dermatoloji Anabilim Dalı, İnciraltı, 35340,
İzmir, Turkey
2Dept of Pathology, Dokuz Eylül University Faculty of
Medicine, İzmir, Turkey
accepté le 7 Decembre 2005
Behçet’s disease (BD) was first described by Hulusi Behçet in 1937
as a triad, characterized by recurrent oral and genital aphthous
ulcerations and uveitis [1]. In the following years, it was
demonstrated that BD could affect almost any organ system [2].
Although BD is a complex multisystem disease, the diagnosis of BD
is still based on clinical criteria because there are no
pathognomonic laboratory findings.The International Study Group
criteria have been widely used for the diagnosis of BD since 1990,
although various criteria were used before. The main criterion of
the International Study Group is recurrent oral ulceration but at
least two of the following clinical manifestations should also be
included: recurrent genital ulceration, typical eye lesions,
typical skin lesions or a positive pathergy test; all of which are
heavily based on the mucocutaneous features of BD [3]. Although
other cutaneous manifestations can occur in BD, the skin lesions
proposed by the International Study Group were confined to erythema
nodosum-like lesions, pseudofolliculitis, papulopustular lesions,
or acneiform nodules. However, pseudofolliculitis, papulopustular
lesions, or acneiform nodules are nonspecific and clinically might
not be differentiated from other diseases with papulopustular
lesions such as folliculitis or acne vulgaris. In addition, the
patients with BD might have other diseases with papulopustular
lesions. Therefore, some authors suggested that, in order to use
pustular lesions as a diagnostic criterion in patients with
suspected BD, histopathological confirmation might be necessary
[4]. In the present study, we aimed to determine whether the
histopathologic evaluation of the papulopustular lesions could be a
useful tool for the diagnosis of BD.
Materials and methods
The protocol and consent forms of this study were approved by the
local ethics committee of Dokuz Eylül University Faculty of
Medicine. Eighteen patients with BD and 16 control patients were
enrolled in the study after their written informed consents were
received. BD was diagnosed by using the International Study Group
criteria. Because the patients with 2 or more symptoms were
regarded as having active Behçet’s disease, those who met at least
one of the other diagnostic criteria in addition to papulopustular
lesions were included in the study. A pathergy test was performed
on Behçet’s disease patients by pricking a 20-gauge disposable
needle in the forearm intradermally. The control group was chosen
among the patients with papulopustular lesions. This group
consisted of eleven patients with bacterial folliculitis and five
patients with acne vulgaris, all of whose diagnoses were made by
clinical examination. None of the patients in the study had
received cytotoxic or immunosuppressive agents such as
corticosteroids, azathioprine or cyclosporine for at least six
months before admission. In addition, patients who had a history of
drug-induced acneiform eruption were excluded from the study. None
of the 18 patients with BD had been taking any systemic treatment
for at least 4 weeks before their inclusion in the study; they
either were just diagnosed or had not taken any treatment.
For the first step, papulopustular lesions over the whole body
were counted. It was considered that taking biopsies from the upper
extremities and trunk would be best, due to cosmetic concerns.
Therefore, papulopustular lesions appropriate for biopsy were
counted and numbered, and then the lesion from which the biopsy
would be taken was identified by drawing lots. In both groups, only
papulopustular lesions with an erythematous base, whether they were
follicular or non follicular, were selected (( figure 1 )). Biopsies from
papulopustular lesions were obtained with a 4 mm punch after
local anesthesia was performed with 2% lidocaine. Punch biopsy
specimens from the skin were immediately transported to the
laboratory in phosphate-buffered saline (PBS) (pH 7.2) at a
low temperature.
For light microscopic examination, a part of the biopsy specimen
was fixed in formalin, embedded in paraffin and stained with
hematoxylin-eosin. Immunofluorescence staining procedure was
performed in the other half of the tissue as previously described
[5]. Histopathological findings were evaluated by two pathologists
(UP and BL); and direct immunofluorescence findings were always
evaluated by a pathologist (BL). The pathologists were blinded to
the patients’ diagnoses. The following histopathologic features
were evaluated; 1. In the epidermis: intraepidermal pustule,
spongiosis, exocytosis, acanthosis, necrotic keratinocytes,
degeneration in the basal layer 2. In the dermis: edema,
lymphohistiocytic or neutrophilic or mixed type perivascular
inflammatory infiltration, lymphohistiocytic or neutrophilic or
mixed type periadnexal inflammatory infiltration, blood vessel
thickening (thickening of the vessel wall caliber), endothelial
swelling (enlargement of the endothelial cell nuclei), fibrin
deposition within the vessel wall, nuclear dust, erythrocyte
extravasation 3. In the follicular epithelium: lymphohistiocytic or
neutrophilic or mixed type follicular and/or perifollicular
inflammatory infiltration; follicular plugging, follicular
epithelial rupture, necrotic keratinocytes or epithelial necrosis.
Then, specimens were grouped into three categories in relation to
their predominant histopathologic findings. The spectrum of
findings within each group is discussed in the following
section.
Pattern I: Vasculitis (lymphocytic or leucocytoclastic) (( figure 2 )).
Pattern II: Folliculitis and/or perifolliculitis (( figure 3 )).
Pattern III: Superficial and/or deep perivascular and/or
interstitial dermatitis
The diagnosis of lymphocytic vasculitis required the presence of
lymphocytic infiltration involving the walls of the small dermal
vessels (comprising superficial and deep dermal vascular plexuses).
Extravasation of fibrin into vessel walls was not a requirement for
this diagnosis [6]. In leukocytoclastic vasculitis infiltration of
vessel walls by neutrophils with associated leukocytoclasis
(nuclear dust), erythrocyte extravasation and fibrin exudation were
prerequisites for the diagnosis [6, 7]. Patterns II and III were
evaluated according to the criteria defined by Ackerman et al.
[7].
Histopathological and immunofluorescence findings were compared
using the chi-square test and P < 0.05 was considered to be
statistically significant.
Results
Eighteen patients with BD (8 women and 10 men) and 16 patients in
the control group (7 women and 9 men) were enrolled in the study.
The mean (± SD) age was 39.2 ± 11.2 years for the patients with BD
and 35.0 ± 10.3 years for the patients in the control group. The BD
and control groups were similar with regard to sex and age. Table
1( Table 1 ) summarizes the clinical
manifestations of the patients with BD during the course of the
disease. The mean number of the papulopustular lesions was higher
in the control group patients than that in the patients with BD
(32.81 ± 30.42, 16.22 ± 19.80, respectively) (P = 0.015).
Specimens of the patients with BD revealed three
histopathological patterns, while those of the control patients
revealed two. The numbers and percentages of histopathological
patterns and positive immunofluorescence deposition in vessels for
both groups are shown in table 2( Table
2 ). While 27.8% of patients with BD showed lymphocytic
vasculitis, the control group did not. No case with
leukocytoclastic vasculitis was noted. The difference between the
two groups was found to be statistically significant (P = 0.046).
The percentage of pattern II (folliculitis and/or perifolliculitis)
was 50.0% in the control patients and 16.7% in the patients with BD
and the difference between the two groups was found to be
statistically significant (P = 0.038). Ten patients in the group
with BD and eight patients in the control group showed pattern III
(superficial and/or deep perivascular and/or interstitial
dermatitis) and no significant difference was found between the two
groups (P > 0.05). Direct immunofluorescence deposition on
vessels of IgM, IgG, IgA, C3, or fibrinogen was found in
4 (three in pattern I, one in pattern II) of 18 patients with BD.
Of 5 cases histopathologically diagnosed as “lymphocytic
vasculitis”, three showed positive immunofluorescence findings in
the vessel walls. In one case immune deposition of IgG, IgM,
C3 and fibrinogen, in one case C3 and
fibrinogen, and in the other case only C3 were observed.
3 out of 16 patients showed positive DIF findings (one in pattern
II, two in pattern III). No significant difference was found
between the two groups (P > 0.05).
Table 1 Demographic data and clinical features of the
patients with BD
|
Patient No.
|
Age/Sex
|
Duration (years)
|
Recurrent oral ulceration
|
Recurrent genital ulceration
|
Erythema nodosum-like lesions
|
Papulo-pustular lesions
|
Pathergy test
|
Eye lesions
|
Other features
|
|
1
|
51/M
|
13
|
+
|
+
|
+
|
+
|
+
|
|
|
|
2
|
52/M
|
2
|
+
|
+
|
|
+
|
+
|
|
|
|
3
|
53/M
|
20
|
+
|
+
|
|
+
|
+
|
+
|
|
|
4
|
32/F
|
18
|
+
|
+
|
|
+
|
+
|
|
Arthritis
|
|
5
|
23/M
|
7
|
+
|
+
|
+
|
+
|
+
|
|
Arthritis
|
|
6
|
39/F
|
5
|
+
|
+
|
|
+
|
+
|
|
|
|
7
|
32/F
|
2
|
+
|
+
|
|
+
|
+
|
|
|
|
8
|
23/F
|
7
|
+
|
+
|
|
+
|
+
|
|
|
|
9
|
40/M
|
2
|
+
|
+
|
+
|
+
|
+
|
+
|
Thrombophlebitis
|
|
10
|
55/F
|
30
|
+
|
|
+
|
+
|
|
+
|
Arthritis
|
|
11
|
46/M
|
20
|
+
|
+
|
+
|
+
|
+
|
+
|
Arthritis, CNS
|
|
12
|
28/F
|
8
|
+
|
+
|
+
|
+
|
|
|
Arthritis
|
|
13
|
32/F
|
14
|
+
|
+
|
+
|
+
|
+
|
+
|
|
|
14
|
31/M
|
20
|
+
|
+
|
|
+
|
+
|
|
|
|
15
|
56/M
|
11
|
+
|
+
|
+
|
+
|
+
|
|
CNS
|
|
16
|
32/M
|
9
|
+
|
+
|
|
+
|
+
|
|
Arthritis
|
|
17
|
49/M
|
30
|
+
|
+
|
|
+
|
|
|
|
|
18
|
32/F
|
1
|
+
|
+
|
|
+
|
+
|
|
|
Table 2 The numbers and percentages of
histopathological patterns and positive immunofluorescence
deposition for both groups
|
Patients in Behçet’s Disease Group (n: 18) n(%)
|
Patients in Control Group (n: 16) n(%)
|
|
Pattern I
|
5 (27.8%)
|
0 (0.0%)
|
|
Pattern II
|
3 (16.7%)
|
8 (50.0%)
|
|
Pattern III
|
10 (55.6%)
|
8 (50.0%)
|
|
DIF
|
4 (22.2%)
|
3 (18.8%)
|
Discussion
Papulopustular lesions in BD arise as papules which become pustules
in the course of 24-48 hrs, and appear as sterile folliculitis or
acne-like lesions on an erythematous base [8]. International Study
Group for BD [9] reported that sensitivity and specificity of
pseudofolliculitis, papulopustular lesions and acneiform nodules
were calculated as 70% and 76% respectively. In another study the
sensitivity and specificity of the papulopustular lesions were
found to be 96% and 11%, respectively. The results of this study
showed that papulopustular lesions appear to be nonspecific for BD.
Authors emphasized that the differences in the clinicians’
experience, in the selection of the patients, or in the methods
used and/or in ethnic origins might account for these differences
in percentages. In this study, the mean number of the
papulopustular lesions was also found to be higher in patients with
BD than that in controls. As a result, it was reported that the
mean number of the lesions is more importance than the frequency of
the lesions in the diagnosis of BD [8]. However, our result does
not support this because the mean number of our patients with BD is
lower than that of the control patients. We believe that the
clinical differentiation of the papulopustular lesions in the
patients with BD is extremely difficult because they have
nonspecific clinical features.
Sometimes the papulopustular lesions may have critical
importance for the diagnosis of patients with BD. Some authors
stated that histopathological confirmation is necessary if
papulopustular lesions are to be used as a diagnostic criterion in
patients with suspected BD. Jorrizzo et al. [4] suggested that only
pustular lesions which reveal vessel-based (not follicle-based)
histopathologic changes should be employed as a diagnostic
criterion for BD.
Therefore, the diagnosis of BD is still based on clinical
criteria because it has no pathognomonic laboratory findings. That
is why studies on biopsies taken from various lesions of BD have
been conducted in order to provide diagnostic histopathological
findings for BD. In a multicentre study, the histopathologic
analysis of cutaneous lesions, including spontaneous pustules,
pathergy test-induced lesions and erythema nodosum-like lesions
taken from 22 patients with BD, showed that perivascular
inflammation is the predominant histopathologic finding [4]. In
another study, it was revealed that 48% of 42 patients with BD with
cutaneous lesions, including erythema nodosum-like lesions,
papulopustular lesions, Sweet’s syndrome-like eruption,
extragenital ulcerations, palpable purpura, and hemorrhagic
blisters, had either a lymphocytic vascultis (31%) or a
leukocytoclastic vascultis (17%) [10].
The positivity of a pathergy test, the occurrence of a sterile
pustular lesion after 24-48 hours at the site of a needle prick to
the skin, is a diagnostic marker for BD [3]. Histopathologic
examination of pathergy lesions taken from patients with BD also
revealed different results. Although some investigators mainly
demonstrated perivascular lymphohistiocytic infiltration [11-14],
Jorizzo et al. [15] demonstrated leucocytoclastic vasculitis or
Sweet’s syndrome-like neutrophilic vascular reaction in a
histopathologic examination of pathergy lesions. In addition, some
authors detected the presence of immunoglobulin, and/or complement,
and/or fibrin deposits on the vessels in pathergy lesions of
patients with BD [15, 16], but some did not [11, 12].
Very few studies exist on the histopathologic features of the
papulopustular lesions, and reports indicate that these lesions
might show either perivascular mixed infiltration with or without
vasculitis or suppurative follicular or perifollicular infiltration
[17, 18]. In a study in which follicle-based papulopustular lesions
were excluded, it was found that 12 of 17 papulopustular lesions of
the patients with BD showed vasculitic changes (11
leucocytocylastic, one lymphocytic). In addition, the presence of
IgM, IgG, C3, and/or fibrin deposits on the vessels in
the papulopustular lesions of 12 patients with BD was detected.
Therefore, the authors stated that immune-complex-mediated
vasculitis was likely to be the main feature of these lesions [17].
In another study, the detailed histopathologic features of
papulopustular lesions in the specimens obtained from 17 patients
with BD and 6 patients with acne vulgaris were studied.
Investigators demonstrated that the papulopustular lesions of acne
vulgaris lesions and BD were characterized by perifolliculitis or
suppurative folliculitis. Therefore, authors stated that the
papulopustular lesions seen in both disorders could not be
distinguished on the basis of clinical and histopathological
findings [18]. Follicle-based lesions were not excluded in this
study because it was considered that the clinical features of these
different patterns could not be differentiated, and besides such a
distinction would be unnecessary, as the way we consider. In
addition, the International Study Group criteria do not include
such a distinction for papulopustular lesions.
The results we obtained were different from the results of both
studies above. We suppose that the differences in the clinical
features and the age of the lesions, whether the disease is severe
or not, whether the disease is in an active period or not, and
whether the patients undergo any treatment or not might lead to
these differences between three studies. In our study, we found
lymphocytic vasculitis in 27.8% of the patients with BD, but not in
any patient in the control group. In addition, while 50% of our
control group patients revealed follicular or perifollicular
involvement, only 16.7% patients with BD revealed the same
histopathological findings. The unexpected direct
immunofluorescence positivity in various histopathologic patterns
was considered to be nonspecific.
The results of our controlled study show that pattern II
(folliculitis and/or perifolliculitis) and pattern III (superficial
and/or deep perivascular and/or interstitial dermatitis) may be
nonspecific findings not representative of papulopustular lesions
in BD. In conclusion, our results show that, of the
histopathological changes encountered in the lesions, only
vasculitic ones (pattern I) can be useful in the employment of
papulopustular lesions as a diagnostic criterion in patients with
suspected BD.
References
1 Behçet H. Über rezidivierende Aphthöse durch ein Virus
verursachte Geschwüre am Mund, am Auge und an den Genitalien.
Dermatol Monatsschr Wochenschr 1937; 105: 1152-7.
2 Mangelsdorf HC, White WL, Jorizzo JL. Behçet’s
disease. Report of twenty-five patients from the United States with
prominent mucocutaneous involvement. J Am Acad Dermatol 1996; 34:
745-50.
3 International Study Group for Behçet’s Disease. Criteria for
diagnosis of Beçet’s disease. Lancet 1990; 335: 1078-80.
4 Jorizzo JL, Abernethy JL, White WL,
Mangelsdorf HC, Zouboulis CC, Sarica R,
Gaffney K, Mat C, Yazici H, Al Lalaan A,
Assad-Khalil SH, Kaneko F, Jorizzo EAF.
Mucocutaneous criteria for the diagnosis of Behçet’s disease: an
analysis of clinicopathologic data from multiple international
centers. J Am Acad Dermatol 1995; 32: 968-76.
5 Dikicioglu E, Meteoglu I, Okyay P,
Culhaci N, Kacar F. The reliability of long-term storage
of direct immunofluorescent staining slides at room temperature. J
Cutan Pathol 2003; 30: 430-6.
6 Weedon D. In: Skin pathology. London: Churchill
Lvingstone, 2002: 242-3.
7 Ackerman AB, Chongchitnant N, Sanchez J,
Guo Y, Bennin B, Reichel M, Randall MB. In:
Histologic Diagnosis of Inflammatory Skin Diseases. Baltimore:
Williams & Wilkins, 1997: 107-44.
8 Alpsoy E, Aktekin M, Er H, Durusoy Ç,
Yılmaz E. A randomized, controlled and blinded study of
papulopustular lesions in Turkish Behçet’s patients. Int J Dermatol
1998; 37: 839-42.
9 International Study Group for Behçet’s Disease. Evaluation of
diagnostic (‘classification’) criteria in Behçet’s disease-towards
internationally agreed criteria. Br J Dermatol 1992; 31:
299-308.
10 Chen K-R, Kawahara Y, Miyakawa S,
Nishikawa T. Cutaneous vasculitis in Behçet’s disease: a
clinical and histopathologic study of 20 patients. J Am Acad
Dermatol 1997; 36: 689-96.
11 Sobel JD, Haim S, Shafrir A, Gellei B.
Cutaneous hyperreactivity in Behçet’s disease. Dermatologica 1973;
146: 350-6.
12 Haim S, Sobel JD, Friedman-Birnbaum R,
Lichtic C. Histological and direct immunofluorescence study of
cutaneous hyperreactivity in Behçet’s disease. Br J Dermatol 1976;
95: 631-6.
13 Gilhar A, Winterstein G, Turani H,
Landau J, Etzioni A. Skin hyperreactivity response
(pathergy) in Behçet’s disease. J Am Acad Dermatol 1989; 21:
547-52.
14 Akmaz Ö, Erel A, Gürer MA. Comparison of
histopathologic and clinical evaluations of pathergy test in
Behçet’s disease. Int J Dermatol 2000; 39: 121-5.
15 Jorizzo JL, Solomon AR, Cavallo T. Behçet’s
syndrome. Immunopathologic and histopathologic assessment of
pathergy lesions is useful in diagnosis and follow-up. Arch Pathol
Lab Med 1985; 109: 747-51.
16 Wechsler J, Wechsler B, Revuz J,
Godeau P. Skin hyperreactivity response (pathergy) in Behçet’s
disease: usefulness of direct immunofluorescence. J Am Acad
Dermatol 1990; 23: 329-30.
17 Alpsoy E, Uzun S, Akman A, Acar MA,
Memişoğlu HR, Başaran E. Histological and
immunofluorescence findings of non-follicular papulopustular
lesions in patients with Behçet’s disease. J Eur Acad Dermatol
Venereol 2003; 17: 521-4.
18 Ergun T, Gürbüz O, Doğusoy G, Mat C,
Yazıcı H. Histopathologic features of the spontaneous pustular
lesions of Behçet’s syndrome. Int J Dermatol 1996; 37: 194-6.
|
Printable version
Version PDF
