Challenges to the Laboratory Diagnosis of Yersinia pestis in the Clinical Laboratory
By Luciana L. Borio, M.D., December 29, 2005
The CBN editors would like to thank Drs. Moon Kim, Lee Borenstein, David Dassey (Los Angeles County Department of Health Services), and Michael Ascher (Lawrence Livermore National Laboratory) who provided the information for this report, and Dr. Vee J. Gill (Suburban Hospital, Bethesda, MD) who critically reviewed this report.
Recently, a 72 year old man presented to a hospital complaining of nausea, vomiting, diarrhea, fever and weakness for 2 days. He was admitted with a presumptive diagnosis of viral gastroenteritis. He developed sepsis and was transferred to the intensive care unit. Chest radiographs showed no evidence of pneumonia. CT scans of his chest, abdomen, and pelvis were unrevealing. He was treated empirically with gatifloxacin and ceftriaxone.
Blood cultures obtained two days after admission grew gram-negative rods. Two commercial biochemical systems identified the organism as Aeromonas spp. and Yersinia pseudotuberculosis. The isolate was then sent to a reference laboratory which undertook Gram stain, subcultures, and 16S sequence analysis; results were strongly suggestive of Y. pestis. The reference laboratory then submitted the specimen to a Laboratory Response Network (LRN) lab for confirmation. At the LRN lab, direct fluorescent antibody (DFA), PCR tests, bacteriophage lysis, and serology confirmed the diagnosis of Y. pestis.
On further review of the history prior to clinical illness, the patient reported crawling underneath his house to work on pipes where rodents were present. Subsequent blood cultures did not grow Y. pestis and he is now recovering.
This patient did not have signs of a bubo or pneumonia on presentation. In a small number of patients who are infected by fleas, septicemia may develop in the absence of a bubo; this presentation is termed primary septicemic plague (Inglesby TV et al. Plague as a Biological Weapon. JAMA 2000 May;283:2281-2290). The absence of these classical features made the clinical diagnosis more challenging. In such cases, an accurate and timely diagnosis relies heavily on the microbiology laboratory, and any suspicion of plague requires prompt forwarding of samples to reference laboratories and notification of public health authorities. Such prompt notification is important because there are times when the administration of antimicrobial prophylaxis to individuals having close contact with an ill, untreated person may be warranted.
Yersinia pestis is a nonmotile, slow-growing, aerobic and facultative anaerobic organism classified in the family Enterobacteriaceae. It appears as plump, gram-negative coccobacilli that are seen mostly as single cells or pairs and short chains in liquid media, and which may exhibit bipolar staining (better visualized if either Wright-Giemsa or Wayson’s stain is used). Basic protocols for the presumptive identification of Y. pestis have been developed by the CDC, ASM, and APHL (http://www.asm.org/ASM/files/LEFTMARGINHEADERLIST/DOWNLOADFILENAME/0000000524/plagueprotocol%5B1%5D.pdf). However, once Y. pestis is suspected, the organism should be preserved, all further work suspended, and the state public health laboratory/department should be consulted immediately.
Y. pestis grows slowly in culture, with only pinpoint, gray-white translucent colonies on SBA after 24 hours, with little or no hemolysis. After 48 to72 hours of incubation, colonies have a raised, irregular "fried egg" appearance under magnification. Colonies are at times described as having a "hammered copper" shiny surface. Y. pestis grows as small, non-lactose fermenting colonies on MAC or EMB agar.
In broth, Y. pestis grows in clumps typically described as flocculant when the broth culture is not shaken. At 24 hours, the growth forms clumps that hang along the side of the tube. After 24 hours the growth settles to the bottom of the tube and is described as “cotton fluff.” Biochemical tests are negative for oxidase and urease, positive for catalase, and negative for indole. Automated identification systems may be misleading and report results as non-Y. pestis Enterobacteriaceae (e.g. Shigella, H2S-negative Salmonella, Acinetobacter or Y. pseudotuberculosis). (Wilmoth BA et al. Identification of Yersinia pestis by BBL Crystal Enteric/Nonfermenter Identification System. J Clin Microbiol. 1996 Nov;34:2829-30.)
In summary, isolates revealing the following characteristics should be suspected as Y. pestis and a reference laboratory should be consulted immediately:
Bipolar staining rod (Wright-Giemsa) on direct smear
Pinpoint colony at 24 h on SBA
Non-lactose fermenter, may not be visible on MAC or EMB at 24 h
Oxidase and urease negative
Growth often better at 28oC
Clinicians should be aware of the limitations of commercial identification systems, which rely on pre-determined algorithms that may not include differential tests adequate for accurate identification of less common pathogens, such as Y. pestis and F. tularensis. Clinicians should maintain open lines of communication with clinical microbiology laboratories to facilitate timely and accurate diagnosis of unusual infections. In addition, laboratory notification allows for the implementation of the laboratory’s biosafety plans. In a bioterrorism event, time to diagnosis is critical. Thus, it is also important for clinical laboratories to keep public health laboratories informed to avoid delays that may have serious public health implications.
FDA Approves Tamiflu for the Prevention of Influenza in Children Under Age 12
On 12/21/2005, the FDA approved the use of Tamiflu (oseltamivir phosphate) for prevention of seasonal influenza in children 1-12 years of age who had close contact with an infected individual (http://www.fda.gov/bbs/topics/news/2005/NEW01285.html). This is the 1st drug approved for prevention of both influenza A and B in children. Prior to this, Tamiflu was approved only for treatment of influenza in this age group.