Tularemia with Vesicular Skin Lesions May Be Mistaken for Infection with Herpes Viruses

Most physicians associate tularemia with ulcerative lesions due to local adenopathy. However, the original reports of tularemia noted vesicular skin rash as a manifestation [1, 2]. Bacterium tularense, a gram-negative, pleomorphic rod, was first identified as the cause of the disease in humans in 1914 by Wherry and Lamb [3]. Utah physicians were among the first to recognize human tularemia in the early 20th century. What is now recognized as the first description of human cases of tularemia in the United States was published in 1911 by Pearse, a physician from Brigham City, Utah, who described multiple cases associated with deerfly bites and who named the disease “deer fly fever” [2]. Elizabeth Tracy of Woodrow, Utah, also described a similar illness and alerted the US Public Health Service [4].

Edward Francis was dispatched by the US Public Health Service to Utah to investigate the illness. He isolated the causative bacterium and identified the animal reservoirs. He demonstrated the mode of transmission to humans through direct contact and via vectors in Utah in the early 1920s [1, 5]. He further described multiple cases of human disease in Utah, including his own [1, 5, 6].

In a summary of tularemia published in 1928, Francis described the skin findings in his review of 654 cases. He noted that a skin eruption was present in 32 cases, and “extreme herpes was noted in one case” [7, p. 425]. We present 2 cases of pediatric tularemia that were initially diagnosed as infection with herpes simplex virus (HSV) or varicella zoster virus (VZV). Clinicians must recognize the atypical cutaneous manifestations of tularemia, as described in the past, and be able to distinguish those from lesions seen with herpes viruses.

Methods. The Institutional Review Board of the University of Utah (Salt Lake City) approved this study. Patients were identified by the pediatric infectious diseases consult service at Primary Children's Medical Center (Salt Lake City). Laboratory identification was based on culture from the patients' blood, eschars, and vesicular lesions. The medical records were reviewed and abstracted.

Wound cultures were processed in the Primary Children's Medical Center microbiology laboratory with use of standardized culture media (MacConkey agar, sheep blood agar, chocolate agar, colistin nalidixic acid agar, and thioglycolate broth). Phenotypic identification of Francisella tularensis was based on Gram-stain results and rapid gram-negative identification, as described by the Centers for Disease Control and Prevention (CDC) tularemia flow chart for Level A Sentinel Laboratories (Laboratory Response Network, Bioterrorism Preparedness Response Program; CDC). All isolates were confirmed by the Utah Public Health Laboratory.

Case reports. Case 1 involved a 6-week-old infant who was hospitalized during July 1998 for evaluation of fever and a vesicular rash of the foot (figure 1A). The initial diagnosis was HSV infection. Specimens from the skin lesion were submitted for viral culture, and blood and CSF were tested by PCR for HSV. Intravenous administration of acyclovir was initiated. Results of all laboratory tests for HSV were negative. The infant remained febrile after 48 h of acyclovir therapy, and the vesicular fluid became turbid. Infectious diseases consultation was requested.

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Figure 1

A, Photograph of a 6-week-old infant with vesicular tularemia initially diagnosed as herpes simplex infection. Results of complete herpes evaluation were negative. B and C, Photographs of a 10-year-old child with vesicular lesions on arms and legs thought to be varicella. The arrow in panel B indicates a vesicle near the primary eschar. Evaluation was negative for varicella and herpes infection. Culture of the eschar and vesicles confirmed tularemia. D, Varicella lesions, shown for comparison.

The consultant elicited a history of deerfly bite to the foot during a camping trip 4 days before hospital admission. An evaluation for bacterial infection was performed, including bacterial cultures of blood, CSF, and vesicular fluid. Blood and vesicular cultures yielded F. tularensis. The patient was treated with gentamicin and clarithromycin for 10 days. The patient responded well but, 2 weeks after completing therapy, had recurrence of fever and lesions on the foot. The patient was readmitted to the hospital and was treated with rifampin and gentamicin for 2 weeks, with complete resolution. Evaluation for immune deficiency, including chronic granulomatous disease, revealed no abnormalities.

Case 2 involved a 10-year-old child who was admitted to the hospital in August 2006 with a 9-day history of fever, vesicles on the extremities and face, an eschar surrounded by vesicles on the left arm (figure 1B and 1C), and a possible seizure on the day of hospital admission. The patient had been evaluated for the same illness in an outpatient setting and had been given a diagnosis of VZV infection. The diagnosis at hospital admission was “complicated varicella.” The patient had received varicella immunization at age 15 months. Infectious diseases and dermatology consultations were requested, and an evaluation for tularemia and varicella was performed. The patient was treated with doxycycline, gentamicin, and acyclovir.

Just before the illness, the patient reported multiple fly bites sustained while working in a livestock barn. Bacterial cultures of blood, vesicular fluid, and eschar specimens yielded F. tularensis. CSF bacterial culture was sterile. All studies for VZV yielded negative results. The patient completed 2 weeks of antibiotic therapy with gentamicin and doxycycline for tularemia, with no recurrence of symptoms.

Discussion. Although the causative organism of tularemia was initially isolated by McCoy in 1911 in Tulare County, California, and described as being the cause of a “plague-like disease of rodents” [8], the human disease now known as tularemia was described by Edward Francis, building on the observations of Pearse, Tracy, and other physicians in Utah [2, 4]. The 2 cases described in this report serve to present to physicians one of the atypical cutaneous manifestations of tularemia—specifically, vesicles—as originally described in the early 20th century by Pearse, Tracy, and Francis [2, 4].

Tularemia is a zoonotic infection caused by F. tularensis. Although it most commonly causes ulceroglandular disease, more-severe forms include pneumonic, typhoidal, and meningitic tularemia. Transmission is primarily through direct contact with infected animals, such as rabbits (members of the Leporidae family), or through vectors such as ticks (Ioxidae) and biting flies (Tabanidae).

F. tularensis has been designated a Category A bioterrorism agent by the CDC because of its unique ability to disseminate and its potential use as an agent of bioterrorism [9]. Clinicians need to be aware of the potential for F. tularensis to be associated with acts of bioterrorism. Aerosol dissemination of F. tularensis would probably result in pneumonic disease, but cutaneous lesions associated with bacteremia might also result from an intentional release of the bacterium. Several agents of bioterrorism commonly present with cutaneous manifestations—smallpox presents with vesicular lesions, and cutaneous anthrax presents with eschars but no vesicles. Tularemia could be mistaken for both conditions. Because of the threat of biological terrorism, the astute clinician must be aware of the possibility of tularemia presenting with vesicular skin lesions and fever.

Although F. tularensis may be used as an agent of bioterrorism, the organism is endemic to the United States, and infection is more likely to be a result of natural exposure. In a review of tularemia cases, 44 states reported cases to the CDC during 1990–2000 [10], the same geographic distribution noted by Edward Francis in 1928 [7]. Fewer than 200 cases are reported to the CDC each year [10]. The Utah Department of Health reports 2–3 cases per year, with an average rate of 0.12 cases per 100,000 population, compared with the national average of 0.03 cases per 100,000 population. Tularemia has also been linked to outbreaks. In 1935, 30 cases were reported in a Civilian Conservation Corps camp in Northern Utah [11]. More-recent outbreaks include 1 on Martha's Vineyard in 2000 in which 11 patients, primarily landscapers, developed severe pneumonic tularemia [12]. In the summer of 2007, the Utah Department of Health investigated a cluster of 14 tularemia cases associated with camping near Utah Lake [13].

Because tularemia is rarely reported in the United States, individual clinicians may be unfamiliar with even the classic findings of tularemia. Appreciation of the more rare manifestations of the disease, including vesicular skin lesions, is, in our experience, infrequent. Pearse described “water blisters” associated with infection following the bite of the deerfly [2]. Francis also described vesicular lesions [7]. A study of cases seen in Tennessee in the 1940s described an associated erythematous maculopapular skin rash [14]. Rash was also noted by investigators in Finland who reported a study of tularemia during 1967–1983; of the 88 patients included, 37 (42%) had evidence of a papular or vesicular skin eruption. [15].

Distinguishing between herpes virus infections and tularemia can be difficult by visual inspection alone, as demonstrated in figure 1. A history of summer outdoor activity with insect exposure, especially biting flies or ticks, is suggestive of tularemia. The vesicles associated with tularemia initially contain clear fluid but may become turbid, as shown in figure 1A, suggesting bacterial infection. Vesicles surrounding an ulcerated eschar (figure 1B) are more suggestive of tularemia than of herpes virus infection. Furthermore, patients with tularemia have persistently high fevers and lymphadenopathy—features that are less common with viral infections. Bacterial cultures of eschar, vesicular fluid, and blood specimens are helpful in establishing the diagnosis.

The original reports of human tularemia described vesicular skin rash as a manifestation, but this presentation may lead to delayed diagnosis [7]. Today, tularemia continues to be an endemic disease in the United States, and F. tularensis is a possible bioterrorism agent. Therefore, it is important that clinicians recognize that the cutaneous manifestations of tularemia include vesicular lesions and that they be able to distinguish these from lesions seen with herpes viruses.

• Received December 12, 2007.
• Revision received February 20, 2008.

• © 2008 by the Infectious Diseases Society of America