Monday, 15 June 2009

Answer To June 2009 Case

Dear Bloggers

Welcome back. Today I will discuss the answer to the recent case. This was one of the most difficult cases I have ever experienced and I thought it would be nice to share this with you all. I was reminded by a colleague also involved with the case, that the rash was in fact painful rather than painless and hence, the initial description of her symptoms was inaccurate for which I apologise :-o

Professor Matsumura has kindly provided his opinion on the case (see below) in respect of the revised history of the patient. His ideas were very much in keeping with the team looking after the patient on initial presentation.

The patient was a 43-year-old woman. I would describe problem list first.

Problem list

#1 Skin rash

The lesions were circular, purplish, discrete, raised with an umbilicated central area. The lesions began as smaller discrete nodules that rapidly expanded. After several days, existing lesions developed a necrotic centre with circumferential erythema. The lesions eventually crusted over leaving large areas of eschar.

#2 Fever

#3 Tachycardia

#4 History of right breast cancer

#5 Acyclovir and ciprofloxacin administration

The main problem of this patient is fever and rash. Differential diagnosis is broad in fever and rash. However, I would think three main category, infection, neoplastic, and autoimmune. Characteristics of skin rash are likely infectious disease. Neoplastic or autoimmune is less likely. But neoplastic disease should be considered for background of this patient


1) From the history and physical examination what would be your likely differential diagnosis?

Vascular: Less likely

Infection: Varicella, HIV, molluscum contagiosum, disseminated cryptococcosis, eosinophilic folliculitis, cutaneous anthrax

Neoplastic: (Recurrence of breast cancer)

Autoimmune: Sarcoidosis

Toxic: Less likely

Metabolic: Less likely

Trauma: Less likely

Degenerative: Less likely

Iatrogenic: Less likely

Idiopathic: Less likely

Congenital: Less likely

I would think varicella is the most likely based on the character of skin rash. The problem is this patient was a 43-year-old woman. Varicella is common in children. Moreover, this patient had history of breast cancer and acyclovir administration.

2) What are the two most important tests to confirm the diagnosis?

Tzanck smear for detecting multinucleated giant calls.

Direct immunofluorescent antibody (DFA) staining.

3) What is your chosen treatment?

Depend on following assessment, question 4.

4) What addition problem should you consider screening for in this patient?

Interstitial pneumonia as complication of adult varicella should be carefully monitored. Cancer screening including recurrence of breast cancer should be evaluated. Moreover, I would confirm pregnancy test.

The Answer

The team caring for this patient were very concerned about an infectious disease such as varicella and even cutaneous anthrax, despite no history of there being any exposure to the latter infectious disease. Therfore, she was nursed with full barrier protection as a precautionary measure.

Her laboratory studies were normal except for a neutrophilia. Varicella IgG was present and IgM was absent suggesting previous infection. Autoimmune serology including ANCA was negative.

Blood cultures were negative.

The patient had the lesions tested for varicella zoster with viral cultures, which were negative.

Chest radiograph was normal.

CRP was 13.6 and ESR was 17mm/hr.

She developed secondary staphylococcal infection as evidenced by impetigo and positive skin cultures. She was commenced on ceftriaxone.

She was reviewed by the hospital infectious disease specialist and dermatologist respectively, who both considered infectious, autoimmune and primary dermatologic disease.

The dermatologist also considered Acute Febrile Neutrophilic Dermatosis (Sweet's Syndrome) in view of the presentation, background history, fever and neutrophilia.

The patient therefore underwent a skin biopsy which confirmed the histopathology consistent with Sweet's Syndrome. She also met the proposed criteria for Sweet's Syndrome (see below).

She was commenced on corticosteroids orally (Prednisolone 60mg per day) with resolution of her symptoms and the lesions over several days. She was discharged home with outpatient follow up.

In view of her history of metastatic breast cancer, a search for secondary causes of Sweet's Syndrome such as a recurrent malignancy, would have been considered appropriate.

Sweet's Syndrome

Described by Robert Sweet 1964: A reactive process characterised by the abrupt onset of tender red-to-purple circinate plaques.

Usually occur on the
  • extremities
  • head
  • neck
and typically are accompanied by fever and peripheral neutrophilia.


Neutrophil mediated hypersensitivity reaction to systemic factors, which may include, haematologic disease, infection, or drug exposure.

Evidence suggests that Sweet’s Syndrome is a cytokine T-cell-mediated disease with secondary and only temporary activation and participation of neutrophils.

Primarily skin affected but lung and kidney can be involved.


Uncommon condition; several hundred reported cases in literature ~2.7 cases/million (Scotland) Kemmett et al 1990

~10-15% cases in setting of malignancy (which includes recurrent breast cancer) but most cases are idiopathic and benign.

Females predominantly affected (F:M 2-3:1) esp. non-malignancy associated Sweet syndrome.

Age at onset
  • Females 30-50
  • Males 60-90
  • Idiopathic (~70% cases)
  • Haematologic Malignancy: CML, AML, MDS. Hodgkin disease, cutaneous T-cell lymphoma, NHL, multiple myeloma, and hairy cell leukemia.
  • Non-Haematological Malignancy: Slight increase in genitourinary cancers, rectal cancer, oral / tonsilar cancer, breast cancer and osteosarcoma.
  • Infection: Streptococcal pneumonia, Yersinia infection. Atypical mycobacteria.


G-CSF- well established cause

all-trans retinoic acid

Anecdotal - Limited reports of drug associations include
levonorgestrel/ethinyl estradiol.

Systemic Disorders

15% cases of Sweet’s Syndrome

Most common association
Crohn’s Disease
Ulcerative Colitis

Sjogren’s, Behcet’s, RA, SLE , undifferentiated connective tissue disease all associated with Sweet’s.

Proposed diagnostic criteria include the following:

Both major criteria must be met in addition to two of the four minor criteria.

Major criteria

1 Acute onset of typical skin lesions
2 Histopathological findings consistent with Sweet’s syndrome

Minor criteria

1 Fever > 38°C or general malaise
2 Association with malignancy, inflammatory disorder or pregnancy or antecedent respiratory or gastrointestinal infection
3 Excellent response to systemic corticosteroids or potassium iodide (KI)
4 Abnormal laboratory values at presentation (three of four required: ESR > 20 mm; leukocytes > 8000; neutrophils > 70%; positive C-reactive protein)

(Rook's Textbook of Dermatology, Seventh Edition)


Steroids!! Usually Prednisolone 1.0mg/kg/day -> Lesions should resolve <14>

Other treatment modalities
  • Colchicine 1.5 mg/day
  • Dapsone 100-200 mg/day
  • Clofazimine 200 mg/day
  • Cyclosporin A 5-10 mg/kg/ body weight/day
  • Indomethacin 50-100 mg/day
  • Naproxen 750 mg/day
  • Doxycycline 200 mg/day

von den Driesch P. 1994


  • Outcome depends on the underlying condition
  • Recurrence may occur in up to 50% of patients
  • Most likely in cases associated with haematological malignancy or drug reaction.
Take Home Message

When you see a patient with an unusual condition such as this, please involve the dermatologist as soon as possible for advice on the likely cause and for a skin biopsy to be taken.

Without a biopsy, despite the best problem list and differential diagnosis it can be difficult to establish the diagnosis even for the best of diagnosticians.

Certainly the history of the acute onset, location and progression of the disorder help to point to Sweet's syndrome as a possible diagnosis but a skin rash with fever has a wide differential. Hence, a biopsy is the very best way to establish the cause -- tissue is the issue!

Much of dermatology is pattern recongition and it can be difficult even for most senior doctors.

I would like to thank Prof Matsumura for his excellent attempt at diagnosis in this rare case. His suggestions were very good and would have led to the diagnosis through a skin biopsy as inferred from the two tests proposed. Moreover, checking for recurrent cancer and pregnancy showed good lateral thinking and were correct. Well done.