Plague: Current, comprehensive information on pathogenesis, microbiology, epidemiology, diagnosis, and treatment

Last updated December 17, 2008

Agent
Pathogenesis
Epidemiology
—Reservoirs/Vectors/Modes of Transmission
—Historical Perspective
—Naturally Occurring Plague in the United States
—Naturally Occurring Plague Worldwide
—Plague as a Biological Weapon
Clinical Syndromes
—Overview
—Bubonic Plague
—Septicemic Plague
—Pneumonic Plague
Pediatric Considerations
Differential Diagnosis
Laboratory Diagnosis
—Specimen Collection and Transport
— Laboratory Biosafety Information
—Laboratory Response Network
—Standard Tests for Detection of Y pestis
—Additional Tests for Detection, Confirmation, and Characterization of Y pestis
—Antimicrobial Susceptibility Studies
—Environmental Testing
Postexposure Prophylaxis for Pneumonic Plague
Treatment
—General Considerations
—Pneumonic Plague
—Bubonic Plague
Plague Vaccines
Hospital Infection Control
Issues Related to Autopsies and Burial
Public Health Reporting and Case Definitions
Images
References

Agent

Key microbiologic characteristics of Yersinia pestis include the following (see References: CDC/ASM/APHL: Basic protocols for level A laboratories for the presumptive identification of Yersinia pestis; Sneath 1986):

Pleomorphic gram-negative bacillus (1.0 to 2.0 mcm x 0.5 mcm); single cells or short chains in direct smears
Bipolar ("closed safety pin") staining with Giemsa, Wright's, or Wayson stains (may not be visible on Gram stain)
Facultative anaerobe
Nonmotile, nonsporulating
Non–lactose fermenter
Slow-growing in culture (colonies are pinpoint after 24 hours on sheep blood agar [SBA] and much smaller than other Enterobacteriaceae growing for 24 hours on SBA; colonies may not be visible on MacConkey or eosin methylene blue agar at 24 hours)
Catalase-positive, oxidase- and urease-negative (rarely, strains may be urease-positive)
Optimal growth at 28°C
"Stalactite pattern" in broth culture with clumps of cells from the side of the tube settling to the bottom if disturbed
At 48 to 72 hours of incubation on solid media, colonies have a raised, irregular, "fried egg" appearance under 4x enlargement, which becomes more pronounced as the culture ages; colonies also have been described as having a "hammered copper" shiny surface
Alkaline slant/acid butt (K/A) on triple sugar iron agar (TSI) without gas or H₂S
Generally susceptible to tetracyclines, chloramphenicol, aminoglycosides, sulfonamides (with or without trimethoprim), and fluoroquinolone antibiotics

Y pestis is divided into three classic biovars (see References: Dennis 1997).

Biovar antiqua (Africa, southeastern Russia, central Asia; thought to be the cause of the first pandemic)
Biovar medievalis (Caspian Sea; thought to be the cause of the second pandemic)
Biovar orientalis (Asia, Western Hemisphere; cause of the third pandemic)

Other classification and diversity information:

A nonvirulent strain, microtus, has been proposed as a fourth biovar (see References: Zhou 2004).
Y pestis is thought to have evolved from Yersinia pseudotuberculosis 1,500 to 20,000 years ago, and the two species remain closely related (see References: Achtman 2000). Whole-genome sequence comparisons have identified 32 chromosomal genes and 2 plasmids in Y pestis but not Y pseudotuberculosis (see References: Chain 2004).
The complete genomes of several strains have been sequenced and are available online (see References: National Center for Biotechnology Information, Parkhill 2001, Song 2004, Zhou 2002).

Pathogenesis

Virulence Factors

Virulence factors for Y pestis are primarily encoded on the chromosome and on three plasmids (the Pst plasmid, the Lcr plasmid, and the pFra plasmid) (see References: Dennis 1997).

The major virulence factors for Y pestis are responsible for the following activities (see References: Dennis 1997, McGovern 1997, Perry 1997, Titball 2003):

The ability of Y pestis organisms to adhere to cell surfaces is a key step in pathogenesis. Irreversible binding to host cell receptors via adhesins allows the organisms to then penetrate the cell surfaces (see References: Zhou 2006).
The F1 antigen is antiphagocytic, elicits a humoral response, and is a target for immunologic-based diagnostic tests. Most pathogenic Y pestis strains isolated from humans contain the F1 antigen.
Plasminogen activator (Pla) is a protease that appears to degrade fibrin and other extracellular proteins and to facilitate systemic spread from the inoculation site. Expression of Pla allows Y pestis to replicate rapidly in the airways. Pla is essential for Y pestis to cause primary pneumonic plague but is less important for dissemination during pneumonic than bubonic plague (see References: Lathem 2007).
The V and W antigens (produced at 37°C) cause the organisms to be resistant to phagocytosis; the V antigen is important for survival of Y pestis in macrophages.
Yersinia outer proteins (Yops) have a variety of activities, including inhibiting phagocytosis, inhibiting platelet aggregation, and preventing an effective inflammatory response.
Lipopolysaccharide (LPS) endotoxin (encoded on the chromosome) causes the classic features of endotoxic shock. LPS consists of three domains: the hydrophobic membrane anchor (lipid A), the surface-exposed O-antigen polysaccharide, and the core sugar region connecting the other two. Most of the effects of LPS are caused by lipid A (LPS-lipid A) (see References: Zhou 2006).
Phospholipase D (PLD) allows the bacilli to survive in the flea midgut.
Yersinia murine toxin (Ymt) is one of the factors required for maintaining Y pestis in fleas. Ymt is highly toxic for mice and rats but less active in other animals (see References: Zhou 2006).

Bubonic Plague

After a flea initially ingests Y pestis, the organisms elaborate a coagulase that clots ingested blood in the proventriculus (an organ between the esophagus and stomach) of the flea, thus blocking passage of the next blood meal into the flea's stomach. Fleas with this blockage regurgitate Y pestis into the bite wound while attempting to feed (see References: Perry 1997).

In a recent study, unblocked fleas given a single infectious blood meal transmitted Y pestis for up to 7 days following the meal and fleas given a "booster" infectious blood meal 5 days after the first meal transmitted Y pestis for the full 9-day duration of the study (see References: Eisen 2007: Temporal dynamics of early-phase transmission of Yersinia pestis by unblocked fleas).
Such research may explain mathematical models that predict rapidly spreading epizootics and epidemics (see References: Eisen 2006: Early-phase transmission), as well as discrepancies observed in plague epizootics among prairie dogs (see References: Webb 2006).

Between 25,000 and 100,000 Y pestis organisms are inoculated into the skin via the bite of an infected flea (see References: Reed 1970).
A papule, vesicle, pustule, or furuncle may occur at the site of the fleabite but is noted in less than 10% of patients (see References: Dennis 1997).
The organisms migrate through the cutaneous lymphatics to regional lymph nodes. Comparative studies in mice reveal that Y pestis virulence is associated with a distinct ability to massively infiltrate the draining lymph node without inducing an organized polymorphonuclear cell reaction (see References: Guinet 2008).
Once in the lymph nodes, they are phagocytized by polymorphonuclear leukocytes (PMNs) and mononuclear phagocytes. Organisms that are phagocytized by PMNs generally are destroyed, whereas those phagocytized by mononuclear cells proliferate intracellularly and develop resistance to further phagocytosis (see References: Perry 1997). These organisms are released when cell lysis occurs.
Initially, a thick, proteinaceous exudate that includes plague bacilli, PMNs, lymphocytes, and fewer macrophages can be found in affected nodes (see References: Dennis 1997).
Subsequently, the exudative pattern gives way to lakes of hemorrhagic necrosis, which obliterate the underlying lymph node architecture. A ground-glass amphophilic material that represents masses of bacilli may be present (see References: CDC: Medical examiners, coroners, and biologic terrorism: a guidebook for surveillance and case management).
The inflammatory process creates swollen painful buboes and surrounding edematous tissues that are characteristic of bubonic plague.
The organisms often enter the bloodstream, causing hemorrhagic lesions in other lymph nodes and in organs throughout the body (initially the liver and spleen). Findings from a recent study using a mouse model suggest that the organisms replicate in splenic macrophages during the later stages of infection (see References: Lukaszewski 2005).
Septicemia, disseminated intravascular coagulation (DIC), and shock can ensue.
Unless treated promptly with appropriate antibiotic therapy, death usually results from overwhelming sepsis.

Pneumonic Plague

Y pestis can enter the lungs either through direct inhalation (primary pneumonic plague) or through hematogenous spread as a complication of bubonic or septicemic plague (secondary pneumonic plague).
Primary pneumonic plague is acquired naturally by inhaling respiratory droplets from infected humans or animals (such as cats).
The infectious dose by inhalation is estimated to be 100 to 500 organisms (see References: Franz 1997).
Marked intra-alveolar edema and congestion of the lungs are common (see References: CDC: Medical examiners, coroners, and biologic terrorism: a guidebook for surveillance and case management). Pulmonary lesions include areas of central exudate with peripheral congestion. This pattern initially is lobular, but usually progresses to lobar consolidation (see References: Dennis 1997).
Distinguishing primary pneumonic plague from secondary hematogenous spread to the lungs can be difficult. Features that occur more commonly with primary pneumonic plague include the following (see References: Dennis 1997):

Tracheal and bronchial mucosal hemorrhages
Fibrinous pleuritis and subpleural hemorrhages overlying areas of exudative pneumonia
Less inflammation and necrosis and more exudation in lobular foci of the parenchyma
Foci of pneumonia along medium and large bronchi
More involvement of hilar lymph nodes
Less severe evidence of disease in organs other than the lungs, if such evidence is present

In primary pneumonic plague, as with bubonic plague, organisms often enter the bloodstream and cause multiorgan involvement, DIC, and shock.
In the absence of early antibiotic therapy (ie, within the first 24 hours), death occurs from overwhelming sepsis (usually within several days after illness onset). Without therapy, mortality approaches 100%.

Septicemic Plague

Primary septicemic plague is defined as systemic toxicity caused by Y pestis infection but without apparent preceding lymph node involvement. Secondary septicemic plague occurs commonly with either bubonic or primary pneumonic plague.
In primary septicemic plague, Y pestis organisms can disseminate from a fleabitesite through thelymphatic system (but without clinically apparent involvement of the lymph nodes), directly through the circulatory system, orboth (see References: Sebbane 2006).
Septicemic plague causes sepsis syndrome with multiorgan involvement, DIC, and shock. In the late stages of infection, high-density bacteremia often occurs, leading to identification of organisms on peripheral blood smears (see References: Butler 1991).
Spleen, liver, kidneys, skin, and brain are the most commonly affected organs. Meningitis can occur and is characterized by a thick, yellow, fibrinous-purulent exudate. Foci of necrosis with hemorrhage are common, as are characteristic lesions of DIC (such as fibrin thrombi in glomerular capillaries or purpuric skin lesions) (see References: Dennis 1997).

Y pestis may persist in necrotic tissues after antibiotic treatment despite negative blood cultures. Presumably, Y pestis becomes trapped in hypoperfused tissues and is able to persist because of: (1) inadequate delivery of antibiotics to affected areas and (2) the ability of the organisms to overcome local host defenses (see References: Guarner 2005).

Epidemiology

Reservoirs/Vectors/Modes of Transmission

Reservoirs

Animals (predominantly wild rodents) are considered to be the primary natural reservoirs for Y pestis. More than 200 mammalian species have been shown to be infected with Y pestis.

In addition, soil may be an important reservoir. In one model, soil serves as the ultimate reservoir, with burrowing animals as the first link in the chain of transmission, followed by spread to other animals and humans through ectoparasites (via fleaborne transmission) (see References: Drancourt 2006). Experimental studies document survival of Y pestis in soil from 24 days (see References: Eisen 2008: Persistence of Yersinia pestis) to 40 weeks (see References: Ayyadurai 2008); however, further study is needed to asses the role of soil in maintaining Y pestis in the natural environment (see References: Drancourt 2006).

Some animal populations are relatively resistant to the effects of Y pestis infection and serve as enzootic reservoirs (see References: Dennis 1997). For example, in one study, swift foxes seropositive for Y pestis were found to be spatially correlated with epizootic plague activity in prairie dog colonies (see References: Salkeld 2007).

Other animal species are more susceptible to disease caused by Y pestis and serve as epizootic hosts (see References: Butler 1991, Gage 1998). Examples include the following (see References: Dennis 1997, Gabastou 2000, Reed 1970):

Urban and domestic rats
Ground squirrels
Rock squirrels
Prairie dogs
Deer mice
Field mice
Gerbils
Voles
Chipmunks
Marmots
Guinea pigs
Kangaroo rats

Humans are incidental hosts for Y pestis and are not part of the natural life cycle of the organisms. Disease occurrence in humans is dependent on the frequency of infection in local rodent populations and the degree of contact between rodents and humans. Human outbreaks usually are preceded by epizootics with increased deaths in susceptible animal hosts (see References: Butler 1991, Perry 1997). Most human exposures to plague occur in the peridomestic environment, and free-roaming pets that bring infected rodent fleas into the home are considered to be a potential source of infection as well (see References: CDC: Human plague—four states, 2006).

Exposure to dogs was found to be a significant risk factor for plague among infected patients in New Mexico. Patients with plague were more likely to report having a sick dog or having slept in the same bed with a pet dog than were controls (see References: Gould 2008). Infected cats were the source of 7.7% of the 297 cases of plague in the United States from 1977 through 1998 (see References: Gage 2000).

In Africa, a study of fleas that have humans as their host (the "human flea," Pulex irritans) revealed that they may be an indicator of plague potential in rural areas and may play a role in plague epidemiology (see References: Laudisoit 2007). Cat fleas may also be secondary vectors for plague in Africa (see References Eisen 2008: Early-phase transmission of Yersinia pestis by cat fleas).

In another recent study, logistic regression models were used to identify landscape features associated with areas where humans have acquired plague from 1957 to 2004 in the four-corners region of the United States (Arizona, Colorado, New Mexico, and Utah). The overall accuracy of the model was >82% and the most conservative model predicted that 14.4% of the four-corners region represented a high risk area of peridomestic exposure to Y pestis (see References: Eisen 2007: Human plague in the southwestern United States, 1957-2004). Such information can be used to identify target areas for surveillance and control (see References: Eisen 2007: Residence-linked human plague in New Mexico).

Plague dynamics appear to be driven by climate variation and seasonal influences.

Studies of plague movement eastward from California show that factors such as climatic and environmental variables can influence spread (see References: Adjemian 2007).
Models of climate change suggest that over the next 50 years geographic shifts of zoonotic diseases such as tularemia and plague will occur. Plague disease ranges in the United States have the potential to expand from a current central focus in New Mexico north into Wyoming and Idaho (see References: Nakazawa 2007).
A recent study found that environmental conditions present during the emergence of recent plague epidemics (wetter springs and warmer summers) may be more common in the future and make outbreaks more likely in both endemic and new areas (see References: Stenseth 2008).
Field data from central Asia from 1949 to 1995 reveal that Y pestis prevalence in gerbils increases with warmer springs and wetter summers (see References: Stenseth 2006). A 1ºC increase during spring is predicted to yield a more than 50% increase in prevalence. Climatic conditions favoring plague are thought to have existed in this region at the onset of the Black Death (between 1280 and 1350) as well as during the third plague pandemic in the 1800s, which occurred in the same region. Furthermore, modeling studies suggest that plague outbreaks in central Asia may become more frequent, based on predicted climate change (see References: Kausrud 2007).
In a 3-year longitudinal study, serologic testing of gerbils in Kazakhstan suggested that there was a mid-summer peak in the abundance of infectious hosts and possible transmission from the reservoir to humans (see References: Begon 2006). Statistical analysis of dynamics of immune response in great gerbils shows that infection and recovery from plague are seasonal (see References: Park 2007).

Like humans, mammalian species other than rodents generally are incidental hosts for Y pestis. However, such animals also can serve as sources of human exposure (either through direct contact or through flea vectors). Examples of other animals that are susceptible to plague include the following (see References: Christie 1980, Dennis 1997, Gage 2000, Palmer 1971, Reed 1970, von Reyn 1976, Wild 2006):

Domestic and feral cats
Dogs
Lagomorphs (rabbits and hares)
Coyotes
Camels
Goats
Deer
Antelope
Lynx

Vectors

The organisms most commonly are transmitted between animal reservoirs and to humans via bites of infected fleas. In order to survive in the flea midgut, Y pestis organisms require phospholipase D (PLD; formerly referred to as Yersinia murine toxin), which allows the organisms to be resistant to a cytotoxic digestion product of blood plasma in the flea gut. A recent study demonstrated that Y pestis acquired the PLD gene at some point in the past, which allowed transformation from a rather benign species of gut bacteria to a major global pathogen (see References: Hinnebusch 2002).

Of the more than 1,500 flea species, about 30 are known to be vectors for Y pestis. Examples of major flea vectors include the following (see References: Perry 1997):

Xenopsylla cheopis (the oriental rat flea; nearly worldwide in moderate climates)
Oropsylla montanus (United States)
Nosopsyllus fasciatus (nearly worldwide in temperate climates)
Xenopsylla brasiliensis (Africa, India, South America)
Xenopsylla astia (Indonesia and Southeast Asia)
Xenopsylla vexabilis (Pacific Islands)

Because of the poor vector competence of fleas, plague epizootics require a high flea burden per host, even when the susceptible host population density is high (see References: Lorange 2005).

Experimental studies of human body lice have demonstrated that lice can also serve as vectors of Y pestis. Infected lice were able to transmit two virulent strains of plague to uninfected rabbits that subsequently became septicemic and died of plague. Infections were transmitted to naïve lice that fed on infected rabbits (see References: Houhamdi 2006).

Modes of Transmission

Bubonic plague is transmitted from animal reservoirs via:

Bites from flea vectors
Bites or scratches from infected animals, such as cats (see References: Gage 2000)
Direct contact with infected animal carcasses, such as rodents (especially marmots), rabbits, hares, carnivores (eg, wild cats, coyotes), and goats (see References: Christie 1980, Reed 1970, von Reyn 1976)

Pneumonic plague is transmitted via:

Inhalation of respiratory droplets (ie, large droplets [>5 microns]) from infected animals such as cats (see References: Gage 2000)
Inhalation of respiratory droplets from a person with primary or secondary pneumonic plague
Handling Y pestis cultures in the laboratory setting (see References: Burmeister 1962)

Pharyngeal plague can be transmitted via ingestion of Y pestis organisms. A recent report identified five patients with plague who acquired infection after eating raw camel liver; four developed severe pharyngitis and one developed submandibular lymphadenitis (see References: Bin Saeed 2005). In another report from Jordan, pharyngeal plague developed in 12 people after they consumed contaminated camel meat; 11 ate the meat raw and one ate cooked meat (see References: Arbaji 2005)

The risk of infection among contacts of cases of pneumonic plague has not been well quantified and likely varies with intensity of exposure and possibly with environmental factors.

During one recent outbreak of pneumonic plague in Madagascar, investigators measured serum F1 antibodies among contacts and estimated that the secondary infection rate was 8.4% (see References: Ratsitorahina 2000).
Another recent report reviewed eight documented outbreaks of pneumonic plague and found that the average number of secondary transmissions per primary transmission ranged from 0.8 to 3.0 (mean, 1.5; median, 1.3) (see References: Gani 2004).
A third study involved one definite and three probable plague cases (two concurrent index patient-caregiver pairs) (see References: Begier 2006). Each index case transmitted infection to one caregiver, although there were 23 additional close contacts for the two index cases (yielding a secondary infection rate of 8%). Transmission was consistent with large-droplet spread.
Several experimental studies have been conducted to determine how far Y pestis organisms will spread from patients with pneumonic plague. These studies involved placing agar plates at various distances from the mouths of infected patients. The farthest distance of spread was 3.7 feet, indicating that close contact is necessary for transmission (see References: Kool 2005).

Historical Perspective

Three plague pandemics have been recorded throughout history (see References: WHO: Report on global surveillance 2000), with an estimated 200 million deaths (see References: Perry 1997). Brief descriptions of the three pandemics follow.

The first pandemic started in Egypt in 542 AD and continued for more than a century. Outbreaks in Europe, Central and Southern Asia, and Africa killed an estimated 100 million people.
The second pandemic began in Italy in 1347 and rapidly spread throughout Europe over the next several years, killing an estimated one third of the European population. Paleodemographic studies suggest that mortality was partially determined by relative health of those infected (see References: DeWitte 2008). During that time, plague became known as the Black Death. Outbreaks of plague continued to occur sporadically in Europe over the next several centuries.
The third pandemic began in 1894 in China and spread around the world over a 10-year period, predominantly by infected rats and their fleas aboard steamships. An estimated 12 million deaths occurred, mostly in India.

Although bubonic plague historically has been the most common form of disease, large outbreaks of pneumonic plague (with person-to-person transmission as the primary mode of spread) also have been reported (see References: Meyer 1961, Kool 2005).

Two large outbreaks of pneumonic plague occurred in Manchuria in the early 20th century (1910-1911 and 1920-1921). An estimated 60,000 deaths occurred in the former and an estimated 9,300 in the latter.
Two pneumonic plague outbreaks occurred in the United States in the early 1900s (see References: Anderson 1978, Kool 2005). The first occurred in 1919 in Oakland, California. The index case was a hunter who contracted bubonic plague from an infected squirrel. He subsequently developed plague pneumonia and transmitted the disease to 12 or 13 other persons. A second outbreak occurred in Los Angeles in 1924 and involved 39 cases of pneumonic plague.

Naturally Occurring Plague in the United States

Plague was first introduced into the United States in 1900 from China; the first plague epidemic occurred in San Francisco from 1900 to 1904 (see References: Caten 1966).
Up through 1926, plague occurred most commonly in urban settings (particularly in California) and was associated with infections in urban rat populations (see References: Kaufmann 1980).
After 1926, plague gradually became endemic in wild animal populations in the Western United States (generally in wild rodents), and human cases continued to occur in persons with exposure to such populations.
Cases also have occurred following exposure to infected cats (see References: Eidson 1988, Gage 2000). A recent report described 23 cases of cat-associated plague in the western United States from 1977 through 1998; five were cases of primary pneumonic plague (see References: Gage 2000).
Since the 1920s, most human plague cases in the United States have occurred in California, New Mexico, Arizona, and Colorado (see References: CDC: Human plague—United States, 1993-1994).
Plague in the United States generally is seasonal, with a higher incidence in the summer months (see References: Caten 1966, Kaufmann 1980).
From 1947 through 1996, 390 cases of plague were reported to CDC (see References: CDC: Fatal human plague—Arizona and Colorado, 1996):

The overall case-fatality rate was 15.4%.
Bubonic plague accounted for 327 (83.9%) cases.
Primary septicemic plague accounted for 49 (12.6%) cases.
Primary pneumonic plague accounted for seven (1.8%) cases.
Seven cases were unclassified.

From 1990 through 2005, 107 cases of plague were reported to CDC, with a median of 7 cases per year (see References: CDC: Summary of notifiable diseases, 1999; CDC: Human plague—Four states, 2006).
Most human cases in the United States occur in two regions: (1) northern New Mexico, northern Arizona, and southern Colorado and (2) California, southern Oregon, and far western Nevada (see References: CDC: Plague: epidemiology). In April 2006, Los Angeles County health officials confirmed a case of bubonic plague in a local resident; this is the first human case of plague in a Los Angeles County resident since 1984 (see References: County of Los Angeles Department of Health Services).
Between February and July 2006, 13 human cases of plague were reported among residents of four states: New Mexico, 7; Colorado, 3; California, 2; and Texas, 1. This total is the largest number of cases reported in a single year in the United States since 1994. The increased number of cases during this time frame is consistent with the predicted relationship between climate and frequency of human plague in the southwestern United States.

Two consecutive February-March periods with high precipitation and an intervening cool summer favor an increased number of cases of plague the following summer. The net effect is increased reproduction and survival rates among rodents and fleas (see References: CDC 2006: Human plague—Four state; Enscore 2002).
A recent study used logistic regression and geographic information system (GIS)–based modeling to identify environmental predictors of elevated risk for plague in the southwestern United States. Results showed that two factors (distance to pinon-juniper ecotones and amount of precipitation) accurately identified case locations as suitable for plague (producer accuracy, 93%) (see References: Eisen 2007: A spatial model of shared risk).

Naturally Occurring Plague Worldwide

Natural enzootic foci of plague (and, therefore, areas with the highest incidence of human disease) include the following countries or regions (see References: CDC: Information on plague; WHO: Report on global surveillance 2000):

Madagascar
Eastern and southern Africa (eg, Uganda, Kenya, Tanzania, Mozambique, Botswana)
Southeast Asia (particularly Vietnam and Myanmar)
Pockets in South America (including areas in the Andean mountain regions of Peru, Bolivia, and Ecuador and in northeastern Brazil)
The western United States
Mongolia and northern China
Russia (the area of the Caucasus Mountains) and central Asia into the Middle East

Between 1954 and 1997, 38 countries reported cases of human plague to the World Health Organization (WHO) (see References: WHO: Report on global surveillance 2000). No plague cases have been reported from Europe since shortly after World War II and no cases have been reported from Australia. Overall, 80,613 cases were reported, with a mean of 1,832 cases per year (range, 200 to 6,004 cases).

The overall average case-fatality rate among reported cases was 11.8%.
Seven countries reported cases of plague for each of the 44 years: Brazil, Democratic Republic of the Congo, Madagascar, Myanmar, Peru, the United States, and Vietnam.
Large outbreaks occurred in Vietnam (from 1966 through 1972), India (1954, 1963, and 1994), Tanzania (1990 through 1992), and Madagascar (1994 through 1996).
The 1994 outbreak in India included cases of pneumonic plague and raised concerns about the spread of pneumonic plague to other areas of the world through airline travel (See References: Campbell 1995, Titball 1998).
In 1997, a localized outbreak of pneumonic plague involving 18 patients occurred in Madagascar; most of the patients had exposure to a traditional healer who died of the disease after treating the index patient (see References: Ratsitorahina 2000).

An outbreak of pneumonic plague was reported in the Democratic Republic of the Congo (DRC) in early 2005. Cases occurred between December 2004 and March 2005 in workers of a diamond mine where about 7,000 people worked under crowded conditions with poor sanitation. WHO reported 130 cases with 57 deaths in their last bulletin on the subject (see References: WHO: Plague in the Democratic Republic of the Congo). It is likely that the index case acquired Y pestis infection through a fleabite and then developed the pneumonic form of the disease and subsequently transmitted infection to other workers through the respiratory route. Additional outbreaks of pneumonic plague have occurred in 2006 in the Oriental province of the DRC (see References: WHO: Plague in the Democratic Republic of the Congo). More than 620 cases, including 42 deaths, were reported from the end of July to mid-October; however, investigators suspect that the total may be overestimated because of the low fatality rate observed among cases.

Following an outbreak of plague in Algeria in June 2003, investigators collected fleas taken from rodents and tested them for the presence of Y pestis (see References: Bitam 2006). The fleas were collected between September 2004 and May 2005. Results demonstrated that Y pestis had persisted in fleas, indicating an ongoing zoonotic focus of plague in the area. The authors concluded that flea studies may be a useful tool for epidemiologic surveillance and for identifying areas where plague may reemerge. Epidemiologic and biomolecular findings strongly suggest the existence of a local animal reservoir, although the origin (resurgence or reimportation) could not be determined (see References: Bertherat 2007).

Plague as a Biological Weapon

Experience with plague as a biological weapon is limited; however, the following information supports the perspective that plague deserves serious consideration as a bioterrorist agent:

Plague was used as a biological weapon in the Middle Ages when armies catapulted dead plague victims into cities under siege in order to spread the disease (see References: Osterholm 2000).
Japan used plague as a biological weapon against the Chinese during World War II by dropping plague-infected fleas over populated areas and causing outbreaks of the disease (see References: McGovern 1997, Osterholm 2000).
In the years following World War II, biological weapons programs in the United States and the Soviet Union developed techniques for aerosolizing Y pestis, thus enhancing the effectiveness of this agent as a potential biological weapon (see References: Inglesby 2000: Plague as a biological weapon).
A 1970 WHO report estimated that an aerosol release of 50 kg of dried powder containing 6 x 10¹⁵ Y pestis spores over a city of 5 million people in an economically developed country (such as the United States) would produce 150,000 incapacitating illnesses and up to 36,000 deaths (see References: WHO: Health aspects of chemical and biological weapons). These estimates did not take into consideration secondary cases that would occur through subsequent person-to-person contact.
Plague is a suitable pathogen for use as a biological weapon because:

The organisms can be delivered in an aerosol form.
Pneumonic plague causes a serious illness with a high case-fatality rate.
Pneumonic plague is communicable, and large outbreaks have occurred in the past.
A bioterrorist attack involving pneumonic plague would cause widespread fear and panic that would be difficult to contain, partly because of the communicable nature of the disease (see References: Campbell 1995; Inglesby 2001: A plague on your city).
Y pestis could potentially be genetically altered to enhance virulence or create antibiotic-resistant strains (see References: Gilsdorf 2005).

Plague used as a bioterrorist weapon would be expected to have the following features:

Previously healthy patients would present with a severe and rapidly progressive pneumonia.
An acute multilobar pneumonia accompanied by hemoptysis, associated gastrointestinal symptoms, and a fulminant clinical course would be very suspicious for pneumonic plague.
Many similar cases would present over several days.
Illness onsets would generally occur 2 to 4 days after release, but could occur as soon as 1 day and up to 6 days later.
Buboes characteristic of bubonic plague would not be present.
Illness would likely occur in an urban area and patient would not have a history of recent travel to a plague-endemic region (ie, southwestern United States).
Patients would not necessarily have risk factors for plague exposure (eg, outdoor field work, veterinary work, recent outdoor recreational activity).
There would be no indication of a prior recent plague epizootic with rodent deaths in the affected community.
Antibiotic resistance may be present.

Investigators recently used univariate and multivariate modeling to assess key parameters for controlling a pneumonic plague outbreak (see References: Massin 2007). Using a hypothetical reference scenario of 1000 index cases of plague pneumonia in Paris, if interventions were taken 10 days after an attack, an estimated 2,500 deaths would occur. Rapidity of implementing interventions offered the greatest effect on final epidemic size. Other measures, in order, were wearing masks, treating contacts preventively, and quarantine. Limiting inter-regional mixing confined casualties to the region but did not reduce casualties in the model.

Clinical Syndromes

Overview

Yersinia pestis infection can cause the following clinical syndromes:

Bubonic plague
Primary septicemic plague
Primary pneumonic plague
Plague meningitis
Plague pharyngitis
Pestis minor
Subclinical infection

The classic forms of plague are bubonic plague, septicemic plague, and pneumonic plague; these are outlined in the tables below. Septicemic plague can be either primary or secondary to bubonic plague. Similarly, pneumonic plague can be either primary or secondary to septicemic plague or bubonic plague (ie, following hematogenous spread).

Brief information about other syndromes caused by Y pestis infection follows:

Plague meningitis occurs as a complication of bacteremia and may be the presenting clinical syndrome for some cases. Symptoms are typical of meningitis from other etiologies and include fever, headache, meningismus, and mental status changes. If meningitis occurs as a complication of bubonic plague, some data suggest that a bubo in the axillary region is a predisposing factor (see References: Butler 1976). The cerebrospinal fluid demonstrates PMNs; characteristic gram-negative organisms usually can be seen on Gram stain (see References: Butler 1991).
Plague pharyngitis occurs as a result of inhaling or ingesting Y pestis organisms. The clinical illness is similar to severe pharyngitis or acute tonsillitis of other causes (eg, streptococcal infection); inflamed cervical nodes usually are present (and usually have the features of a characteristic bubo or buboes). As with bubonic plague, septicemia can occur.
Pestis minor is a milder form of bubonic plague. Patients usually have a febrile illness with localized lymphadenopathy. The nodes drain and patients recover without therapy. Patients with this form of plague are more likely to have some preexisting immunity to Y pestis (see References: Legters 1970).
Subclinical infections can occur as evidenced by cross-sectional surveys of serum antibody titers in populations living in endemic areas (see References: Ratsitorahina 2000).

Clinical Features of Bubonic Plague
Feature	Characteristics
Incubation period	1-7 days
Presenting features	—Sudden onset of fever, chills, weakness —Usually within 1 day, painful swollen lymph node or group of nodes (bubo) occurs in groin, axilla, or cervical region: ~1-10 cm, smooth, uniform, unfixed, egg-shaped mass or irregular cluster of several nodes ~Extremely tender ~Region may be erythematous, with surrounding edema ~Buboes usually occur in only one location, but multiple buboes may be seen ~Rarely, buboes may suppurate and rupture —Skin lesions may occur at site of flea bite (ie, papules, vesicles, pustules) but are present in <10% of cases —Associated lymphangitis uncommonly occurs —Presenting symptoms for case series of 40 Vietnamese patients with bubonic plague^a: ~Fever (100%) (mean temperature for 32 patients: 102.9°F [39.4°C]) ~Chills (40%) ~Bubo (100%—groin, 88%; axilla, 15%; cervical, 5%; epitrochlear, 3%) ~Headache (85%) ~Prostration (75%) ~Altered mental status (38%) (lethargy, confusion, delirium, seizures) ~Anorexia (33%) ~Vomiting (25%) ~Abdominal pain (18%) ~Cough (25%) ~Chest pain (13%) ~Skin rash (23%) (petechiae, purpura, papular eruptions)
Laboratory features	—Laboratory features for case series of 40 Vietnamese patients^a: ~Mean WBC count: 21,500/mm³ (range, 6,000/mm³-100,000/mm³) (most patients had left shifts and 3 had leukemoid reactions) ~PMNs showed cytoplasmic vacuolation in 24 patients, Dohle bodies in 20 patients, toxic granules in 8 patients ~Mean platelet count: 210,000/mm³ (range, 72,000/mm³-496,000/mm³) (18 patients had platelet counts <150,000/mm³) ~SGOT elevated in 13 patients (20-92 M-IU) ~LDH elevated in 7 patients (308-900 units) ~Alkaline phosphatase elevated in 9 patients (33-116 units) ~PTT >10 seconds over control In 6 patients
Complications	—Secondary septicemia (can lead to DIC, shock, multisystem involvement) —Secondary pneumonic plague (5%-15% of patients)^b —Meningitis (may occur in patients with bubonic plague that was not adequately treated)^b —Buboes may become infected with other bacterial pathogens^a
Case-fatality rate	—Over 50% without antibiotic therapy^c —With appropriate antibiotic therapy, <5%^d
Abbreviations: DIC, disseminated intravascular coagulation; LDH, lactic dehydrogenase; PMNs, polymorphonuclear neutrophils; PTT, partial thromboplastin time; SGOT, serum glutamic oxaloacetic transaminase; WBC, white blood cell. ^aSee References: Butler 1972. ^bSee References: McGovern 1997. ^cSee References: Dennis 1997. ^dSee References: Butler 1991.

Clinical Features of Primary Septicemic Plague
Feature	Characteristics
Incubation period	1-4 days
Presenting features^a,b	—10%-25% of US plague cases present with primary septicemic plague^b —Presenting symptoms for 18 cases of primary septicemic plague in New Mexico: ~Fever (100%) ~Chills (61%) ~Nausea (44%) ~Headache (44%) ~Vomiting (50%) ~Diarrhea (39%) ~Abdominal pain (39%) ~Any gastrointestinal symptom (72%) —Presenting signs for 18 cases of primary septicemic plague in New Mexico^a: ~Mean temperature: 38.5°C (range, 35.4°C-40.4°C) ~Mean pulse: 109 (range, 72-160) ~Mean respiratory rate: 31 (range, 16-60) ~Mean systolic BP: 104 (range, 80-130) ~Mean diastolic BP: 66 (range, 36-80) —Mental status changes commonly occur (delirium, obtundation, coma)
Laboratory features	—Laboratory features consistent with severe bacterial infection and sepsis syndrome (as often seen with bubonic plague and secondary septicemia) —Leukocytosis, leukopenia, or normal WBC count may be seen —If plague pneumonia present, CXR shows patchy alveolar infiltrates (usually bilateral), often with consolidation —Findings noted for 18 patients with septicemic plague: ~Mean WBC count: 18,950/mm³ (range, 3,000/mm³-68,700/mm³); all had marked left shifts ~Bacteria seen on peripheral blood smear (17.6%)
Complications^a,c,d	—Illness rapidly progresses to sepsis syndrome often with DIC, shock, and multisystem involvement —Skin lesions reflect DIC (may be similar to meningococcemia)^c: ~Purpura ~Petechiae ~Ecchymoses ~Gangrene of acral regions (caused by small artery thromboses) ~Ecthyma gangrenosum (rare) —Meningitis^d —Secondary plague pneumonia (about 25% of patients)*
Case-fatality rate^d	—Overall 30%-50%^b —High CFR related to delay in appropriate diagnosis and antibiotic therapy —Without antibiotic therapy, CFR approaches 100%
Abbreviations: BP, blood pressure; WBC, white blood cell; CXR, chest x-ray. ^aSee References: Hull 1987. ^bSee References: Perry 1997. ^cSee References: McGovern 1999. ^dSee References: Dennis 1997.

Clinical Features of Primary Pneumonic Plague^a
Feature	Characteristics
Incubation period	1-4 days
Presenting features	—Symptoms of primary plague pneumonia^b: ~Fever ~Chest pain ~Dyspnea ~Productive cough (sputum may be purulent or watery, frothy, blood-tinged) ~Hemoptysis ~Tachypnea (particularly in young children) ~Cyanosis ~Bubo not present (rarely, cervical bubo may be noted) —Gastrointestinal symptoms (nausea, vomiting, abdominal pain, diarrhea) common
Laboratory features	—Findings consistent with severe bacterial infection and sepsis syndrome (as often seen with bubonic and primary septicemic plague) —CXR findings in series of 9 cases of secondary pneumonic plague^c: ~Alveolar infiltrates (100%) ~Pleural effusion (55%) ~One patient developed cavitary lesion 3 weeks after illness onset —Consolidation common on CXR; massive mediastinal adenopathy occurs rarely —Organisms usually seen on sputum Gram stain
Complications	—Septicemia with sepsis syndrome —Meningitis
Case-fatality rate	—Close to 100% without appropriate antibiotic therapy (generally, fatality rates are high if antibiotic therapy is not instituted soon after symptom onset [ie, within 24 hr]; however, patients may survive even if appropriate therapy is instituted beyond 24 hr)^d —Death often occurs 2-5 days after illness onset^e
Abbreviations: CXR, chest x-ray. ^aNote: Few detailed clinical descriptions of primary plague pneumonia are readily available since the condition is relatively rare. ^bSee References: Dennis 1997. ^cSee References: Alsofrom 1981. ^dSee References: Begier 2006, Gage 2000, Butler 1991. ^eSee References: Inglesby 2000: Plague as a biological weapon.

Pediatric Considerations

The clinical presentation of plague in children is similar to that in adults. Several studies have made the following observations about pediatric plague:

Children with bubonic plague may have a slightly increased risk for development of secondary pneumonic plague or meningitis. In one study of 38 pediatric patients with plague, 16% developed pneumonia and 11% developed meningitis (see References: Mann 1982). In two other case series, most of the meningitis cases occurred among children (see References: Crook 1992, Reed 1970).
Vomiting may be more common in children at the time of presentation of illness than in adults (ie, about 50% and 30%, respectively) (see References: Burkle 1973, Mann 1982).
In cases of bubonic plague, node pain is more common in children than lymph node swelling or a bubo; this may manifest as limb immobility (such as from painful axillary nodes).
Retroperitoneal adenopathy may be responsible for vomiting and/or abdominal pain.
Children may be more likely to have seizures as part of the presenting symptom complex (see References: Butler 1991). Most often these are febrile seizures, although they may be caused by plague meningitis, which may be more common in children.
The diagnosis often is not considered at time of initial presentation for pediatric cases, even in a plague-endemic area (see References: Mann 1982).

Presenting symptoms for 38 pediatric patients with bubonic or septicemic plague diagnosed in New Mexico between 1970 and 1980 are shown in the table below. The overall case-fatality rate was 15.8%.

Presenting Symptoms for 38 Pediatric Patients With Bubonic or Septicemic Plague
Symptom	Patients With Bubonic Plague (N=31)	Patients With Septicemic Plague (N=7)
Fever Chills Vomiting Headache Abdominal distress or nausea Diarrhea Lethargy, malaise, anorexia	30 (97%) 11 (35%) 16 (52%) 11 (35%) 8 (26%) 3 (10%) 12 (39%)	36 (95%) 0 3 (43%) 0 2 (29%) 0 3 (43%)
Adapted from Mann JM, et al. Pediatric plague (see References).

In the pre-antibiotic era, pregnant women with plague usually either died or had an adverse outcome of pregnancy (eg, spontaneous abortion, stillbirth). However, more recent reports have demonstrated successful outcomes with antibiotic therapy, including normal gestational periods and delivery of healthy infants (see References: Mann 1977, Welty 1985). Potential adverse effects to the fetus are governed by the time of antibiotic therapy; however, during outbreaks and bioterrorism emergencies, treatment benefits for the mother outweigh fetal risk (see References: Cono 2006). Antibiotics should be administered to infants born to infected mothers (see References: Welty 1985).

Breast-feeding women and their infants should be treated with the same antibiotic. The medication that is safest for the infant generally should be considered the first choice (ie, gentamicin in the contained casualty setting and doxycycline in the mass casualty setting). Fluoroquinolone antibiotics would be the recommended alternative in both settings (see References: Inglesby 2000: Plague as a biological weapon).

Differential Diagnosis

Differential Diagnosis for Pneumonic, Bubonic, and Septicemic Plague
Condition	Distinguishing Features
PNEUMONIC PLAGUE^a
Inhalational anthrax (Bacillus anthracis)	—Widened mediastinum and pleural effusions seen on CXR or chest CT —Not true pneumonia; minimal sputum production —Hemoptysis uncommon (if present, suggests diagnosis of plague)
Tularemia (Francisella tularensis)	Clinical course not as fulminant as pneumonic plague
Community-acquired bacterial pneumonia —Mycoplasmal pneumonia (Mycoplasma pneumoniae) —Pneumonia caused by Chlamydia pneumoniae —Legionnaires' disease (Legionella pneumophila or other Legionella species) —Psittacosis (Chlamydia psittaci) —Other bacterial agents (eg, Staphyloccocus aureus, Streptococcus pneumoniae, Haemophilus influenzae, Klebsiella pneumoniae, Moraxella catarrhalis)	—Rarely as fulminant as pneumonic plague —Legionellosis and many other bacterial agents (S aureus, S pneumoniae, H influenzae, K pneumoniae, M catarrhalis) usually occur in persons with underlying pulmonary or other disease or in the elderly —Bird exposure with psittacosis —Gram stain of sputum may be useful —Community outbreaks caused by other etiologic agents not likely to be as explosive as pneumonic plague outbreak —Outbreaks of S pneumoniae usually institutional —Community outbreaks of Legionnaires' disease often involve exposure to cooling towers
Viral pneumonia —Influenza —Hantavirus —RSV —CMV	—Influenza generally seasonal (October-March in United States) or involves history of recent cruise ship travel or travel to tropics —Exposure to excrement (urine or feces) of mice with Hantavirus —RSV usually occurs in children (although may be cause of pneumonia in elderly); tends to be seasonal (winter/spring) —CMV usually occurs in immunocompromised patients
Q fever (Coxiella burnetii)	—Exposure to infected parturient cats, cattle, sheep, goats —Severe pneumonia not prominent feature
BUBONIC PLAGUE^b
Streptococcal or staphylococcal adenitis (S aureus, Streptococcus pyogenes)	—Purulent or inflamed lesion often noted distal to involved nodes (ie, pustule, infected traumatic lesion) —Involved nodes more likely to be fluctuant —Associated ascending lymphangitis or cellulitis may be present (generally not seen with plague)
Tularemia (F tularensis)^c	—Ulcer or pustule often present distal to involved nodes —Clinical course rarely as fulminant as in plague —Systemic toxicity uncommon
Cat scratch disease (Bartonella henselae)	—History of contact with cats; usually history of cat scratch —Indolent clinical course; progresses over weeks —Primary lesion at site of scratch often present (small papule, vesicle) —Systemic toxicity not present
Mycobacterial infection, including scrofula (Mycobacterium tuberculosis and other Mycobacterium species)	—With scrofula, adenitis occurs in cervical region —Usually painless —Indolent clinical course —Infections with species other than M tuberculosis more likely to occur in immunocompromised patients
Lymphogranuloma venereum (Chlamydia trachomatis)	—Adenitis occurs in the inguinal region —History of sexual exposure 10-30 days previously —Suppuration, fistula tracts common —Although LGV buboes may be somewhat tender, exquisite tenderness usually absent —Although patients may appear ill (headache, fever, myalgias), systemic toxicity not present
Chancroid (Hemophilus ducreyi)	—Adenitis occurs in the inguinal region —Ulcerative lesion present —Systemic symptoms uncommon; toxicity does not occur
Primary genital herpes	—Herpes lesions present in genital area —Adenitis occurs in the inguinal region —Although patients may be ill (fever, headache), severe systemic toxicity not present
Primary or secondary syphilis (Treponema pallidum)	—Enlarged lymph nodes in the inguinal region —Lymph nodes generally painless —Chancre may be noted with primary syphilis
Strangulated inguinal hernias	Evidence of bowel involvement
SEPTICEMIC PLAGUE
Meningococcemia	More likely to have evidence of meningitis (but not always present)
Septicemia caused by other gram-negative bacteria	Underlying illness usually present
Abbreviations: CMV, cytomegalovirus; CT, computed tomography; CXR, chest x-ray; LGV, lymphogranuloma venereum; RSV, respiratory syncytial virus. ^aOther causes of pneumonia also may be considered, depending on the clinical presentation and setting (eg, tuberculosis, fungal infections). ^bInfectious causes of generalized lymphadenopathy (eg, CMV infection, toxoplasmosis, mononucleosis) also may be considered, depending on the clinical presentation. ^cSee References: Butler 1979; Cleri 1997.

Laboratory Diagnosis

Specimen Collection and Transport

Collection and Transport of Clinical Laboratory Specimens for the Diagnosis of Plague
Condition	Collection and Transport^a,b,c
Pneumonia	Sputum: —May be collected for culture/direct examination, but yield on culture may be low owing to likely overgrowth of normal flora^d —Transport at room temperature (22°C-28°C) if transport <2 hr —If transport expected to be 2-24 hr, refrigerate (2C°C-8°C) —Order culture, Gram stain, and Giemsa, Wright’s, or Wayson stain —If suspicion of plague high, contact local health department and LRN laboratory for instructions on ordering DFA or other tests Bronchial wash (>1.0 mL): —Bronchoscopy may be indicated in certain situations where sputum specimens are negative [Note: In such situations, the bronchoscopy team should follow appropriate barrier precautions, including use of masks and other personal protective equipment; if a bronchoscopy is performed on a patient who later is found to have pneumonic plague and the team did not wear respiratory protection, then post-exposure prophylaxis is indicated for members of the team] —Use same specimen handling conditions and test orders as described for sputum specimens Blood: —Collect volume and number of sets per institution’s standard protocol —Transport to laboratory and hold at ambient temperature until placed into incubator or blood culture instrument —DO NOT REFRIGERATE —Follow established laboratory protocol for processing blood cultures —If high suspicion of plague, order additional blood or broth culture (general nutrient broth) for incubation at room temperature (22°C-28°C), the optimal growth temperature range for Y pestis. Cultures should be incubated without shaking [Note: An additional culture set is needed because holding cultures at room temperature will delay or negate growth of other common bacterial pathogens] Serum: —An acute-phase serum sample may be collected and stored at 4°C until plague can be ruled out —If plague cannot be ruled out, contact public health officials and LRN system for further instructions
Septicemia	Blood: —Collect volume and number of sets per institution’s standard protocol —Transport to laboratory and hold at ambient temperature until placed into incubator or blood culture instrument —DO NOT REFRIGERATE —Follow established laboratory protocol for processing blood cultures —If high suspicion of plague, order additional blood or broth culture (general nutrient broth) for incubation at room temperature (2


Home	Mission & Activities	About Us	Center Support	Contact Us