Specimen Collection, Transport, and Processing

The most common specimens submitted to the laboratory are scrapings from infected ulcers, lymph node biopsies, and sputum. Whole blood is an acceptable specimen for all types of tularemia; however, false-negative results may occur during early stages of disease. Serum is generally collected from all patients early in disease and during convalescence. The blood should be separated from the serum as soon as possible, preferably within 24 hours, and may be stored at 2° to 8°C for up to 10 days. If long-term storage is required, the serum may be frozen. To minimize the loss of viable organisms, samples should be transported to the laboratory within 24 hours. If specimens are to be held longer than 24 hours, specimens should be refrigerated in Amie’s transport medium. F. tularensis should remain viable for up to 7 days stored at ambient temperature in Amie’s medium. Swab specimens should be placed in Amie’s transport media containing charcoal. Specimens for molecular testing should be placed in guanidine isothiocyanate buffer for up to 1 month.

Specimen collection for the identification of F. tularensis is highly dependent on the type of clinical manifestation. A detailed description of the recommended type of specimen associated with the patient’s clinical presentation is presented in Table 38-3. In light of recent events and concerns about bioterrorism, laboratories must keep in mind that isolation of F. tularensis from blood cultures might be considered a potential bioterrorist attack; F. tularensis is considered one of the Select Biological Agents of Human Disease (see Chapter 80).

Direct Detection Methods

Gram staining of clinical material is of little use with primary specimens unless the concentration of organisms is high, as in swabs from wounds or ulcers, tissues, and respiratory aspirates. The organisms tend to counterstain poorly with safranin. Replacing safranin with basic fuchsin may enhance identification. Fluorescent antibody stains and immunohistochemical stains are commercially available for direct detection of the organism in lesion smears and tissues and are typically available in reference laboratories. Conventional and real-time polymerase chain reaction (PCR) assays have been developed to detect F. tularensis directly in clinical specimens. Of significance, several patients with clinically suspected tularemia with negative serology and culture had detectable DNA by PCR. Currently most PCR-based assays are unable to discriminate F. tularensis from F. novicida, which limits the value of the epidemiologic data.

Cultivation

Isolation of F. tularensis is difficult. The organism is strictly aerobic and is enhanced by enriched media containing sulfhydryl compounds (cysteine, cystine, thiosulfate or IsoVitaleX) for primary isolation. Two commercial media for cultivation of the organism are available: glucose cystine agar (BBL; Microbiology Systems, Sparks, Maryland) and cystine-heart agar (Difco Laboratories, Detroit, Michigan); both require the addition of 5% sheep or rabbit blood. F. tularensis also may grow on chocolate agar supplemented with IsoVitaleX, the nonselective buffered charcoal-yeast extract agar (BCYE) used for isolation of legionellae, or modified Mueller-Hinton broth and tryptic soy broth supplemented with 1% to 2% IsoVitaleX. Growth is not enhanced by carbon dioxide.

These slow-growing organisms require 2 to 4 days for maximal colony formation; they are weakly catalase positive and oxidase negative. Some strains may require up to 2 weeks to develop visible colonies. F. philomiragia is less fastidious than F. tularensis. Although F. philomiragia does not require cysteine or cystine for isolation, it is similar to F. tularensis in that it is a small, coccobacillary rod that grows poorly or not at all on MacConkey agar. This organism grows well on heart infusion agar with 5% rabbit blood or BCYE agar with or without cysteine. F. tularensis can be detected in commercial blood culture systems in 2 to 5 days; because these organisms Gram stain poorly, an acridine orange stain may be required to visualize the organisms in a positive blood culture bottle.

Approach to Identification

Colonies are transparent, mucoid, and easily emulsified. Although carbohydrates are fermented, isolates should be identified serologically (by agglutination) or by a fluorescent antibody stain. Ideally, isolates should be sent to a reference laboratory for characterization.

F. philomiragia differs from F. tularensis biochemically; F. philomiragia is oxidase-positive by Kovac’s modification, and most strains produce hydrogen sulfide in triple sugar iron agar medium, hydrolyze gelatin, and grow in 6% sodium chloride (no strains of F. tularensis share these characteristics).

In previous reports, problems have been identified in association with Francisella species isolated from clinical specimens. Twelve microbiology employees were exposed to F. tularensis even though bioterrorism procedures were in place; the organism had been isolated from blood, respiratory, and autopsy specimens and grew on chocolate agar. In this situation, multiple cultures were worked up on open benches without any additional personal protective equipment for what had been thought to be most consistent with a Haemophilus species. As a result of this report, microbiologists must be aware of not only the key characteristics of this group of organisms (Box 38-1), but also the possible pitfalls in their identification (e.g., some strains grow well on sheep blood agar; identification kits may incorrectly suggest an identification of Actinobacillus actinomycetemcomitans). If F. tularensis is suspected, all culture Petri dishes should be taped from the top to the bottom in two places to keep them together for safety purposes.

Serodiagnosis

Because of the risk of infection to laboratory personnel and other inherent difficulties with culture, diagnosis of tularemia is usually accomplished serologically by whole-cell agglutination (febrile agglutinins or newer enzyme-linked immunosorbent assay techniques). Serum antibody detection is useful for all forms of tularemia. After the initial specimen, a convalescent sample should be collected at 14 days and preferable up to 3 to 4 weeks after the appearance of symptoms. A fourfold difference in titers in acute versus convalescent phase serum samples, in conjunction with one additional positive diagnostic test, such as culture or molecular tests, is considered a presumptive diagnosis for tularemia.