On July 8, 2004, a 29-year-old previously healthy white woman presented to the emergency department at the University of California Davis Medical Center for evaluation of a skin rash. Ten days before presentation, she was treated with an oral cephalosporin for sore throat and fever to 38.9°C. A rash developed on her shoulders, and the patient's antibacterial therapy was changed to a fluoroquinolone. The rash worsened, and the patient was referred to our facility for further management.
The patient lives with her husband and young son on a dairy farm in Merced County, Calif. She denied exposure to animals other than the family's pet dog and had no recent travel. She did not have a history of chickenpox as a child but had been told that her titer was positive. She had not been vaccinated against smallpox, nor had she been in contact with a vaccinee. At the time of admission, the patient was taking only oral contraceptive pills. She is reportedly allergic to ciprofloxacin, azithromycin, and sulfa products.
On examination, she was afebrile with normal vital signs and did not appear toxic. Her mucous membranes were clear, but there was shotty cervical lymphadenopathy. Skin examination revealed multiple 2- to 10-mm lesions which appeared to be vesicular (Fig. 1A), with a faintly erythematous base, coalescing on the shoulders and back. There were also some lesions on the anterior chest, a few lesions on her scalp, and rare scattered lesions on her extremities. The lesions were synchronous and exquisitely tender. Complete blood count, routine chemistry analyses, liver function tests, blood cultures, and chest radiograph were unremarkable. An HIV test was negative. Nucleic acid was extracted from biopsies of the skin lesions. At the time of biopsy, it became evident that the lesions were in fact papules and nodules. With the use of protocols validated by the Centers for Disease Control and Prevention (CDC), polymerase chain reaction was performed for variola, vaccinia, varicella, and generic Orthopoxvirus. All polymerase chain reaction tests were negative, as was a direct fluorescent antibody staining for herpes simplex virus and varicella. No evidence of viral particles was seen with transmission electron microscopy.
Histopathologic evaluation of a trephine biopsy specimen from the skin of the back revealed subepidermal vesiculation secondary to marked papillary dermal edema (Fig. 1B). There were dense, predominantly neutrophilic, infiltrates in the upper dermis with prominent leukocytoclastic nuclear debris. Scattered eosinophils were present, but viral cytopathologic features and ballooning degeneration of the epidermis were not identified. Based on the characteristic histopathology, a diagnosis of Sweet's syndrome was made. The patient's oral contraceptive pills were discontinued, and she was discharged home on a prednisone taper. Lesions recurred as the steroids were tapered off, and 2 months after her initial presentation, the patient continues with fever, myalgias, and arthralgias.
The poxviruses are a complex group of DNA viruses that have a propensity to cause cutaneous lesions.1,2 They can be divided into several genera, 4 of which are known to infect humans: Orthopoxvirus, Parapoxvirus, Molluscipoxvirus, and Yatapoxvirus. The genus Orthopoxvirus includes at least 9 closely related species, which are the agents of variola, vaccinia, cowpox, and monkeypox, to name only a few.2 The only specifically human poxviruses are variola virus (smallpox) and Molluscipoxvirus, which causes molluscum contagiosum. Variola no longer exists in an indigenous state in nature, the last reported case having occurred in Somalia in 1977.3 The rash usually progresses in a uniform pattern, from maculopapules to vesicles to pustules and finally scabs. The current vaccinia virus, which is used to immunize against smallpox, is actually thought to be derived from the cowpox virus4 and is highly protective against smallpox. Complications related to vaccination are well known and include postinfectious encephalitis, dermal complications (vaccinia necrosum, eczema vaccinatum, accidental autoinoculation, generalized vaccinia), arthritis, myocarditis, pericarditis, and thrombocytopenia.1
Zoonotic poxvirus infections produce cutaneous lesions very similar to those caused by human poxviruses, often in association with lymphadenopathy and systemic symptoms.5 They are usually acquired through cuts, abrasions, or bites after direct contact with an infected animal.3 In most instances, lesions are few in number, although this is not the case with monkeypox, in which there is widespread centrifugal blistering similar to the rash of smallpox.1 We considered monkeypox as a possible cause of this patient's rash, although her only reported animal exposure was to the family's pet dog. Monkeypox is a vesicular illness that is very similar to variola.8 It is enzootic among squirrels and monkeys in the rainforests of western and central Africa. Infection occurs only sporadically in humans living in these areas, but it carries significant morbidity and childhood mortality.9 The 2003 US monkeypox outbreak in prairie dogs was caused by imported African rodents and led to renewed interest in poxviruses and antipox viral therapies.6 Tanapox virus (Yatapoxvirus genus) behaves clinically like monkey pox, although only 5 documented cases have been reported in the United States since the 1960s.10 We felt that the patient's rash was too extensive to be compatible with a Parapoxvirus infection, and the patient had no history of exposure to cattle.5 When these viruses infect humans, there is usually a history of direct contact with the originally infected species or with contaminated objects (fences, posts, feeding troughs). The lesions are typically found on the hands or arms and tend to be painful and pruritic.5 Ecthyma contagiosum (orf) and milker's nodules (pseudocowpox) are clinically identical, distinguished only by the source of infection (sheep and goats vs. cows). Cowpox is an Orthopoxvirus, which despite its name, is not common in cattle. The most commonly detected source for human infection is the domestic cat, and it now appears to occur only in Europe and the former Soviet Union.4
Although the patient's rash was somewhat reminiscent of smallpox, it was not centrifugal, and the patient appeared much less toxic than we would expect given her degree of skin involvement. It also became apparent at the time of the skin biopsy that the lesions were nodules, not vesicles, as is typically the case with smallpox. However, in view of the current concern over the possible use of variola virus for biologic warfare, we originally considered smallpox as a possible diagnosis.6,7 The patient was maintained in airborne and contact precautions until smallpox was excluded. Several employees who had been in contact with the patient were concerned about potential exposure to smallpox or some other Orthopoxvirus infection and were reluctant to go home for fear of exposing their families. Because smallpox and other Orthopoxvirus infections have an incubation period of 10 to 15 days, health care workers were reassured that they could safely return home.1 Smallpox is, in fact, a relatively noncontagious disease that requires close contact for spread and can be easily contained provided that all contacts are identified and vaccinated.3
The case was immediately reported to the Sacramento and Merced County Health Department, the California Department of Health Services, and the CDC. We used the CDC algorithm to evaluate the patient for smallpox (http://www.bt.cdc.gov/agent/smallpox/training/overview/pdf/differentialdiagnosis.pdf), which yielded in our hands a "high" likelihood. State and CDC officials obtained a "moderate" likelihood of smallpox. In 2002, the CDC Rash Illness Response Team investigated 25 cases of possible smallpox. Twenty-one met low-risk criteria, and 4 met moderate-risk criteria. Histological review of the skin biopsies determined that the lesions were consistent with Sweet's syndrome, a rare disorder originally described in 1964 by Dr Robert Douglas Sweet as "acute febrile neutrophilic dermatosis."
Sweet's syndrome is characterized by an acute febrile illness, leukocytosis, and erythematous plaques infiltrated by neutrophils.11 Skin lesions usually present as painful red or purple-red papules or nodules, which tend to coalesce and form irregular, sharply bordered plaques. Lesions usually appear on the upper extremities, face, and neck but can also develop on the trunk and lower limbs. Because of pronounced edema in the upper dermis, many lesions that are solid on palpation have a transparent vesiclelike appearance, which can be mistaken for pox and other viral infections. Lesions may also mimic those of erythema nodosum, pyoderma gangrenosum, and erythema multiforme. Sweet's syndrome is often classified based on the clinical setting in which it occurs: classical idiopathic (often associated with an underlying infection, inflammatory bowel disease, and pregnancy), malignancy-associated (acute myelogenous leukemia,breast/genitourinary/gastrointestinalcarcinomas), and drug-induced (granulocyte colony-stimulating factor, trimethoprim-sulfamethoxazole, nitrofurantoin, tetracycline, oral contraceptive pills, antiepileptics, and antihypertensives) Sweet's syndrome. Associated conditions include bacterial (streptococcus, mycobacterium, Yersinia, typhus, and Salmonella) and viral (cytomegalovirus, chronic hepatitis, and HIV) infections. The primary therapy for Sweet's syndrome is systemic corticosteroids.12 The oral cephalosporin that our patient was prescribed early in her illness may have caused her to develop Sweet's syndrome, although there is no prior report of such an association. Other possible etiologic agents include the contraceptive pills that she was taking before admission, an unidentified upper respiratory tract infection virus, or a yet-to-be-diagnosed underlying malignancy.
In view of the grave consequences that would likely follow the intentional reintroduction of smallpox, clinicians must maintain a high degree of suspicion. Such attention will also help physicians to recognize zoonotic poxvirus infections, which will likely increase in prevalence because of the importation of exotic animal species and rapid international travel. Sweet's syndrome, although rare, should be included in the differential diagnosis of patients presenting with a poxlike rash.
The authors thank Hien Nguyen for the clinical photograph and Stuart Cohen for review of the manuscript.
1. Diven DG. An overview of poxviruses. J Am Acad Dermatol
2. Moss B. Poxviridae: the viruses and their replication. In: Fields BN, Knipe DM, Howley PM, eds. Fields Virology
. 4th ed. vol 2. Philadelphia, Pa: Lippincott, Williams and Wilkins; 2001:2849-2883.
3. Fenner F, Henderson DA, Arita I, et al. Human monkeypox and other poxvirus infections of man. In: Smallpox and Its Eradication
. Geneva, Switzerland: World Health Organization; 1988:1287-1320.
4. Baxby D, Bennett M. Cowpox: a reevaluation of the risks of human cowpox based on new epidemiological information. Arch Virol
5. Lewis-Jones S. Zoonotic poxvirus infections in humans. Curr Opin Infect Dis
6. Besser JM, Crouch NA, Sullivan M. Laboratory diagnosis to differentiate smallpox, vaccinia, and other vesicular/pustular illnesses. J Lab Clin Med
7. Hazelton PR, Gerderblom HR. Electron microscopy for rapid diagnosis of infectious agents in emergent situations. Emerg Infect Dis
8. Reed KD, Melski JW, Graham MB, et al. The detection of monkeypox in humans in the western hemisphere. N Engl J Med
9. Di Giulio DB, Eckburg PB. Human monkeypox: an emerging zoonosis. Lancet Infect Dis
10. Dhar DA, Werchniak AE, Li Y, et al. Tanapox infection in a college student. N Engl J Med
11. Von den Driesch P. Sweet's syndrome (acute febrile neutrophilic dermatosis). J Am Acad Dermatol
12. Cohen PR, Kurzrock R. Sweet's syndrome: a neutrophilic dermatosis classically associated with acute onset and fever. Clin Dermatol