Tintack' sign in a patient with cutaneous B-cell lymphoma

© 2001 European Academy of Dermatology and Venereology

357

LETTERS

JEADV

(2001)

15

, 357–377

Blackwell Science, LtdOxford, UKJDVJournal of the European Academy of Dermatology and Venereology0926-9959© 2001 European Academy of Dermatology and Venereology15If known2001223LettersLettersLetters1000Graphicraft Limited, Hong Kong

Localized vitiligo successfully treated with

cream-psoralen + ultraviolet A

To the Editor

Oral psoralen plus ultraviolet A photochemotherapy (PUVA)

is one of the most effective treatment modalities available for

vitiligo. However, in a significant number of subjects adverse

events occur and, therefore, this treatment is not considered

appropriate for localized forms of the disease.

1,2

Recently, cream

preparations containing 8-methoxypsoralen (cream PUVA)

were found to be effective in treating a variety of skin diseases,

including palmoplantar dermatoses

3

and granuloma anulare.

4

Here, we report on a patient with acrofacial vitiligo treated

successfully with cream PUVA.

A 39-year-old Indian woman presented because of extensive

exacerbation of depigmented lesions on her face that were

cosmetically disturbing (fig. 1). Initial treatment with topical

glucocorticosteroids over a period of 8 weeks did not stop the

progression of the disease. Clinical examination revealed well-

circumscribed white macules on her forehead and periocular

region. Laboratory examinations, including routine blood

count and antinuclear antibodies, were all within normal limits.

We undertook treatment with cream PUVA (0.0006% 8-

methoxypsoralen containing water-in-oil emulsion, 30% H

2

O).

3

One hour before UVA irradiation a thin layer of the cream was

applied to the depigmented macules. On the basis of the depig-

mented areas, the patient was classified as skin type I (Fitz-

patrick’s classification). The initial UVA dose was 0.5 J/cm

2

.

Treatments were performed four times per week and the UVA

dose was increased by 0.5 J/cm

2

every third session (maximum

UVA dose: 4 J/cm

2

). UVA irradiation was performed with a local

PUVA device (PUVA 200, Waldmann, Villingen-Schwenningen,

Germany) emitting wavelengths in the range of 320–400 nm

with a peak at 365 nm. Photometrically measured irradiance

was 12 mW/cm

2

at a distance of 20 cm from the lamps (UV-

METER, Waldmann, Villingen-Schwenningen, Germany).

After 8 weeks of treatment, there was remarkable repigmenta-

tion of the formerly depigmented macules, and after 80 treat-

ment sessions almost complete repigmentation had been

achieved (cumulative UVA dose: 254 J/cm

2

, fig. 2). Phototoxic

side-effects, such as itching and blistering, were not observed.

Moreover, repigmentation has been sustained during a follow-

up of 1 year.

Apart from UVB therapy, oral PUVA is considered one of the

most effective non-surgical therapeutic options for generalized

vitiligo.

1,2

As vitiligo is regarded as a T-lymphocyte-mediated

autoimmune disease, the therapeutic effect of PUVA can be

explained by its inhibition of T lymphocytes.

5

In addition,

PUVA stimulates inactive melanocytes in the outer root sheath

to migrate upward to the epidermis. Well known side-effects of

oral PUVA include nausea, liver toxicity, pruritus, premature

cataract formation, and development of skin cancer. In local-

ized vitiligo with less than 20% surface involvement, a topical

PUVA therapy should be favoured when topical corticosteroids

are contraindicated or have proven ineffective.

2,6

Before topical

PUVA is initiated, the subject should be made aware of the

realistic expectations of the therapy. A response rate of approxi-

mately 50–60% was previously reported.

6

Usually, topical PUVA is performed with ethanol or aqua-

phor lotions and carbomer or hydrocellulose gels containing

0.001–0.1% 8-methoxypsoralen. The initial UVA doses are

0.25–0.5 J/cm

2

. However, these preparations may provoke

severe phototoxic reactions within a maximum of 48–110 h.

Hyperpigmentation of non-lesional skin is frequently observed,

due to the dripping of lotions containing 8-methoxypsoralen.

6–9

Furthermore, application of these preparations can cause xerosis.

Lotion and gel PUVA preparations are therefore very difficult to

handle. Phototoxicity due to a PUVA cream preparation disappears

after 4 h and severe phototoxic reactions are uncommon.

3,8fig. 1 Vitiligo of the face before treatment.

fig. 2 Almost complete repigmentation after 80 treatment sessions with

cream-psoralen + ultraviolet A treatment.

JDV291.fm Page 357 Tuesday, September 18, 2001 5:52 PM

358

Letters


JEADV

(2001)

15

, 357–377

Another advantage of cream PUVA is that compared with gel

PUVA, 8-methoxypsoralen containing water-in oil emulsions

remain stable much longer.

10

However, higher UVA doses

are usually required for cream PUVA treatments than other

topical PUVA regimens.

3,4,7,8

Further controlled studies on

cream PUVA are necessary before this can be confirmed as a safe

and effective treatment option for localized vitiligo.

A Kreuter,* T Gambichler, A Avermaete, T Jansen, P Altmeyer,

G von Kobyletzki

Department of Dermatology, Ruhr-University Bochum, Gudrunstrasse

56, D-44791 Bochum, Germany.

*

Corresponding author,

tel. +49 234 509 2551; fax +49 234509 3409;

E-mail: [email protected]

References

1 Kovacs S. Vitiligo.

J Am Acad Dermatol

1998;

38

: 647–666.

2 Njoo MD, Spuls PI, Bos JD

et al

. Nonsurgical repigmentation

therapies in vitiligo.

Meta-Anal Literatur Arch Dermatol

1998;

134

:

1532–1540.

3 Stege H, Berneburg M, Ruzicka T, Krutmann J. Cream-PUVA-

Phototherapy.

Hautarzt

1997;

48

: 89–93.

4 Gambichler T, Menzel S. Cream PUVA in granuloma anulare.

Z Dermatol

1999;

185

: 124–127.

5 Fitzpatrick TB. Mechanisms of phototherapy of vitiligo.

Arch Dermatol

1997;

133

: 1591–1592.

6 Halder RM. Topical PUVA therapy for vitiligo.

Dermatol Nurs

1991;

3

: 178–180.

7 Westerhof W, Nieuweboer-Krobotova L. Treatment of vitiligo with

UV-B radiation vs topical psoralen plus UV-A.

Arch Dermatol

1997;

133

: 1525–1528.

8 Grundmann-Kollmann M, Leiter U, Behrens S

et al.

The time

course of phototoxicity of topical PUVA: 8-methoxypsoralen

cream-PUVA vs. 8-methoxypsoralen gel-PUVA.

Br J Dermatol

1999;

140

: 988–990.

9 De Rie MA, Eendenburg van JP, Versnick AC

et al

. A new

psoralen-containing gel for topical PUVA therapy: development,

and treatment results in patients with palmoplantar and

plaque-type psoriasis, and hyperkeratotic eczema.

Br J Dermatol

1995;

132

: 964–969.

10 Martens-Lobenhoffer J, Rinke M, Losche D, Gollnick H. Long-term

stability of 8-methoxypsoralen in ointments for topical PUVA

therapy (‘Cream-PUVA’).

Skin Pharmacol Appl Skin Physiol

1999;

12

: 266–270.

1542001265LettersLetters10.1046/j.1442-2026.2001.00201.xLetters1000Graphicraft Limited, Hong Kong

Diagnostic signs of cutaneous lymphomas

To the Editor

The skin is most frequently affected by proliferations of the

haematopoietic system in extranodal localizations. Their

prevalence corresponds to that of Hodgkin’s disease.

Skin lymphomas are unique in terms of clinical manifesta-

tions. This allows the diagnosis to be made during the early

stages, and because of the easy accessibility of biopsy material,

and in terms of prognosis and of therapy, which can easily and

directly be targeted to the involved organ, i.e. the skin.

On the other hand, haematopoietic cells also colonize the

skin as the result of immune processes against autoantigens or

foreign antigens, infectious or toxic irritants and many other

stimuli. Differentiation between the early stages of cutaneous

lymphomas (CL), especially cutaneous T-cell lymphomas (CTCL)

and reactive proliferations in the skin may be difficult and has

to be based not only on clinical, histological, phenotypical and

genotypical findings, but primarily based on the combination

of several findings with confirmation by thorough follow-ups.

1

Finally, the diagnosis of CL should not be made without one

single clear-cut clinical, histological, phenotypical or genotypi-

cal criterion or combinations of several criteria. Processes that

simulate CL or precursors, potentially developing into a neopla-

sia only after long-term promotion, have to be differentiated

from malignant lymphoma of T-cell or of B-cell type as long as

clear-cut criteria are missing.

In the following, the most reliable clinical, histological,

phenotypical and genotypical criteria for establishing the

diagnosis of CTCL and of CBCL, respectively, will briefly be

outlined. The fact that there are always exceptions, however,

does not change the basic concept of differentiation of malig-

nant lymphomas from precursors, simulators and ‘pitfalls’

of CL.

Clinical signs

In disease the skin provides a limited number of features,

including eczematous patches, infiltrated plaques, tumours,

with or without involvement of the epidermis, erythroderma,

oedema and pruritus, which at early stages of disease (patches

without infiltration or as one single nodule) may not allow

reactive prelymphomatous or pseudolymphomatous processes

from T- and B-cell lymphomas, respectively, to be differentiated.

In more advanced stages, in which the diagnosis usually can be

confirmed by typical histological changes, the clinical feature

allows differentiation of ‘superficial spreading’ plaque-like

lesions in CTCL from more cutaneous and subcutaneous

nodular lesions of CBCL. Moreover, the distribution of the skin

lesions in CTCL – solitary circumscribed, vs. disseminated or

erythroderma – differentiates between subtypes of CTCL,

such as pagetoid reticulosis, mycosis fungoides and Sézary

syndrome. In CBCL, however, subclassification on the basis of

the clinical feature is usually not possible.

2,3

Histomorphology

Apart from premycotic stages with sparce non-specific infiltrates

and B-cell pseudolymphomatous reactions, histomorphology


Letters

359


JEADV

(2001)

15

, 357–377

and cytomorphology provides the gold standard diagnostic tool

for the differentiation of CTCL from CBCL and their subtypes.

4

The growth pattern may be diffuse as seen in most CTCL or

nodular as in CBCL. A network or Indian file-like pattern is

suggestive for a myeloproliferative infiltrate in the skin.

Phenotyping

In early prelymphomatous stages of CTCL there are no specific

phenotypical features. In more advanced stages, loss of antigens

(e.g. CD7), staining for markers of proliferation (Ki 67, PCNA)

or demonstration of activation markers (CD30) may be helpful.

Large cell lymphomas, where the cytomorphology usually does

not allow the differentiation between T- and B-cell lineages,

staining for T-cell (CD3, CD43) and for B-cell markers (CD20,

CD78a) are very helpful. Specific diagnostic information may

be determined by antibodies of bcl-2 (B-cell lymphoma with

secondary skin involvement), CD15 (Hodgkin’s disease) and

CD5 (specific skin infiltrates in chronic lymphatic leukaemia).

To identify rare, but prognostically and therapeutically

important forms of cytotoxic lymphomas, staining for CD8,

CD56 and cytotoxic molecules (TIA-1, granzyme B, Perforine)

is essential.

1

Genotyping

Demonstration of the clonal rearrangement of the T-cell

receptor (TCR) gene or the immunoglobulin gene is the most

specific tool for the differentiation of reactive lympho-

proliferative from oligo-or monoclonal lymphoproliferative

skin infiltrates, especially in the early stages of CTCL or in

pseudolymphomatous-like conditions of CBCL. Even though

monoclonality by itself does not prove malignancy, in con-

junction with appropriate clinical, histological and pheno-

typical features, gene rearrangement is the most reliable tool

for the differentiation of CLs from non-lymphomatous con-

ditions and for the differentiation of CTCL from CBCL. For

subtyping CTCL and CBCL, respectively, genotyping is not

helpful except in rare cases in which the demonstration of

germline configurations of TCR genes is an important feature

in diagnosis [e.g. true natural killer (NK) cell lymphoma].

Other diagnostic signs

Many other clinical and laboratory markers may provide

additional diagnostic confirmation, such as the leukaemic

blood picture (Sézary syndrome), eosinophilia and elevated

levels of immunoglobulins (proliferation of Th2 cells) in

CTCL,

5

pruritus, immunodeficiency [impaired NK cell

function in CTCL], demonstration of viral infection (Epstein–

Barr virus in NK cell lymphoma), human T-cell leukaemia

virus in adult T-cell lymphoma.

Conclusions

In summary, the diagnosis of CL, and the differentiation of

malignant lymphoproliferative skin infiltrates from pre-

lymphomatous or pseudo-lymphomatous conditions has to be

based on the synoptic evaluation of clinical and histological

features in most cases (65%) and on additional phenotypical

and genotypical findings for confirmation and diagnoses in rare

and unusual cases (25%) (Table 1). In about 10% of cases, long-

term follow-up and repeating diagnostic procedures is needed

for more accurate diagnoses.

G Burg,* W Kempf, R Dummer

Clinic of Dermatology, University Hospital of Zürich, Gloriastr. 31,

CH–8091 Zürich/Switzerland.

*


tel. +41 1255 25 50; fax +41 1255 44 03; E-mail: [email protected]

References

1 Kempf W, Dummer R

et al

. Approach to lymphoproliferative

infiltrates of the skin. The difficult lesions.

Am J Clin Pathol

1999;

111

(Suppl. 1): S84–S93.

2 Burg G, Braun-Falco O.

Cutaneous Lymphomas, Pseudolymphomas

and Related Disorders

. Springer-Verlag, Berlin, 1983.

3 Burg G, Kempf W

et al.

Cutaneous lymphomas.

Current Problems

1997;

9

(5): 137–204.

4 Burg G, Dummer R

et al.

Pathology of cutaneous T-cell lymphoma.

Hematol Oncol Clin North Am

1995;

9

(5): 961–995.

5 Dummer R, Heald PW, Nestle FO

et al.

Sézary syndrome T-cell

clones display T-helper 2 cytokines and express the accessory

Clinical signs Histomorphology Phenotyping Genotyping Other signsa

CLb vs. non-lymphoma (+) > + (+) > + (+) > + + > (+) (+)

CTCL vs. CBCL + + + + +

Subtypes CTCL + + (+) − −Subtypes CBCL − + (+) − −

CTCL, cutaneous T-cell lymphoma; CBCL, cutaneous B-cell lymphoma; >, rather thanaEosinophilia, leukaemic blood picture, pruritus, EBV+, HTLV+, Immunodeficiency, etcbCL, cutaneous lymphoma, especially early stages from prelymphomatous and pseudolymphomatous conditions

Table 1 Differential diagnostic impact (good +,

moderate (+), poor –) of diagnostic signs in

cutaneous lymphomas


360

Letters


JEADV

(2001)

15

, 357–377

factor-1 (interferon-gamma receptor beta-chain).

Blood

1996;

88

(4):

1383–1389.

6 Cerroni L, Kerl H, Gatter K.

An Illustrated Guide to Skin Lymphoma

.

Blackwell Science Ltd, Oxford, 1998.

15If known2001241LettersLettersLetters1000Graphicraft Limited, Hong Kong

‘Tin-tack’ sign in a patient with cutaneous

B-cell lymphoma

To the Editor

The ‘tin-tack’ sign is the manifestation of spiny projections from

the undersurface of a scale or a crust when it is mechanically

detached from the skin. Although this sign was first described as

a clinical feature of discoid lupus erythematosus (DLE),

1

it has

also been reported in localized pemphigus foliaceus,

2

drug-

induced lichen planus,

3

seborrhoeic dermatitis,

4

postirradiation

scalp skin scales

5

and cutaneous leishmaniasis (CL).

6,7

We

report a case of cutaneous B-cell lymphoma (CBCL) with a

positive ‘tin-tack’ sign.

A 59-year-old man presented to our clinic for a progressive

mass on the right parieto-occipital scalp region of 4 months

duration. On examination we saw a tumoral lesion character-

ized by indurated plaques and nodules varying from light red to

plum colour. It was asymmetric in shape, 7

×

9 cm in dimen-

sions, firm, non-tender and fixed to subcutaneous tissues.

Although the larger portions of the mass had a smooth

surface, one of the nodules had a crust on top (fig. 1). Removal

of this crust revealed spiny processes attached to the under-

surface (fig. 2). Differential diagnoses of primary cutaneous

lymphoma, metastatic skin disease and pseudolymphoma

were considered.

Histological examination revealed hyperkeratosis and

multiple follicular plugs associated with large follicular open-

ings. There was a diffuse lymphomatous infiltrate composed of

round, large cells with hyperchromatic nuclei and prominent

nucleoli. The cells had abundant, eosinophilic cytoplasm. There

were marked mitotic activity and apoptotic changes in the

neoplastic cells. The tumoral infiltrate filled the dermis entirely

and extended to subcutaneous tissues. Histochemical examina-

tion included staining these cells with anti-CD20 but not with

anti-UCHL-1, and the tumour was diagnosed as CBCL with the

following histological and immunohistochemical findings.

Scans of extracutaneous systems for metastatic involvement

evidenced no significant findings except computed tomography

scan evidence of multiple cervical lymph node enlargements. In

about 5 months, the tumoral lesion regressed gradually to the

skin level with chemotherapy.

The clinical importance of ‘tin-tack’ or ‘carpet tacks’ has been

well known for years, particularly in DLE

1

and CL,

6,7

but rarely

in other skin affections, such as localized pemphigus foliaceus,

2

drug-induced lichen planus,

3

seborrhoeic dermatitis

4

and

postirradiation scalp skin scales.

5

However, we found no

reports in the literature and standard textbooks on the existence

of this sign in cutaneous lymphoma. Accumulation of keratotic

plugs in dilated follicles, possibly due to retention hyper-

keratosis, demonstrates itself clinically as spiny projections

under the surface of the detached crust, i.e. positive ‘tin-tack’

sign.

In clinical practice, this sign is not unfamiliar to us because

CL is endemic in our area.

8

Therefore, we have been aware of

the clinical importance of this frequent sign in CL since it was

first described by Behçet.

6,7

We generally lift crusts with

tweezers from all nodular or noduloulcerative crusted lesions

to show any positive ‘tin-tack’ sign indicating a possible case of

CL. (We have been investigating the importance of ‘tin-tack’

signs in the clinical diagnosis of CL in a large case series to be

reported soon). Hence, detection of the positive ‘tin-tack’ sign

in this previously reported case of CBCL was a coincidental

finding.

We considered that a positive ‘tin-tack’ sign on crusted

lesions localized in areas with multiple large follicular ori-

fices, such as the scalp and face, could be a clue for clinical

diagnosis in certain skin diseases, as mentioned in previous

reports,

1–7

including CBCL; however, it is not a specific diag-

nostic feature.fig. 1 Tumoral lesion presenting indurated plaques and nodules on the

scalp. Note the crust on top of the biggest nodule.

fig. 2 Horny projections attached to the under-surface of the removed crust.


Letters

361


JEADV

(2001)

15

, 357–377

M Baba,† S Uzun,†* MA Acar,† D Gümürdülü,‡ HR Memisoglu†

Departments of

†

Dermatology and

‡

Pathology, Faculty of Medicine,

Cukurova University, 01330 Adana, Turkey.

*


tel. +90 322 338 64 26; fax +90 322338 66 56;


References

1 Rowell NR, Goodfield MJD. The ‘connective tissue diseases’.

In: Champion RH, Burton JL, Burns DA, Breathnach SM, editors.

Textbook of Dermatology

, 6nd edn. Blackwell Scientific Publications,

Oxford, 1998.

2 Paramsothy Y, Lawrence CM. ‘Tin-tack’ sign in localized pemphigus

foliaceus.

Br J Dermatol

1987;

116

: 127–129.

3 Cox NH, Tapson JS, Farr PM. Lichen planus associated with

captopril: a further disorder demonstrating the ‘tin-tack’ sign.

Br J Dermatol

1989;

120

: 319–321.

4 Cowley NC, Lawrence CM. ‘Tin-tack’ sign in seborrhoeic dermatitis.

Br J Dermatol

1991;

124

: 393–394.

5 Thomas RJ, Smith NP, Spittle MF. The ‘tin-tack’ sign in

post-irradiation scalp skin scales.

Br J Dermatol

1992;

126

: 90.

6 Behçet H. Wright çıbanları seririyatında ihmal edilmi3 iki mühim

nokta.

Türkderm

1934;

1

: 16–22.

7 Behçet H. Wright çıbanları hakkında (I) (II).

Türkderm

1935;

2

:

825–836.

8 Uzun S, Uslular C, Yucel A

et al.

Cutaneous leishmaniasis:

evaluation of 3074 cases in the Cukurova region of Turkey.

Br J Dermatol

1999;

140

: 347–350.

15If known2001272LettersLettersLetters1000Graphicraft Limited, Hong Kong

Glucocorticoid action on skin collagen: overview on clinical significance and

consequences

To the Editor

Glucocorticoids (GCs) have been used in the treatment of

inflammatory diseases, such as skin and pulmonary diseases for

over 50 years. In dermatology, hydrocortisone was one of the

first to achieve marked popularity, as its topical use was rapidly

able to alleviate inflammation in eczema and other skin

diseases. Later on, other synthetic compounds were introduced

into clinical practice, such as triamcinolone, betamethasone

and numerous others. Many of these are still widely used.

Topical steroids constitute a substantial portion of dermato-

logical compounds. Another important indication for GCs is

pulmonary asthma. Large clinical trials have shown that inhaled

GCs can efficiently alleviate inflammation in pulmonary airways

and reduce the symptoms of asthma.

During the recent years, new derivatives of GCs have been

developed, such as mometasone furoate (MMF) and methyl-

prednisolone aceponate (MA), and it has been suggested that

these compounds could have fewer side-effects than the older

agents. The ultimate goal in developing new GCs has been to

obtain compounds that have potent anti-inflammatory action

without side-effects. One of the most important side-effects of

GCs is skin atrophy. As collagen is the major component of the

skin, the action of GCs on collagen synthesis in the light of

recent studies will be discussed.

GCs bind to specific receptors (GR).

1

The binding affinities

of different steroids are different, and the affinity is higher with

synthetic potent steroids than with cortisol. The binding affin-

ity roughly correlates with the anti-inflammatory potency of

the steroid. After a GC binds to a receptor the steroid-receptor

complex is able to bind to specific sequences in DNA (response

element) in a dimer form to induce transcription. GC can also

bind to negative GRE sequences. By binding to a negative GRE,

GC inhibits transcription. Interestingly, the promoter of type I

collagen contain no GRE or nGRE, and the mechanisms

described above do not mediate the inhibitory action of GC on

collagen expression.

2

It is, however, possible that GR binds to

some other activating factors that are needed in collagen tran-

scription. GCs can also modulate the level of functional mRNAs

of collagen by increasing the degradation of mRNAs.

Most of the skin collagen consist of type I collagen, which is

composed of two different collagen chains,

α

1(I) and

α

2(I),

and has a structure [

α

1(I)]2

α

2[I]. Another important collagen

is type III collagen, which is composed of three identical

α

chains, having a structure

α

1(III)3.

The level of corresponding mRNA correlates with the

actual collagen synthesis. Accordingly, it has been shown in

skin fibroblast cultures and human skin biopsy specimens

that GCs are able to decrease type I and III mRNAs. Recently,

a newly developed quantitative polymerase chain reaction

method employing synthetic collagen RNA as an internal

standard was used in a clinical trial. Three days’ treatment with

betamethasone-17-valerate caused a > 70% decrease in type I

collagen mRNA (fig. 1).

3

The decrease was similar to the

decrease in collagen propeptides in the same subjects. This

indicates that the suppression of procollagen synthesis by GCs

is almost solely due to a decrease in corresponding functional

mRNA. During collagen synthesis, procollagen is first excreted

into the extracellular space and then converted into collagen,

and N- and C-propeptides are liberated at the same time. These

propeptides can be assayed and they reflect the actual synthesis

of collagen. We have used the suction blister technique, in which

suction blisters are produced with disposable devices (Venti-

press®, Lappeenranta, Finland) and the suction blister fluids are

collected and used for propeptide and protein analysis.

4

When

clobetasol-17-propionate was applied on to the abdominal skin

for 1 day, the collagen propeptides declined by about 60–70%.

4

After 2 and 4 days the reduction was even more pronounced.

This indicates that clobetasol propionate is able to almost

completely block the synthesis of new collagen in the skin.

Accordingly, clobetasol propionate has a marked risk to

induce skin atrophy, even over a relatively short period. A

comparison of various steroids, such as hydrocortisone


362

Letters


JEADV

(2001)

15

, 357–377

(HC), hydrocortisone-17-butyrate (HC-17-B) and betametha-

sone valerate (BM), showed that all of them reduced collagen

propeptides in the skin after 7 days of use, the decline being

about 30–40% after HC, 40–50% after HC-17-B and 50–60%

after BM (fig. 2).

5

The recovery of skin collagen synthesis after topical steroid

treatment is clinically important. The more rapidly the skin

collagen synthesis is recovered, the shorter the steroid-free

intervals can be without harmful side-effects. When BM was

applied for 3 days after a 14-day drug-free interval, collagen

synthesis recovered up to 50% of the corresponding collagen

synthesis in non-treated skin.

6

For HC and other steroids, the

duration of down-regulation is not known.

Newer topical steroids, such as MMF and MA, have been

claimed to be less atrophogenic when compared with other

steroids of equal clinical potency. After 1 week of treatment,

MMF decreased collagen propeptides to the same extent as BM,

which shows that it may have a similar risk to induce skin

atrophy.

7

Furthermore, MA was almost as potent as MMF in

reducing propeptides.

8

As a conclusion, the atrophogenic and

anti-inflammatory effects go hand in hand and there is cur-

rently no steroid that would have potent anti-inflammatory

action without affecting collagen synthesis.

P Nuutinen, P Autio, T Hurskainen,† A Oikarinen†*

Departments of Dermatology, Central Military Hospital, Helsinki and,

†

University of Oulu, Oulu, FIN-90220 Finland.

*


fax +358 8 3153135

References

1 Evans RM. The steroid and thyroid hormone receptor superfamily.

Science

1988;

240

: 889–895.

2 Chu M-L, de Wet WJ, Bernard M

et al.

Fine structural analysis of

the human pro-alpha-1(I) collagen gene: promoter structure, Alu I

repeats, and polymorphic transcripts.

J Biol Chem

1985;

260

: 2315–

2330.

3 Oikarinen A, Haapasaari K-M, Sutinen M et al. Molecular basis

of glucocorticoid-induced skin atrophy: topical glucocorticoid

apparently decreases both synthesis and corresponding mRNA

level in human skin in vivo. Br J Dermatol 1998; 139: 1106–1110.

4 Oikarinen A, Autio P, Kiistala U et al. A new method to measure type

I and III collagen synthesis in human skin in vivo: demonstration of

decreased collagen synthesis after topical glucocorticoid treatment.

J Invest Dermatol 1992; 98: 220–225.

5 Haapasaari K-M, Risteli J, Koivukangas V et al. Comparison of

the effect of hydrocortisone, hydrocortisone-17-butyrate and

betamethasone on collagen synthesis in human skin in vivo.

Acta Derm Venereol 1995; 75: 269–271.

6 Haapasaari K-M, Risteli J, Oikarinen A. Recovery of human skin

collagen synthesis after short-term corticosteroid treatment and

comparison between young and old subjects. Br J Dermatol 1996;

135: 65–69.

7 Koivukangas V, Karvonen J, Risteli J et al. Topical mometasone

furoate and betamethasone-17-valerate decrease similarly

collagen synthesis in human skin in vivo. Br J Dermatol 1995;

132: 66–68.

8 Haapasaari K-M, Risteli J, Karvonen J et al. Effect of hydrocortisone,

methylprednisolone aceponate and mometasone furoate on collagen

synthesis in human skin in vivo. Skin Pharmacol 1997; 10: 261–264.15If known2001282LettersLettersLetters1000Graphicraft Limited, Hong Kong

Localized monilethrix with improvement after treatment of iron deficiency anaemia

To the Editor

A 29-year-old woman presented to the Department of

Dermatology, Adnan Menderes University (Aydın, Turkey)

fig. 1 Topical betamethasone-17-valerate cream used for 3 days reduced

markedly the level of type I collagen mRNA in human skin when assayed by

quantitative polymerase chain reaction technique. Modified from Oikarinen

et al.3

fig. 2 Hydrocortisone (HC), hydrocortisone-17-butyrate (HB) and betameth-

asone valerate (BM) used for 7 days reduced the levels of type I procollagen

propeptides (PINP) in the suction blister fluid in human skin. Modified from

Haapasaari et al.5


Letters 363

© 2001 European Academy of Dermatology and Venereology JEADV (2001) 15, 357–377

with a 1-month history of alopecia and fragile hair on the scalp.

She had no history of contact with chemicals (hair dyes,

bleaching or permanent agents) or application of any other

traumatic procedures. Neither drug intake before the onset of

the disease and menstrual abnormalities nor a positive family

history could be found. She had experienced the same

complaints in the same localization twice before, 3 and 6 years

ago, which had improved spontaneously within 3 months, with

therapy for iron deficiency anaemia caused by rectal bleeding.

Physical examination revealed diffuse alopecia with dry and

brittle hairs with keratotic follicular plugs on otherwise normal

skin of the frontoparietal area (fig. 1). The woman’s skin was

pale, her tongue was smooth, and she had multiple external and

internal haemorrhoids. Hair samples were taken by cutting near

the proximal end from both the apparently normal occipital

area and the alopecic region. Microscopic examination of hairs

taken from the alopecic region showed fusiform nodes with a

regular periodicity of 0.5–0.7 mm separated by narrow segments

on a 1.2-cm fragment proximally (fig. 2), while the distal portions

were normal. Hairs from the occipital area were completely

normal.

Significant laboratory values included 9.5 g/dL haemoglobin

and 15 µg/dL serum ferritin (normal value 50–170 µg/dL).

Peripheral blood smear revealed microcytic-hypochromic

anaemia. White blood cells and platelets, erythrocyte sedimen-

tation rate, haemoglobin electrophoresis, sex hormones, renal,

liver and thyroid function test results were within normal limits.

After administration of 1125 mg of ferro glykokoll sulphate

daily for 4 months, the woman’s haemoglobin value increased

to 11.5 g/dL. On dermatological examination, a 2-cm proximal

segment of hair appeared normal and there were no signs of

follicular keratosis. The fragment with fusiform nodes had

proceeded away as a result of normal growth.

Structural defects of the hair shaft cause significant cosmetic

disability. Monilethrix is a rare defect of the hair shaft resulting

in hair fragility and dystrophic alopecia. Microscopically, hairs

show elliptical nodes with a regular periodicity of 0.7–1 mm

separated by constricted segments, where the hair usually fractures.

Most cases of monilethrix are of autosomal dominant inheritance,

although autosomal recessive cases have been reported.1 Onset

may be delayed until subjects are in their teens, and the loss may

be localized or diffuse.2–4 Bentley-Phillips and Bayles5 described

a syndrome that they termed pseudomonilethrix, now believed to

be artefactual.6 Microscopic changes can be produced in normal

hairs by trauma from tweezers or forceps, and also by compressing

overlapping hairs between two glass slides. The same condition

can occur if the hair is pulled out slowly while taking samples

for examination. Our subject did not describe any traumatic

procedures and the samples were taken by cutting to avoid artefacts.

The hairs were examined separately, to avoid indentation in one

shaft caused by another overlying hair.

Differential diagnosis should include androgenetic alopecia

and telogen effluvium, where iron deficiency anaemia is one of

the best known causes. However, there are no structural hair

abnormalities in these two forms of alopecia. Although at first

glance our subject appeared to have andropenetic alopecia, we

found no signs of either hormonal influence or positive family

history. The interesting point was the repetition of the alopecia

three times, always associated with anaemia and with improve-

ment with iron therapy. This condition could not be termed as

classical monilethrix, and there was not sufficient evidence to

classify it as a different form of pseudomonilethrix. To the best

of our knowledge no similar case has been reported in the literature.

Hence, we believe that this structural hair defect was in some

way associated with iron deficiency anaemia, and that possibly the

anaemia itself might have triggered a pre-existing but silent

monilethrix.

G Can Karaman,* N Sendur, H Basar, E Bozkurt Savk

Department of Dermatology, Adnan Menderes University,

Faculty of Medicine, Aydın, Turkey. *Corresponding author, Adnan

Menderes Üniversitesi, Tıp Fakültesi, Dermatoloji Anabilim Dali,

Aydın, Turkey, tel. +90 2562124078; fax +90 2562120146;


fig. 1 Alopecia with keratotic follicular plugs on the frontoparietal area.

fig. 2 A 1.2-cm fragment showing fusiform nodes with a normal appearing

distal portion (original magnification × 10).


364 Letters


References1 Olsen EA. Hair disorders. In: Freedberg IM, Eisen AZ, Wolff K et al.

editors. Dermatology in General Medicine, 5th edn. McGraw-Hill,

Inc., New York, 1999: 729–751.

2 Gebhardt M, Fischer T, Calussen U, Wollina U, Elsner P.

Monilethrix: improvement by hormonal influences? Pediatr

Dermatol 1999; 16: 297–300.

3 Whiting DA. Structural abnormalities of the hair shaft. J Am Acad

Dermatol 1987; 16: 1–25.

4 De Berker DA, Ferguson DJ, Dawber RP. Monilethrix:

a clinicopathological illustration of a cortical defect. Br J Dermatol

1993; 128: 327–331.

5 Bentley-Phillips B, Bayles MAH. Pseudomonilethrix. Br J Dermatol

1975; 92: 113–120.

6 Zitelli JA. Pseudomonilethrix: an artefact. Arch Dermatol 1986; 122:

688–692.15If known2001280LettersLettersLetters1000Graphicraft Limited, Hong Kong

Lymphoedema associated with Kaposi’s sarcoma

To the Editor

Kaposi’s sarcoma (KS) is one of several neoplasms identified as

part of the spectrum of acquired immunodeficiency syndrome

(AIDS). We present a case of severe lymphoedema of the lower

extremities secondary to disseminated KS in an AIDS patient,

and discuss recent studies concerning the aetiology of KS and its

predilection for involving the lymphatics.

A 41-year-old white male presented with end-stage AIDS

manifested by the following: a CD4 count less than 10, disseminated

KS, cytomegalovirus retinitis, disseminated Mycobacterium

avium intracellulare infection, recurrent herpes simplex infection,

disseminated strongyloides infection, candidal oesophagitis,

pan-sinusitis and cutaneous mucormycosis infection. He was

hospitalized with nephropathy associated anasarca, complain-

ing of extremity pain and discoloration suggestive of vascular

compromise.

On examination, he demonstrated multiple violaceous

plaques and nodules over the entire cutaneous surface, with

severe bilateral weeping and bullous lesions of his lower

extremities.

The patient’s anasarca was felt to be secondary to a protein-losing

nephropathy, most likely due to AIDS or amphotericin-related

damage.

Biopsy from the oedematous regions of the leg revealed massive

oedema and vesiculation within the epidermis, overlying

extensive involvement of the dermis and subcutaneous tissues

with KS. Lymphatic and venous infiltration by tumour cells

was noted in these areas. Perivascular KS lesions were seen sur-

rounding areas of haemorrhagic and coagulative necrosis of

the subcutaneous connective tissues.

In deeper tissues, the KS lesions became more numerous as

the extent of the necrosis increased, and assumed a striking

perilymphatic and perivenous distribution. Together these

findings suggest that lymphatic and venous outflow obstruc-

tion may have caused the lower extremity ischaemia, necrosis

and, in part, the oedema.

The risk of acquiring KS among HIV-positive individuals

varies depending on the mode of infection, with an

extremely high proportion of cases reported among homo-

sexual men. This has led some to suggest that a sexually

transmitted cofactor may play a part in the development of

AIDS-related KS. The recently described KS-associated herpes-

virus (human herpesvirus-8) may be the agent responsible

for the higher rates of KS among patients infected through

sexual contact.1

Lymph node involvement in AIDS-related KS occurs in 50%

of cases. Although lymphadenopathy is the most frequent

manifestation, oedema is estimated to occur in 20% of patients

with KS.2 In addition, there has been at least one case of KS

reported in which lymphatic involvement was the only mani-

festation of the disease.3

The pathophysiology of AIDS-related KS inducing lym-

phoedema is not fully understood, and may be due to a number

of processes. Obstruction due to the external compression

of lymphatics by cutaneous lesions is probably responsible for

some oedema in the extremities. Another proposed mechanism

involves the mixing of blood and lymph secondary to lympha-

ticovenous anastomoses.

A growing amount of research has focused on identifying the

cell of origin in KS. Various theories explaining the cell type

from which the spindle cells characteristic of KS lesions are

derived, have been proposed in the past. Mesenchymal stem

cells and blood vessel endothelium were often cited as possible

sources. However, recent studies using the techniques of immu-

nohistochemistry and enzyme histochemistry have revealed

that the lymphatic endothelium may be the origin of KS.4

Beckstead et al.5 have shown that spindle cells of KS are pheno-

typically similar in terms of several cell markers to normal lym-

phatic endothelial cells, and share very few similarities with

blood vessel endothelium. Using electron microscopy, McNutt

et al.6 found numerous ultrastructural similarities between dermal

lymphatics and early KS lesions.

These studies, indicating lymphatic vessels as the origin of

KS, may help to explain the high frequency of lymph node

involvement by cutaneous KS and the rare case of lymphatic KS

without cutaneous lesions. With regard to the case presented,

our patient had diffuse cutaneous KS involvement with severe

oedema and bullous gangrene of the lower extremities.

At autopsy there was no evidence of significant arterial disease

or vascular compromise. Rather, the perilymphatic and

perivenous distribution of the KS lesions suggest a possible role

for an outflow obstruction as the source of this patient’s vascu-

lar complications.


Letters 365


ES Atillasoy, A Santoro, JM Weinberg*

Department of Dermatology, St Luke’s-Roosevelt Hospital Center,

1090 Amsterdam Avenue, Suite 11D, New York, NY 10025, USA.

*Corresponding author, tel. (212)523 4366; fax (212)523 5027;


References1 Cesarman E, Moore PS, Rao PH et al. In vitro establishment

and characterization of two acquired immunodeficiency

syndrome-related lymphoma cell lines (BC-1 and BC-2) containing

Kaposi’s sarcoma-associated herpesvirus-like (KSHV) DNA

sequences. Blood 1995; 86: 2708–2714.

2 Harrison M, Tomlinson D, Stewart S. Periorbital edema in Kaposi’s

sarcoma (letter). N Engl J Med 1995; 333: 799–800.

3 Frans E, Blockmans D, Peetermans W et al. Kaposi’s sarcoma

presenting as generalized lymphedema. Acta Clin Belg 1994; 49:

19–22.

4 Dorfman RF. Kaposi’s sarcoma: evidence supporting its origin from

the lymphatic system. Lymphology 1988; 21: 45–52.

5 Beckstead JH, Wood GS, Fletcher V. Evidence for the origin of

Kaposi’s sarcoma from lymphatic endothelium. Am J Pathol 1985;

119: 294–300.

6 McNutt NS, Fletcher V, Conant MA. Early lesions of Kaposi’s

sarcoma in homosexual men: an ultrastructural comparison with

other vascular proliferations in the skin. Am J Pathol 1983; 111:

62–77.15If known2001276LettersLettersLetters1000Graphicraft Limited, Hong Kong

Raynaud symptoms as principal signs in a case of Sneddon’s syndrome

To the Editor

Sneddon’s syndrome is characterized by generalized livedo

reticularis associated with cerebrovascular disorders. The

syndrome is uncommon and is often diagnosed very late,

usually only on the basis of examination of biopsy specimens.

A 46-year-old female presented to us with the principal signs

of Raynaud’s phenomenon involving her hands and feet associ-

ated with noticeable flattening of the sphygmo-oscillograph

curves of the toes and various neurological symptoms. She

had complained of Raynaud symptoms since mid-1997, with

bilateral numbness of her fingers and toes and for a short time

also superficial necrosis of the finger pads. The distal phalanxes

of all her fingers and toes were noticeably livid, there were

isolated petechiae on her finger pads and toe pads, and reddish-

blue marks on the skin of her lower legs (fig. 1).

The diagnosis of livedo reticularis was confirmed by histolog-

ical examination. In the area of transition from the dermis to

the subcutis many small arteries showed subendothelial cellular

proliferation, usually originating in the smooth muscle cells

and forming intraluminal, bridge-like structures in the cell

walls. In immunohistochemical tests the cell-wall proliferation

was labelled by anti-actin (HHF 35), an indication that the pro-

liferation originated in the vessel walls (fig. 2). We note that

such tests require biopsy material taken from unaffected skin

in the centre of a focus of livedo reticularis; the samples must

be 1–2 cm in diameter and the serial sections must be adequate

in size.1,2

An oscillogram of the fingers and toes showed flattened

oscillations at D1 and D3 on the left hand and D1 of the right

foot. The values improved with warming, indicating the presence

of vasospasms. Colour duplex examination of the hands showed

fig. 1 Livedo reticularis on the thigh of a 46-year-old woman with Sneddon’s

syndrome.

fig. 2 Histological picture of a skin specimen from a woman with Sneddon’s

syndrome. Note the subendothelial cellular proliferation, that usually originates

in smooth muscle cells, with development of intraluminal, bridge-like structures.


366 Letters


no signs of obliterating endangiitis. Neither the Doppler values

or curves nor nailfold capillaroscopy revealed any other abnor-

malities. Thrombophilia screening tests and all other laboratory

tests gave normal findings. Immunopathological tests evidenced

slightly elevated levels of anticardiolipin antibodies.

The woman also presented cerebrovascular disorders and had

experienced symptomatic convulsions since 1988. Magnetic

resonance imaging (MRI) showed distinct supracranial and

intracranial vascular processes resulting from supratentorial

and infratentorial haemorrhages as well as multiple intra-

cranial stenoses, such as partial stroke involving the left

superior cerebellar artery. MRI is the method of choice for

establishing the extent of cerebrovascular damage,1 although

brain imaging studies can be used to detect cerebral circulatory

disorders at an early stage before irreversible ischaemic damage

takes place.3,4

In Sneddon’s syndrome the spectrum of possible neurological

symptoms ranges from subtle, non-specific complaints, such as

dizziness, headaches, partial short- and long-term memory loss

and loss of fine motor control, to cerebral infarction with hemi-

plegia or hemiparesis.5 Also other organs (heart, kidneys) may

be affected, as well as the peripheral nerves and fundus of the

eyes, but the clinical course is often asymptomatic.6

Because the aetiology is unknown there is to date no satisfactory

therapy for this chronic disease, and controlled medication

studies have yet to be performed on livedo racemosa. There

have been reports of positive results with therapy using

anticoagulants, thrombocyte aggregation inhibitors, oral pre-

dnisolone and intravenous cyclophosphamide.1,7–9 Female sex

hormones are considered an important stimulatory factor

in this disorder and therefore oral contraceptives should not

be taken. Other risk factors (high blood pressure, smoking,

overweight) should be avoided or minimized.

A Schlez,* G Lischka, G Schaumburg-Lever, T Ganten, M Jünger

Department of Dermatology, University of Tübingen; Department of

Internal Medicine, University of Heidelberg, Germany. *Corresponding

author, University Medical Center Tuebingen, Department of

Dermatology, University of Tuebingen, Liebermeister Str. 25, 72076

Tuebingen, Germany, tel. +49 7071 298555; fax +49 7071 550134;


References1 Stockhaminer G, Felber SR, Zegler B et al. Sneddon’s syndrome:

diagnosis by skin biopsy and MRI in 17 patients. Stroke 1993; 24:

685–690.

2 Zeiger B, Sepp N, Schmid KW et al. Life history of cutaneous

vascular lesions in Sneddon’s syndrome. Hum Pathol 1992; 23:

438.

3 Tourbah A, Piette JC, IbaZizen MT et al. The natural course of

cerebral lesions in Sneddon syndrome. Arch Neurol 1997; 54:

53–60.

4 Menzel G, Reinhold U, Grunwaid F et al. Cerebral blood flow in

Sneddon’s syndrome. J Nuclear Med 1994; 35: 461–464.

5 Weissenborn K, Ruckert N, Ehrenheim C et al. Neuropsychological

deficits in patients with Sneddon’s syndrome. J Neurol 1996; 243:

357–363.

6 Zeiger B, Sepp N, Stockmann G et al. Sneddon’s syndrome.

A long-term follow-up of 21 patients. Arch Dermatol 1994; 129:

437–447.

7 Yamamoto M, Danno K, Shio H, Imamura S. Antithrombotic

treatment in livideo vasculitis. J Am Acad Dermatol 1988; 18:

57–62.

8 Kume M, Imai H, Miura AB, Namura I. Sneddon’s syndrome

presenting psychiatric disturbance and shortening of fingers and

toes. Intern Med 1996; 35: 668–673.

9 Daoud MS, Wilmoth G, Su WP, Pittelkow MR. Sneddon’s syndrome.

Semin Dermatol 1995; 14: 166–172.15If known2001293LettersLettersLetters1000Graphicraft Limited, Hong Kong

Primary cutaneous anaplastic large cell lymphoma in a young girl

To the Editor

Primary cutaneous large cell lymphoma is characterized by the

absence of nodal and visceral involvement, spontaneous

remission, low recurrence rate and being uncommon in

subjects under 20 years of age.1

A 17-year-old girl presented to us with a 6-month history of

non-healing skin ulcer on her left thigh. She had initially

noticed a painful, highly sensitive nodule that later developed

into a 10-cm plaque with ulceration. A clinical diagnosis of

pyoderma gangrenosum was made at the local hospital without

biopsy. The lesion was treated by chemical cauterization using

silver nitrate applications. Upon referral, physical examination

revealed a large ulcer of 7 × 6 cm with 1 cm nodular infiltrates

surrounding an atrophic mid-line scar (fig. 1). Laboratory

work-up gave unremarkable findings.

A punch biopsy from the edge of the ulcer revealed a dense

polymorphous infiltrate with neither Grenz zone nor epider-

motropism. The dermal infiltrate deeply invaded the skin adnexia

and subcutaneous fat, and there were reactive lymphocytes,

plasma cells and some neutrophils, large mitotically active cells

with coarse chromatin pattern and nucleoli (fig. 2). Immuno-

histochemical studies revealed strong positivity for CD3,

CD45-RO and, especially, CD30 (over 75% of the pleomorphic

cells). Few cells were positive for CD15 and EMA, and there

were a few reactive CD20-positive B cells. The differential count

of peripheral blood showed 34% lymphocytes with 80% CD80,

64% CD4 and 30% CD8 positive cells. A second biopsy taken

from this centre revealed the same findings with over 75%

CD30 and 40% Ki-67 positivity.

The subject has remained in complete remission for 2 years

with interferon (IFN)-α and chemotherapy.


Letters 367


In addition to the above, the girl presented erythemat-

ous scaling lesions on her scalp and body, diagnosed as

psoriasis vulgaris and treated with acitretin and ultraviolet

A phototherapy.

CD30-positive anaplastic large cell lymphomas (ALL),

a morphologically distinct group of large cell lymphomas,

rarely present with primary skin manifestation. Usually ALL

skin lesions present as solitary lesions in elderly males; however,

our case was a young female. Lesions are usually firm red to

violaceous tumours, up to 10 cm in diameter. Histologically,

there is a diffuse dermal and subcutaneous non-epidermotropic

infiltrate with atypical tumour cells.2 Most primary T-cell

lymphomas, other than mycosis fungoides and Sézary syn-

drome, are large cell lymphomas and an important propor-

tion of these express a CD30 lymphocyte activation marker.3,4

Histopathology of large cell cutaneous lymphomas shows very

dense infiltrate frequently extending into the subcutaneous fat,

and there is often ulceration. Generally, epidermotropism and

the Grenz zone are not evidenced.5

The initial diagnosis of pyoderma gangrenosum led to treat-

ment with chemical (silver nitrate) cauterization.

We also describe the de-novo development of psoriasis in the

patient, who was under treatment with IFN-α.6

AcknowledgementsWe thank Dr Peter Soyer and Dr Murat Okçu for their assist-

ance with the immunohistochemical staining and diagnosis of

this case.

E Hazneci,‡* NE Aydin,‡ G Dogan,† IO Turhan‡

Departments of †Dermatology and ‡Pathology, Inönü University Faculty

of Medicine, Turgut Özal Medical Center, Malatya, Turkey.

*Corresponding author, Turgut Özal Tip Merkezi, Dermatoloji Klinigi,

44069 Malatya Turkey, tel. +422 3410660/3508; fax +422 3410036;

E-mail: [email protected] [email protected]

References1 Willemze RK, Bejaards RC, Meijer CJLM. Classification of

primary cutaneous T cell lymphoma. Histopathology 1994; 24:

405–415.

2 Odom RB, James WD, Berger TG. Cutaneous lymphoid hyperplasia,

cutaneous t-cell lymphoma, other malignant lymphomas, and allied

diseases. Andrews’ Diseases of the Skin. Clinical Dermatology, 9th edn.

WB Saunders, Philadelphia, 2000.

3 Bernier M, Bapot M, Broyer M et al. Distinctive clinicopathologic

features associated with regressive primary CD30 positive

cutaneous lymphomas: analysis of 6 cases. J Cutan Pathol 1997; 24:

157–163.

4 Beljaards RC, Kaudewitz P, Berti E et al. Primary cutaneous large cell

lymphomas: definition of a new type of cutaneous lymphoma with a

favourable prognosis: a European multi-centre study on 47 patients.

Cancer 1993; 71: 2097–2104.

5 McKee PH. Cutaneous lymphoproliferative diseases and allied

disorders. In: Pathology of the Skin with Clinical Correlations,

2nd edn. Mosby-Wolfe, London, 1996: 12.1–22.

6 Wolfe JT, Singh A, Lessin AR et al. De novo development of

psoriatic plaques in patients receiving interferon alfa for treatment of

erythrodermic cutaneous T-cell lymphoma. J Am Acad Dermatol

1995; 32: 887–893.

fig. 1 Clinical appearance of the Ki-1 lymphoma lesion; a large healed ulcer

(9 × 8 cm) with 1 cm nodular infiltrates around an atrophic mid-line scar.

fig. 2 Histopathological picture showing large nucleoli and pleomorphic

nuclei in the anaplastic tumour cells (haematoxylin and eosin, original mag-

nification × 1000 oil immersion).


368 Letters


15If known2001269LettersLettersLetters1000Graphicraft Limited, Hong KongResponse of chronic bullous dermatosis of childhood to a combination of dapsone and nicotinamide

To the Editor

Chronic bullous dermatosis of childhood (CBDC) is a chronic

self-limiting blistering disorder usually seen in children less

than 6 years of age.1 It is characterized by deposition of a linear

band of IgA at the dermoepidermal junction evidenced by

direct immunofluorescence (DIF) of lesional and perilesional

skin.2 The treatment of choice for this condition is dapsone

(1.5–3 mg/kg per day) or sulphapyridine (150 mg/kg per day),

whereas resistant cases may require therapy with steroids.3 We

report a case of CBDC resistant to combined dapsone and

steroid therapy, but the patient was successfully treated with

dapsone associated with nicotinamide.

A 4-year-old girl presented to us with a 2-year history of itchy,

recurrent extensive, tense vesiculobullous lesions. The bullae

ruptured within a few days of formation, becoming crusted

lesions that eventually healed with hyperpigmentation. There

was no history of mucosal involvement or history of bulky

stools or diarrhoea. The child had been treated with dapsone

(100 mg daily, 5 mg/kg body weight) plus betamethasone

(2 mg daily) for more than 3 months, but without significant

improvement and with continued appearance of new bullae.

General physical and systemic examination gave normal

findings. The child weighed 20 kg. Cutaneous examination

revealed multiple, grouped, tense vesiculobullous lesions

containing clear or haemorrhagic fluid; the bullae ranged in size

from 0.5 to 1.5 cm and were on an erythematous base. The

lesions were seen predominantly on the lower part of the trunk,

upper and lower extremities and on the face, especially in the

perioral area. There were a few pustules and several hyper-

pigmented macules on the scalp, face, trunk and the upper

extremities. The Nikolsky sign was negative.

Haemogram, liver and kidney function parameters and

X-ray of the chest were within normal limits. Glucose-6-

phosphate dehydrogenase levels in the blood were normal.

The electrocardiogram (ECG) did not reveal any abnormality.

Skin biopsy from the forearm revealed a subepidermal split

with eosinophils and neutrophils in the bulla cavity and in the

dermis. DIF study showed a linear band of IgA along the base-

ment membrane zone.

Treatment was started with 50 mg (2.5 mg/kg body weight)

dapsone and 600 mg of nicotinamide daily (30 mg/kg body

weight). The patient was monitored for side-effects of dapsone

by monthly haematological and liver function tests. Through-

out the duration of therapy the child’s pulse rate and ECG

(done every fortnight) remained normal.

After 1 week of therapy, the old lesions had healed and the

appearance of new lesions had become sporadic. In a fortnight,

all the lesions had cleared. The dose of nicotinamide was grad-

ually tapered off over a period of 12 weeks. The patient was then

maintained on dapsone (25 mg) for an additional 12 weeks. She

has since remained lesion-free for the last 9 months.

Drugs of choice for CBDC are dapsone and sulphapyridine.3

Corticosteroids may be required in resistant cases but our

patient had failed to show any response to a combination of

dapsone and oral steroids. A combination of tetracycline and

nicotinamide has been found useful in the management of

bullous pemphigoid and linear IgA bullous dermatosis in

adults.4,5 The proposed mechanisms of action of tetracycline

and nicotinamide in autoimmune bullous diseases may include

their ability to inhibit neutrophil and eosinophil chemotaxis/

secretion, inhibit histamine release, inhibit phosphodiesterase

and suppress lymphocyte transformation.4

As tetracyclines should not be administered to children, we

decided to substitute it with dapsone. Our patient showed

an excellent response to this combination of dapsone and

nicotinamide, that could well be used as an important and safe

therapy in subjects unresponsive to dapsone alone. The mech-

anism of action of a combination of dapsone and nicotinamide

in this case of CBDC can be speculative at best: this combina-

tion has effects on lymphocyte transformation, decreases IgA

antibasement membrane zone antibodies and decreases

neutrophil chemotaxis.4

N Khanna,* RK Pandhi, S Gupta, MK Singh

Department of Dermatology and Venereology and Department of

Pathology, All India Institute of Medical Sciences, New Delhi, 110 029,

India. *Corresponding author, fax +91 11 4620665;


References1 Rabinowitz LG, Esterly NB. Inflammatory bullous diseases in chil-

dren. Dermatol Clin 1993; 11: 565–581.

2 Wojnarowska F, Marsden RA, Bhogal B, Black MM. The spectrum

of liner IgA dermatosis. In: Mac-Donald DM editor. Immunoderma-

tology. Butterworths, London, 1984: 245–249.

3 Chorzelski TP, Jablonska S. IgA linear dermatosis of childhood

(chronic bullous dermatosis of childhood). Br J Dermatol 1979; 102:

535–542.

4 Berk MA, Lorincz AL. The treatment of bullous pemphigoid with tet-

racycline and niacinamide. Arch Dermatol 1986; 122: 670–674.

5 Peoples D, Fivenson DP. Linear IgA bullous dermatosis: successful

treatment with tetracycline and nicotinamide. J Am Acad Dermatol

1992; 26: 498–499.15If known2001284LettersLettersLetters1000Graphicraft Limited, Hong Kong

Dermatitis artefacta

To the Editor

A 25-year-old woman with multiple, bilateral and symmetrical,

linear crusted erosions on her skin presented to the Department

of Dermatology, University of Dicle, Diyarbakır, Turkey in


Letters 369


June 1999, accompanied by her mother. Physical examination

revealed multiple linear crusted erosions on her extremities,

face and anterior aspect of the trunk. The lesions were scattered

on parts easily reached by the hands and had a tendency to

be symmetrical, with a regular and linear arrangement (fig. 1).

The back of the trunk was spared. According to the young

woman, these lesions had been developing naturally for

2 years, and there were only very short periods of recovery.

Her previous medical history was instructive. From 1989

to 1997 she had complained repeatedly of abdominal pain,

associated with constipation, nausea, vomiting and anorexia.

She had been diagnosed with Salmonella infection several

times, and treated by a general practitioner with no relief.

She was the eldest of six children, with four sisters and one

brother, and she was unmarried and lived with her mother. Her

father had died in a traffic accident in 1984. The young woman

claimed that due to her mother’s excessively authoritarian

character and inadequate socio-economic conditions, she herself

had not been able to develop fully her own personality and that

she had been alienated from the community. She was seen by a

clinical psychologist. The psychological examination suggested

a personality structure characterized by inward-looking

self-centred attitudes leading to increasing isolation. Electroen-

cephalography and cranial magnetic resonance were done to

evidence any possible brain tumour or damage, but with no

findings of note.

We prescribed amitriptyline HCl (50 mg/day) and 1% silver

sulphadiazine cream to be applied topically twice a day. On a

follow-up visit in August 1999, the old lesions showed healing

and re-epithelialization, but there were also new lesions. The

young woman did not return to our clinic after that. Follow-up

telephone conversations with her mother indicated that she

had discontinued the therapy, and continued to develop new

lesions. She is under follow-up at present.

Dermatitis artefacta is a psychocutaneous syndrome

included in the general spectrum of factitial disease. These self-

inflicted skin lesions are often difficult to treat.1 The disorder

is relatively rare, with a female/male ratio of up to 8 : 1. It is seen

most commonly in adolescents and young adults, but can occur

at any age.1–5

The diagnosis of factitial dermatitis should be based on the

appearance of the lesions, and on the subject’s aspect, personality

and history.1,2,6 The shape of artefact lesions is usually bizarre;

they have linear or geometric outlines, and superficial necrosis

is common.4,5 The lesions are distributed on body areas easily

reached by the hands. A detailed history allows the physician/

dermatologist to discover inconsistencies and contradictions

in the patient’s story.1 Subjects typically describe lesions that

appeared suddenly and give few details of how they evolved.1 If

an inconsistency is exposed, the subject rarely admits to lying,

but rather modifies the story, and when the inconsistency is too

great, typically becomes angry.1 The young woman in our case

presented numerous features typical of dermatitis artefacta,

including bizarre patterns not characteristic of any disease,

sharply defined lesions at different stages of healing at accessible

sites, sudden appearance of lesions in crops, only very short

periods of improvement, and a hollow history of how these

lesions evolved.

In most cases the psychiatric investigations reveal a back-

ground of emotional disturbances during the formative

years and in later life often resulting in feelings of isolation and

insecurity. The onset of dermatitis artefacta is very frequently

related to definable precipitating events, such as the loss of a

parent or sibling in early childhood, marital or family conflicts,

and depression.1,3 Some psychiatrists believe that subjects

may have an underlying immature personality, with self-injury

representing an appeal for help in moments of stress.1,3–5,7

A supportive and empathic approach that fosters a stable

patient–doctor relationship is advocated by most specialists,

rather than direct confrontation regarding the self-inflicted

injury.1,2,5,7 Once the subject is engaged in a stable therapeutic

relationship, the possibility of more long-term, directed therapy

should be considered, such as psychotherapy, which can be

a crucial element for recovery.1,8 However, if confrontation

occurs too soon, the subject may interrupt contact with the

physician and seek care elsewhere.1,5,8,9

In many reports, recovery has seemed to depend more

on changes in life circumstances than on actual medical

management.1,3,5,6,9fig. 1 Multiple crusted erosions on parts easily reached by the hands.


370 Letters


M Harman,†* S Akdeniz,† Y Bayram‡

†Department of Dermatology and Venereology, Faculty of Medicine,

University of Dicle, Diyarbakır, Turkey, ‡Department of Psychiatry,

State Hospital, Diyarbakır, Turkey. *Corresponding author, Dicle

Üniversitesi Tıp Fakültesi Dermatoloji Anabilim Dalı, 21280 Diyarbakır,

Turkey, tel. +90 412 2488662; fax +90 412 2488440;


References1 Joe EK, Li VW, Magro CM et al. Diagnostic clues of dermatitis

artefacta. Cutis 1999; 63(4): 209–214.

2 Koblenzer CS. Psychological aspects of skin disease. In: Fitzpatrick

TB, Eisen AZ, Wolff K et al. editors. Dermatology in General Medicine.

McGraw-Hill, New York, 1993: 14–26.

3 Fabisch W. Psychiatric aspects of dermatitis artefacta. Br J Dermatol

1980; 102: 29–34.

4 Koblenzer CS. Cutaneous manifestations of psychiatric disease that

commonly presented to the dermatologist–diagnosis and treatment.

Int J Psychiatry Med 1992; 22: 47–63.

5 Sneddon I, Sneddon J. Self-inflicted injury: a follow-up study of

43 patients. Br Med J 1975; 3(5982): 527–530.

6 Carew-McColl M, O’Dwyer S. A case of dermatitis artefacta.

Practitioner 1982; 226(1368): 1170–1172.

7 Gupta MA, Gupta AK, Haberman HF. The self-inflicted dermatoses:

a critical review. Gen Hosp Psychiatry 1987; 9: 45–52.

8 Gieler U. Factitious disease in the field of dermatology. Psychother

Psychosom 1994; 62(1–2): 48–55.

9 Edlich RF, Tafel JA, Smith JF et al. Factitious skin wounds.

Comp Ther 1987; 13(8): 57–61.15If known2001279LettersLettersLetters1000Graphicraft Limited, Hong Kong

Angiosarcoma of the scalp: new treatment modalities

To the Editor

Angiosarcoma of the head and neck region is characterized by

an aggressive course with early metastatic spread. The prognosis

of cutaneous angiosarcoma of the scalp and face is poor, with a

5-year survival rate of 10–22% and a mean survival length of

24 months.1 The most important prognostic factor is the size

of the lesion on presentation. Patients with lesions larger than

10 cm in diameter have a 5-year survival of 0%.2 If a complete

excision with wide margins is not possible, the local recurrence

rate is high.

Chemotherapy or interferon-α alone are of limited value.

Radiation without surgery is palliative. The 5-year actuarial

survival rate for patients irradiated with clinical disease is 13%,

for those without clinical disease it is 40%. The 5-year actuarial

incidence of distant metastases is 63%.3 The available adjuvant

treatment leads to a 5-year survival rate of 33% with only 20%

of patients being disease free.2

In a recent issue of the Journal of the European Academy of

Dermatology and Venereology, Ulrich et al.4 reported on the

long-term survival in an 88-year-old woman with angiosarcoma

of the scalp using a combination of intralesional cytokines

(interferon-α and interleukin-2) and surface irradiation with

several courses. It is remarkable that a patient with a very

extended lesion (20 × 15 cm) could achieve a 2-year survival.

Cytokine therapy, however, may cause some problems with

side-effects in the long term as reported by Ulrich et al.

We would like to refer to another therapeutic modality,

which showed an excellent tolerability and comparable

efficacy.5 A 79-year-old woman was referred to the Department

of Dermatology and Allergology, Friedrich-Schiller-University

(Jena, Germany) because of a 3-year history of a painless, growing

bluish plaque on her scalp. On examination we observed a palpable

bluish plaque of about 20 cm in diameter on the right side of the

scalp extending from the occipital region to the lateral forehead

with a central ulceration of about 5 cm in diameter. Chest

X-ray, lymph node sonography and cranial magnetic resonance

imaging provided no evidence of metastatic disease. We treated

her with intravenous liposomal doxorubicin (Caelyx®, Schering-

Plough, Essex, UK) 20 mg/m2 body surface (i.e. 30 mg) once a

month. Side-effects were monitored by clinical and laboratory

examinations, electrocardiography and cardial sonography.

After 12 infusions, we observed a partial remission with a

> 50% decrease of the affected area and disappearance of

ulceration. A total of 21 infusions was given and a stable disease

was achieved. After this we continued with electron beam

radiotherapy using fractionated doses of 2 Gy five times per

week for up to a total of 40 Gy. To ensure the maximum dose in

the upper layer of the dermis a bolus technique was used. At the

end of treatment a remarkable regression of the cutaneous lesion

was noted. Over the last 26 months she has not developed any

metastatic spread. Side-effects were only temporary in nature

and minor in intensity, i.e. increase of transaminases and alkaline

phosphatase (grade 1) and there was no sign of cardiotoxicity.

In our hands, low-dose liposomal doxorubicin followed by

fractionated electron beam irradiation was a safe and effective

therapy.

U Wollina,†* T Graefe,† J Füller‡

Departments of †Dermatology and Allergology and ‡Radiology,

Friedrich-Schiller-University, Erfurter Street 35, 07740 Jena, Germany.

*Corresponding author, tel. +49 3641 937357; fax +49 3641 937364;


References1 Caldwell JB, Ryan MT, Benson PM, James WD. Cutaneous

angiosarcoma arising in the radiation site of a congenital hem

angioma. J Am Acad Dermatol 1995; 33: 865–870.

2 Lydiatt WM, Shaha AR, Shah JP. Angiosarcoma of the head and neck.

Am J Surg 1994; 168: 451–454.


Letters 371


3 Morrison WH, Byers RM, Garden AS et al. Cutaneous angiosarcoma

of the head and neck. A therapeutic dilemma. Cancer 1995; 76:

319–327.

4 Ulrich J, Krause M, Brachmann A et al. Successful treatment of

angiosarcoma of the scalp by intralesional cytokine therapy and sur-

face irradiation. J Eur Acad Dermatol Venereol 2000; 14: 412–415.

5 Wollina U, Füller J, Graefe T et al. Angiosarcoma of the

scalp–treatment with liposomal doxorubicin and radiotherapy.

J Cancer Res Clin Oncol 2001; 127: in press.15If known2001299LettersLettersLetters1000Graphicraft Limited, Hong Kong

Sun-protective clothes: accuracy of laboratory testing

To the Editor

Ultraviolet (UV) radiation protection provided by clothes is a

subject of considerable current interest.1 According to the

Australian/New Zealand Standard,2 the determination of the

UV protection factor (UPF) of textiles has become an accepted

laboratory-based method using spectrophotometric measure-

ments (in vitro). On one hand, stringent requirements for the

measurement devices are necessary to determine precisely the

UPF of textiles, on the other hand, practical and time-saving

methods are required. We here report on in vitro measurements

of UPF with the aid of a newly developed automated sampling

device.

Fifteen different plain woven summer fabrics (4 × cotton,

4 × viscose, 7 × polyester) with uniform colours were investig-

ated with a Cary 500 UV/V is /NIR spectrophotometer (Varian

Deutschland GmbH, Darmstadt, Germany). The transmission

measurements were performed at wavelengths ranging from

280 to 400 nm (interval: 1 nm) using a fluorescence filter (UG-11;

Schott, Mainz, Germany). The spectrophotometer sample

compartment was specially customized with a diffuse reflect-

ance accessory and an integrated autosampler containing a

magazine for 100 samples (ecb ONLINE Analysentechnik

GmbH, Schwerin, Germany). One specimen of each textile was

fixed in a common slide frame put into the slide frame magazine,

which was placed in the autosampler. All the specimens were

automatically transported from the sample magazine into the

measurement position in front of the integrating sphere. In

this position the samples were measured four times by sub-

sequent 90° rotations after each computer-controlled scan.

After the measurement and calculation of the mean UPF of the

four scans the 15 samples were removed from the slide frame

magazine. Subsequently, the samples were replaced in the same

way for reassessment.

The UPF was repetitively determined eight times for all the

textiles. All system functions (e.g. calculation of the UPF)1 were

performed with a specifically dedicated Windows program.

Data were assessed by analysis of variance, Pearson’s correlation

and two-tailed paired Student’s t-test. Differences were con-

sidered significant for P < 0.05.

The UPFs of the different textiles ranged from 3.3 to 57.5.

UVA transmission ranged from 4.8 to 31.4% and UVB trans-

mission from 1.4 to 31.7%. Mean EUPF (standard error × 100/

UPF) of the UPFs was 1% (range: 0.1–3.6%). A significantly

higher EUPF was found for seven textiles with UPFs > 30

(r = 0.78; P < 0.001). The mean EUV (standard error × 100/UV

transmission) of the UVA transmission (3.9%; range: 0.2–

8.4%) differed significantly (P < 0.05) from the mean EUV of the

UVB transmission (1.1%; range: 0.01–0.9%). Percentage of

UVA transmission was significantly higher than percentage

of UVB transmission (r = 0.98; P < 0.001). The mean of

UVA/UVB ratio for polyester was 2.9 and for cotton and viscose

1.5 (P < 0.05). The measurement of the 15 textile samples took

about 45 min (total time for 120 measurements was 6 h).

In their comparisons of measurements done in several differ-

ent several laboratories, both Gies et al.3 and Laperre et al.4

reported that the reproducibility of measurements clearly

decreases with an increasing UPF level. We also observed an

increasing variability for higher UPFs in repeated measure-

ments done in a single laboratory, although the mean EUPF of

the UPFs (1%) indicates high measurement precision. It is

known that greater variability at higher UPFs could be due

to instrumental inaccuracies at very low transmittances. In

accordance with previous studies we found relatively high UPFs

and significantly increased UVA/UVB transmission ratios for

polyester.5,6 The mean EUV for UVB transmission was 1.1%, and

for UVA transmission 3.9%. This disparity could be due to the

use of a fluorescence filter generating apparent noise in the

380–400 nm range.

In contrast to autosampler loading, with manual introduc-

tion of the samples it can be difficult to place the textile samples

exactly vertical to the radiation beam (‘worst case principle’)

and in a well reproducible position in front of the integrating

sphere.1,6 Correspondingly, our previous spectrophotometric

investigations with manual sample introduction resulted in a

mean EUPF of 3.8% (unpublished data). Even a slight inclination

of the sample can lead to a decrease in UV transmission and

thus to the false calculation of UPF value (too high). This source

of error was avoided by the automated transport of the samples

to a sample holder precisely positioned directly at the aperture

of the integrating sphere. In this position the samples were

rotated by 90° and measured four times by computerized con-

trol, allowing for assessment of each specimen in both the

machine and cross-machine directions.1,2 Using a slightly off-

centred spectrophotometer beam allows the more representa-

tive assessment of the sample in different locations thus yielding

a high degree of accuracy of measurement.

From a clinical perspective it hardly matters whether the UPF

of a fabric is, for instance, 31.4 or 29.5. However, for rating or

not rating a textile as UV protective clothing, such small differ-

ences may be significant. In particular, for UPF classification

schemes based on class width of 5 or 10, i.e. AS/NZS,2 a high

degree of precision is required.


372 Letters


Moreover, the autosampler offers the advantage of un-

attended time-saving measurements once the samples are

logged into the software and loaded in the magazine; this is very

important for high output manufacturers and test laboratories

performing routine spectrophotometric measurements on a

large number of fabrics.

In conclusion, this practical automated device for spectro-

photometric measurements gives repeatable, precision measure-

ments of UPF values for textiles.

T Gambichler,* A Bader, A Avermaete, P Altmeyer, K Hoffmann

Department of Dermatology, Ruhr-University Bochum, Gudrunstr. 56,

D-44791 Bochum, Germany. *Corresponding author, tel. +49 2345093444;

fax +49 2345093445; E-mail: [email protected]

References1 Hoffmann K, Kaspar K, Gambichler T, Altmeyer P. In vitro and

in vivo determination of the UV protection factor for lightweight

cotton and viscose summer fabrics: a preliminary study. J Am Acad

Dermatol 2000; 43: 1009–1016.

2 Australian/New Zealand Standard. Sun protective clothing–

evaluation and classification. AS/NZS 4399: 1996.

3 Gies HP, Roy CR, McLennan A et al. UV protection by clothing: an

intercomparison of measurements and methods. Health Phys 1997;

73: 456–464.

4 Laperre J, Gambichler T, Driscoll C et al. Determination of

the ultraviolet protection factor of textile materials: measurement

precision. Photodermatol Photoimmunol Photomed 2001,

in press.

5 Davis S, Capjack L, Kerr N, Fedosejevs R. Clothing as protection

from radiation: which fabric is most effective? Int J Dermatol 1997;

36: 374–379.

6 Robson J, Diffey BL. Textiles and sun protection. Photodermatol

Photoimmunol Photomed 1990; 7: 32–34.15If known2001297LettersLettersLetters1000Graphicraft Limited, Hong Kong

Asymmetric periflexural exanthem of childhood in an adult patient with parvovirus B19

To the Editor

Asymmetric periflexural exanthem of childhood (APEC) is an

exanthem of unknown aetiology that was described in young

children by Taieb et al. in 1986;1 however, a case of the disease

was first reported by Brunner et al. in 1962.2 APEC is typically

described as a macular-papular scarlatiniform eruption or an

eczematous dermatitis that involves one axillary fold with

centrifugal spreading on the thorax and the proximal part of

the corresponding arm. The alternative name of unilateral

laterothoracic exanthem of childhood (ULEC) was proposed by

Bodemer and de Prost et al. in 1992.3 The disease does not affect

the subject’s general health, but it can provoke moderate itching

and mild local lymphadenopathy and is found in about 50% of

cases. Sometimes the exanthem diffuses with minor lesions

elsewhere on the body. The condition resolves in about

1 month, leaving mild hyperpigmentation or furfuraceous

scaling. Skin biopsy usually is not contributory, in particular

histological findings are non-specific, showing areas of epi-

dermal spongiosis with exocytosis of mononuclear cells and

in the dermis moderate perivascular and periappendageal

lymphohistiocytic infiltrate. The lack of response to antibiotics,

the age of affected individuals, typically children between 1 and

4 years old, the occurrence of small epidemics, the frequency of

associated upper respiratory tract prodromes and the possible

seasonal nature of the eruption, all suggest a viral origin. This

is the third reported case of APEC in an adult subject, but the

first in which seroconversion for parvovirus B19 has been

demonstrated.

A 41-year-old, previously healthy, Caucasian man was

admitted to our hospital for fever (39–40 °C) associated with

malaise, myalgia, sternalgia and headache. Four days later, when

these symptoms were diminishing, a papular–purpuric eruption

appeared around the right axillary flexure (fig. 1). This dermatitis

fig. 1 Papular–purpuric eruption involving the right axillary flexure, extend-

ing along the lateral thoracic wall and the proximal flexural part of the arm.


Letters 373


had not been preceded or accompanied by any other dermato-

logical signs, such as the herald patch of pityriasis rosea, and there

was no indication of insect bites or drug administration. The

lesions were discrete tiny red papules that did not tend to

coalesce; within a few days they spread centrifugally, extending

along the lateral thoracic wall and the proximal flexural part of

the arm. The rash remained unilateral until it completely dis-

appeared spontaneously within 2 weeks and without sequelae.

No associated symptoms or signs were noted and regional

lymph nodes were not enlarged.

Besides transient neutropenia and thrombocytopenia (white

blood count: 1.6 × 103/µL; neutrophilic count: 550 × 10/µL;

platelets: 93 × 103/µL), routine screening, including chest X-ray,

were within normal limits or negative. Bacteriological blood,

urine and throat cultures were negative. Bone marrow smear

was normal except for a reduction of the megakaryocytes and

signs of dysmegakaryopoiesis. The man refused to allow a skin

biopsy. Laboratory investigations revealed positive IgM and

negative IgG (enzyme immunoassay) for human parvovirus

B19, indicating recent infection. Titres for antibodies against

Borrelia burgdorferi, hepatitis B and C viruses, human immuno-

deficiency virus and Toxoplasma gondii were negative, while titres

against Epstein–Barr virus and cytomegalovirus were positive,

but only for IgG, ruling out recent infection. Follow-up titres for

antibodies against human parvovirus B19 were IgG-positive

and IgM-positive after 2 months, but IgG-positive and IgM-

negative after 4 months. We did not administer any medication,

both because the symptoms were minimal and because of the

inefficacy of treatments in previously reported cases to modify

the clinical course of the disease. No recurrence of the exanthem

was observed during this period.

When Taieb et al.1 first described the entity in five children,

they called it ‘erythème localisé avec adénopathie règionale de

l’enfant’ (‘localized erythema with regional adenopathy of the

child’). Bodemer and de Prost in 1992 gave the disease the

English name ‘unilateral laterothoracic exanthem in children’

(ULEC).3 Since then, further cases of the same disease have

been reported in Europe and North America with various

names, including ‘unilateral periflexural exanthem of child-

hood’ (UPEC).

We believe that the name ‘asymmetrical periflexural exan-

them of childhood’ (APEC) is preferable because the exanthem

is not always unilateral, although it does start specifically on

one side of the body, and is not always laterothoracic, as has

been recently documented in a case with limb involvement but

without involvement of the thorax.7

To date, despite the fact that hundreds of patients have been

studied and reported1–10 the aetiology of the disorder remains

obscure. As the initial hypothesis of bacterial origin has not

been confirmed, a possible viral aetiology has been suggested.

The differential diagnosis of APEC includes: irritant and

allergic contact dermatitis, miliaria, and, particularly, pityriasis

rosea, and Gianotti–Crosti syndrome.

Contact dermatitis can be ruled out by the subject’s history

because its lesions are more vesicular and pruritic and are sens-

itive to topical corticosteroids. Miliaria usually affects small infants,

is symmetrical and almost always affects the forehead and the

neck, in the warm season. Pityriasis rosea is also symmetrical

and presents larger oval, centripetally scaling lesions preceded by

the classic herald patch. The most difficult differential diagnosis

is with atypical pityriasis rosea, as both atypical pityriasis rosea

and APEC may lack the herald patch and present a papular

aspect. Nevertheless, APEC is more clearly asymmetric while

pityriasis rosea is more clearly symmetric, except for the herald

patch, which, in fact, is lateralized. The clinical course of APEC

is generally shorter, 4 weeks compared with 6–8 weeks for

pityriasis rosea. Furthermore, on histological examination

dyskeratotic cells of pityriasis rosea have never been found in

APEC. It is also true that both pityriasis rosea and APEC have

biphasic eruption, but in the former the first lesion (i.e. the herald

patch) is minimal compared with the subsequent ‘explosion’ of

the rash, while in the latter the first rash is, by far, the most

evident dermatological sign. This explains also why both the

‘herald patch’ and the small, late, controlateral lesions of APEC

can be easily ignored.

Individual lesions of Gianotti–Crosti syndrome may resemble

those of APEC, but they are symmetrical and, although the

trunk may be slightly affected at the beginning of the eruption,

the lesions are predominantly on the limbs and face. Thus, the

differential diagnosis between Gianotti–Crosti syndrome and

APEC would be very easy if APEC always involved lesions on the

trunk (the so-called, laterothoracic). However, this is not always

so, and a small percentage of subjects, from the first description

by Brunner et al.2 to a more recent publication, present lesions

on the limbs.7 In these cases there are two major differences

between Gianotti–Crosti syndrome and APEC: in Gianotti–

Crosti the lesions on the limbs are accompanied by lesions on

the face, whereas in APEC, when the lesions are located on the

limbs, but not the thorax, there are no facial lesions; further-

more, the rash of Gianotti–Crosti starts contemporaneously on

both sides, whereas in APEC the rash always affects the limbs

in two different phases, even in those rare cases where it may

appear to be practically symmetrical. This observation is

important because it can be recorded from the history, also in

the late phase of the eruption when the objective lesions are

fading or have been altered by treatment. As also happened for

Gianotti–Crosti syndrome, also in APEC after many observa-

tions in children, the disease has been described in adults.6,8

However, our case is the first in which a viral aetiology has been

confirmed. It is not surprising that parvovirus B19 was found

only in our case because many exanthems can be provoked by

different viruses, as in the case of Gianotti–Crosti syndrome or

‘gloves and socks’ syndrome, and in Gianotti–Crosti syndrome

the aetiology is unknown in almost 50% of cases. Thus, APEC

is a cutaneous affection that can probably be caused by various

microorganisms, including parvovirus B19.


374 Letters


P Pauluzzi,†* G Festini,‡ C Gelmetti§

†Clinica Dermatologica dell’Università di Trieste, Italy, ‡Î Div. Medica

‘Azienda Ospedaliera’ di Trieste, Italy, §Istituto di Scienze Dermatologiche

dell’Università di Milano, IRCCS ‘Ospedale Maggiore’ di Milano, Italy.

*Corresponding author, Istitute of Dermatology, University of Trieste,

Ospedale di Cattinara, Strada per Fiume, I-34149 Trieste, Italy,

tel. +39 040 399 43 27; fax +39 040 910 415;


References1 Taieb A, Megraud F, Le Roy JM et al. Erythème localisé avec

adénopathie régional de l’infant: une maladie d’inoculation?

Ann Dermatol Venereol 1986; 113: 1023–1025.

2 Brunner MJ, Rubin L, Dunlap E. A new papular erythema of

childhood. Arch Dermatol 1962; 85: 539–540.

3 Bodemer C, de Prost Y. Unilateral laterothoracic exanthem in

children: a new disease? J Am Acad Dermatol 1992; 27: 693–696.

4 Gelmetti C, Grimalt R, Cambiaghi S, Caputo S. Asymmetric

periflexural exanthem of childhood: report of two new cases.

Pediatr Dermatol 1994; 11: 42–45.

5 Harangi F, Varszegi D, Szucs G. Asymmetric periflexural exanthem of

childood and viral examinations. Pediatr Dermatol 1995; 12: 112–115.

6 Corazza M, Virgili A. Asymmetric periflexural exanthem in an

adult. Acta Derm Venereol (Stockh) 1997; 77: 79–80.

7 Gelmetti C. Asymmetrical acrodermatitis. J Eur Acad Dermatol

Venereol 1996; 7 (Suppl. 2): 10.

8 Gutzmer R, Herbst RA, Kiel P, Kapp A. Unilateral laterothoracic

exanthem (asymmetric periflexural exanthem of childhood):

report of an adult patient. J Am Acad Dermatol 1997; 37: 484–485.

9 Maroon M, Billingsley EM. Unilateral laterothoracic exanthem

(letter). J Am Acad Dermatol 1993; 29: 130.

10 Fort DW, Greer KE. Unilateral laterothoracic exanthem in a child

with acute lymphoblastic leukemia. Pediatr Dermatol 1998; 15: 51–52.15If known2001294LettersLettersLetters1000Graphicraft Limited, Hong Kong

Atypical tinea corporis caused by Microsporum gypseum in a subject with acquired immune deficiency syndrome

To the Editor

Microsporum gypseum is the most common species of geophilic

dermatophytes that cause human disease and it is isolated

from soil throughout the world.1,2 In Rosario metropolitan area

and surroundings it is found in a high percentage of samples

from areas with vegetation.2 Studies of the aetiology of

dermatophytosis in Argentina have shown a low percentage of

M. gypseum infections. It is an unusual agent in tinea corporis

and tinea capitis.2

During the clinical course of acquired immune deficiency

syndrome (AIDS), mycoses such as candidiasis, cryptococcosis

and histoplasmosis are some of the most important complica-

tions.3,4 The prevalence of dermatophytosis in patients infected

with the human immunodeficiency virus (HIV) is probably

similar to that in the non-HIV-infected population.5,6

A survey done in our centre (CEREMIC, Faculty of Biochem-

istry and Pharmacology, the National University of Rosario,

Argentina) involving over 576 clinical samples from 255 HIV+

subjects in recent years showed that 1.02% of the diagnosed

mycoses were dermatophytosis. In all these cases the aetiologi-

cal agent was Trichophyton rubrum.4

Dermotophytic infections usually present similar symptoms

in HIV+ and non-HIV patients. Tinea corporis presents annu-

lar plaques with central clearing and scales at the edges. In HIV-

infected patients the dermatophytosis occasionally exhibits

different clinical characteristics: large and exuberant lesions,

without clear peripheral delimitation, associated with a strong

tendency for dissemination and that are usually refractory to

topical and systemic therapy.7 In these cases T. rubrum is the

fungus most frequently involved.5,7,8

We describe the first report in Argentina of a patient

with AIDS who had an atypical tinea corporis produced by

M. gypseum.

A 39-year-old white male diagnosed as HIV+ in August 1996

had developed a Pneumocystis carinii infection in December

1996 and bilateral chorioretinitis in April 1998. He received

antiretroviral treatment with zidovudine, dideoxyinosine and

Indinavir, and trimethoprim/sulphamethoxazole as P. carinii

pneumonia prophylaxis.

In June 1998 the man developed seborrhoeic dermatitis,

oropharyngeal candidiasis and genital herpes. His viral load

increased to 155 000 copies/mL with a decrease in CD4 to

125 cells/mL. Meanwhile, he presented erythematous, vesicular

and granulomatous skin lesions with clear, lightly infiltrated

edges on his lower legs. The skin surrounding the lesions was

scaly and pruritic (fig. 1).

We scraped both types of lesions (vesicular and scaly) with a

blunt scalpel for mycological examination. Direct microscopic

examination with (20%) potassium hydroxide showed abund-

ant hyaline, septated hyphae typical of dermatophyte

fig. 1 Erythematous, vesicular, granulomatous and scaly skin lesions on the

right lower leg.


Letters 375


infection. A portion of the scrapings was inoculated in tubes,

with Sabouraud dextrose agar (DIFCO Laboratories, Detroit)

(with and without chloramphenicol), Mycosel agar (DIFCO)

and V-8 juice agar and incubated at 28 °C. The culture developed

a fungus identified as M. gypseum based on morphological

features as seen in fig. 2.

Based on laboratory findings, fluconazole treatment was

started (200 mg/day). A slow regression of the lesions was

observed after 2 months of treatment.

A review of the literature showed five reported cases of tinea

corporis caused by M. gypseum in subjects with AIDS, one in

France9 and four involving Brazilians.6,7,10 All of these subjects

presented extensive, scaly, circinated and erythematous lesions

that were very resistant to treatment. In our case, however, the

antifungal treatment with fluconazole was successful.

The atypical clinical manifestations presented by our subject

can be attributed to the deterioration of his immune system;

although he was in a state of advanced immunosuppression and

antiretroviral treatment was unsuccessful he experienced no

relapses of the dermatophytosis before his death 1 year later.

AG Luque,† MS Biasoli,†* MA Sortino,† SH Lupo,‡ RF Bussy§

†CEREMIC, Facultad de Ciencias Bioquímicas y Farmacéuticas

Universidad Nacional de Rosario, Suipacha 531, Rosario, Rep. Argentina,

‡Primera Cátedra de Clínica Médica and §Cátedra de Dermatología,

Facultad de Ciencias Médicas, Universidad Nacional de Rosario, Santa Fe

3100, Rosario, Rep. Argentina. *Corresponding author, Sarmiento 308,

cuarto piso, 2000, Rosario, República Argentina;


References1 Kwon-Chung KJ. Medical Mycology. Lea & Febiger, Philadelphia,

PA, 1992: 138.

2 Bracalenti BJC, Alvarez DP, Colella MG. Ecología de los dermatofitos.

Correlación entre dermatofitias y hongos queratinolíticos de suelos

de Rosario. Sabouraudia 1975; 13: 255–262.

3 Negroni R. Micosis en pacientes con SIDA. Rev Argent Micol 1990;

13: 3–14.

4 Luque AG, Tosello ME, Gómez CR, Bortolozzi R. Manifestaciones

cutáneas de histoplasmosis en pacientes infectados con el virus

de inmunodeficiencia humana. Infectol Microbiol Clín 1999; 11:

30–35.

5 Ramonda S, Salerni G, Macia A et al. Micosis superficiales más

frecuentes en pacientes infectados por HIV. Rev Argent Dermatol

1994; 75: 34–35.

6 Porro AM, Yoshiola MCN, Kaminski SK et al. Disseminated

dermatophytosis caused by Microsporum gypseum in two patients

with the acquired immunodeficiency syndrome. Mycopathologia

1997; 137: 9–12.

7 Giudice MC, Szeszs MW, Scarpini RL et al. Clinical and

epidemiological study in an AIDS patient with Microsporum

gypseum infection. Rev Iberoam Micol 1997; 14: 184–187.

8 Tsang P, Hopkins T, Jimenez-Lucho V. Deep dermatophytosis

caused by Trichophyton rubrum in a patient with AIDS. J Am Acad

Dermatol 1996; 34: 1090–1091.

9 Blanc V, Cremer G, Benkhraba F et al. Dermatophytie á

Microsporum gypseum chez un patient atteint du syndrome

D’immunodéficience acquise (SIDA). J Mycol Méd 1994; 4:

172–174.

10 Fernandes NC, Lamy F, Akiti T, da Barreiros M, GC. Microsporum

gypseum infection in AIDS patient: a case report. An Bras Dermatol

1998; 73: 39–41.15If known2001300LettersLettersLetters1000Graphicraft Limited, Hong Kong

Tinea capitis in two young women: possible favouring role of hair styling products

To the Editor

Tinea capitis is a very contagious disease caused by

dermatophytes belonging to the Trichophyton and Microsporum

genera, and only exceptionally to those of the Epidermophyton

genus. It usually affects children between 3 and 8 years old, who

are often infected at school or in family epidemics.1

About 3 weeks before coming to our attention an 18-year-old

young woman noticed a round, slightly inflamed spot on her

scalp in the proximity of the anterior hairline. Clinical exami-

nation revealed an erythematous lesion with sharp margins

covered at intervals with scales and hairs broken off at 4–5 mm

from the follicular ostium, intermingled with normal appearing

hair with no resistance to traction (fig. 1).

A second young woman, 21-year-old, came to us for a lesion

presenting clinical features similar to those of the case previ-

ously described. This lesion had arisen about a month before

(fig. 2).

In both cases, microscopic examination of the scales and

hair taken from the lesion showed pilar trunks surrounded

by spore muffs, sometimes containing mycetium hyphae in

the inner part (endo-ectotrix parasitism). Cultures on Sabour-

aud medium identified Microsporum canis as the aetiological

fig. 2 Microsporum gypseum slide culture preparation with typical macroconidia

(cotton-blue lactophenol, original magnification × 400).


376 Letters


agent in both cases. Routine tests of blood chemistry gave

unremarkable findings, and endocrine and immune profiles

were normal.

Both subjects were treated systemically with terbinafine

(250 mg/d for 15 days) and local therapy with naftifine, with

complete clinical and laboratory remission in 6 weeks.

Tinea capitis is a typical childhood problem caused by

dermatophytes.1 Dermatophytosis of the scalp is uncommon in

otherwise healthy adult persons,2–5 and when it does occur in

this group it usually involves women during the menopause

(because of involution of the sebaceous glands and reduced

secretion of antimicotic fatty acids at infection sites).6–10

The cases reported here are worth noting because they

involve young women in good health with regular menstrual

cycles. A careful review of both histories revealed year-long use

of several unspecified hair-styling products until about a week

before our observation. It is very likely that these cosmetic

products changed the quantity and/or the quality of local

sebaceous secretion, favouring susceptibility to infection by

Microsporum canis. Because of the widespread use of such hair

products, especially by teenagers, we thought these unusual

clinical cases should be reported.

A Vozza, E Fiorentini, F Tripodi Cutrì, F Di Girolamo,* RA Satriano

Second University of Naples, School of Medicine and Surgery, Department

of Dermatology, Via S. Pansini, 5–80131 Naples, Italy. *Corresponding

author, tel. +39 081 5666828; fax +39 081 5468759;


References1 Pinetti P. Le dermatofizie superficiali. In: Le dermatofizie. Piccin,

Padova, 1997.

2 Albanese G, Crippa D, Giorgetti P, Santagostino L. Tinea capitis

dell’adulto da Microsporum canis. G Ital Dermatol Venereol 1987;

125: 507–509.

3 Chernov A, Alteras I, Shohat B et al. An unusual case report: Tinea

capitis, verrucae vulgaris and other infections in a girl with T and B

cell disturbances. Mycorpathologia 1980; 72: 143–145.

4 Rozzoni M, Tribbia G, Marchesi L, Resegetti A, Cainelli T. Su un

caso di Tinea capitis et corporis microsporica nell’adulto. Chron

Dermatol 1980; 11: 389–390.

5 Vena G, Barile F. Epidermomicosi del cuoio capelluto in adulto con

parziale difetto dell’immunità cellulare. Dermatol Clin 1982; 2:

75–77.

6 Barile F, Loconsole F, Cantuccio F. Osservazioni clinico-

epidemiologiche in tema di Tinea capitis dell’adulto.

G Ital Dermatol Venereol 1980; 125: 507–509.

7 Cervetti O, Forte M, Paggio A. Tinea capitis dell’adulto. Descrizione

di tre casi. G Ital Dermatol Venereol 1990; 125: 27–28.

8 Difonzo EM, Vannini P, Poli M et al. Tinea capitis dell’adulto.

Micol Dermatol 1989; 3: 15–22.

9 Donofrio P, Montesano M. Tinea capitis in una donna anziana.

Ann It Dermatol Clin Sper 1982; 36: 89–90.

10 Lasagni A, Graziani F, Bassi G. Tinea capitis microsporica

dell’adulto. G It Dermatol Venereol 1977; 112: 627–630.15If known2001150LettersLettersLetters1000Graphicraft Limited, Hong Kong

Unilateral hyperkeratosis of nipple and areola associated with androgen insensitivity and oestrogen replacement therapy

To the Editor

Hyperkeratosis of the nipple and areola is a benign rare

dermatological disorder characterized by asymptomatic hyper-

pigmented and verrucous hyperkeratosis.

This condition is mainly of cosmetic importance and diagnosis

is usually by exclusion. It should be distinguished from the long

list of diseases of the nipple–areolar complex,1 including erosive

adenomatosis of the nipple, Paget’s disease, perifollicular elas-

tolysis, Mondor’s disease and eczema.

We report a unique case of unilateral hyperkeratosis of the

nipple and areola in a mature androgen-insensitive individual

with a totally female phenotype, on long-term oestrogen

replacement therapy.

fig. 1 Clinical presentation: presence of a round, slightly inflamed spot on

the scalp in proximity of the anterior hairline.

fig. 2 Clinical presentation: a second young woman with similar clinical lesion.


Letters 377


This 41-year-old happily married woman was referred to

the dermatology clinic with a 13-month history of yellow–

brown discoloration and thickening of her right nipple

and areola. At the age of 18 she had been diagnosed to have

incomplete androgen insensitivity syndrome (46XY karyotype)

following investigation of her primary amenorrhoea. Her

gonads (testes) were then removed and oestrogen replacement

therapy was commenced and continued without interruption

thereafter.

On presentation, she had a normal female body configura-

tion with normal breast development. On physical examina-

tion approximately one-third of the right areolar surface was

thickened, hyperpigmented and covered with papular warty

excrescences (fig. 1). The right nipple also appeared hyper-

keratotic with central desquamation. None of the involved

skin was indurated; it was not tender and not adherent to the

underlying structures.

No changes suggestive of acanthosis nigricans were noted

in the intertriginous zones or oral mucosa. Further cutaneous

examination did not reveal epidermal naevi, ichthyosis, chronic

dermatitis or cutaneous T-cell lymphoma.2 Histological exam-

ination showed papillomatous skin with orthokeratotic hyper-

keratosis and follicular plugging as well as acanthosis. Curettage

was performed and successfully removed most of the pigmented

warty growths, achieving a satisfactory cosmetic result. One

year later, there was no recurrence and the patient remained

satisfied with the aesthetic outcome.

In androgen-insensitive individuals, the androgen receptor

is either absent or has an altered amino acid sequence, such

that it can no longer bind androgens. These individuals are

brought up as females and breasts develop normally at puberty,

because there is effectively no testosterone to oppose the circu-

lating oestrogen that is derived from the aromatization of

testosterone.

Hyperkeratosis of the nipple and areola is a rare mammary

condition. The classical 1934 Levy-Franckel3 classification

refers to three variants: (i) unilateral, due to extension of an

epidermal naevus; (ii) bilateral, associated with a disseminated

dermatosis such as ichthyosis, acanthosis nigricans, Darier’s

disease or lymphoma; and (iii) the usually bilateral naevoid

form in young women, appearing after menarche or pregnancy,4

suggesting that hormones may play a part in the pathogenesis.

A fourth type of acquired hyperkeratosis of the nipples and

areolae associated with endocrinopathy, has been postulated

by Banuchi et al.,5 Schwartz6 and Mold and Jegasothy.7 They

reported cases of men with prostatic adenocarcinoma treated

with diethyl-stilboestrol who subsequently developed bilateral

hyperkeratoses of the nipples and areolae.

Our case seems to belong to this last group with the distinctive

feature of the asymmetrical unilateral distribution, also offering

further evidence to strengthen the hypothesis that hormonal

factors, oestrogens in particular, play a key role in the pathogenesis.

Regarding treatment,8 response to topical steroids, keratolytics,

retinoids, cryotherapy, surgical excision and carbon dioxide

laser9 has been reported as variable. In 1998, Marin-Bertolin

et al.10 reported that it was difficult for plastic surgeons to deal

with this condition due to the lack of literature on its surgical

treatment. Curettage is now demonstrated to be a simple and

very effective alternative for the treatment of idiopathic hyper-

keratosis and papillomatosis mammae.

AG Lambiris,†* F McCormick‡

Departments of †Dermatology and ‡Pathology, Derriford Hospital,

Plymouth, PL6 8DH, UK. *Corresponding author, tel. +1752 777111;

fax +1752 768976; E-mail: [email protected]

References1 Ward K, Burton J. Dermatologic diseases of the breast in young

women. Clin Dermatol 1997; 15: 45–52.

2 English JC, 3rd Coots NV. A man with nevoid hyperkeratosis of the

areola. Cutis 1996; 57: 354–356.

3 Levy-Franckel A. Les hyperkeratoses de l’areole et du mamelon.

Paris Med 1938; 28: 63–66.

4 Alpsoy E, Yilmaz E, Aykol A. Hyperkeratosis of the nipple: report

of two cases. Br J Dermatol 1997; 24: 43–45.

5 Banuchi SR, Cohen L, Lorincz AL. Acanthosis nigricans following

diethylstilbestrol therapy. Arch Dermatol 1974; 109: 545.

6 Schwartz RA. Hyperkeratosis of the nipple and areola. Arch

Dermatol 1978; 114: 1844–1845.

7 Mold DE, Jegasothy BV. Estrogen-induced hyperkeratosis of the

nipple. Cutis 1980; 26: 95–96.

8 Kuhlman DS, Hodge SJ, Owen LG. Hyperkeratosis of the nipple and

areola. J Am Acad Dermatol 1985; 13: 596–598.

9 Busse A, Peschen M, Schöpf E, Vanscheidt W. Treatment of

hyperkeratosis areolae mammae naeviformis with the carbon

dioxide laser. J Am Acad Dermatol 1999; 41: 274–276.

10 Marin-Bertolin S, Gonzalez-Martinez R, Marquina Vila P. Nevoid

hyperkeratosis of the areola. Plast Reconstr Surg 1998; 102: 275–276.fig. 1 Hyperkeratotic nipple and one-third of right areola thickened, hyper-

pigmented and covered with papular warty excrescences.


Tintack' sign in a patient with cutaneous B-cell lymphoma

Documents