Clinical manifestations, diagnosis, and treatment of plague (Yersinia pestis infection)

INTRODUCTION — In the genus Yersinia, three species are important human pathogens: Yersinia pestis, Yersinia enterocolitica, and Yersinia pseudotuberculosis. The yersinioses are zoonotic infections of domestic and wild animals; humans are considered incidental hosts that do not contribute to the natural disease cycle.

Y. pestis causes plague and is transmitted by fleas. The most common clinical manifestation is acute febrile lymphadenitis, called bubonic plague. Less common forms include septicemia, pneumonia, pharyngeal, and meningeal plague.

The clinical features, diagnosis, and treatment of plague will be reviewed here. The epidemiology, microbiology, and pathogenesis of Y. pestis are discussed separately. (See "Epidemiology, microbiology and pathogenesis of plague (Yersinia pestis infection)".)

Issues related to other Yersinia species are discussed separately. (See "Yersiniosis: Infection due to Yersinia enterocolitica and Yersinia pseudotuberculosis".)

EPIDEMIOLOGY — Plague is a murine zoonosis; humans are incidental hosts. Humans acquire plague via bites of rodent fleas, scratches or bites from infected domestic cats, direct handling of infected animal tissues, inhalation of respiratory secretions from infected animals, inhalation of aerosolized droplets from infected humans, consumption of contaminated food, or by laboratory exposure [1-5]. The incubation period is generally two to eight days. (See "Epidemiology, microbiology and pathogenesis of plague (Yersinia pestis infection)".)

CLINICAL MANIFESTATIONS — There are three major clinical syndromes associated with plague: bubonic plague, septicemic plague, and pneumonic plague [1]. Bubonic plague accounts for 80 to 95 percent of cases; septicemic plague accounts for 10 to 20 percent of cases [1,6]. Pneumonic plague is generally rare [7]. Overall, the estimated mortality is 60 to 100 percent in untreated plague compared with less than 15 percent with treatment [8]. In a series of 23 cases of plague in the United States between 1977 and 1998, 17 were bubonic plague and the mortality rate was 22 percent [9].

Bubonic plague — Bubonic plague is the most common form of plague; it accounts for 80 to 95 percent of cases. Skin lesions at the site of flea bite are usually inapparent and thus ignored or forgotten. However, some patients may have eschars, pustules, or even necrotic lesions resembling ecthyma gangrenosum [10]. Flea bites on the extremities may be associated with lymph nodes swellings in the region draining the inoculation site. A minority of patients develop skin lesions, such as purpura, associated with disseminated intravascular coagulation [11].

Bubonic plague is clinically characterized by the sudden onset of fever, chills, weakness, and headache, followed by intense pain and swelling in a lymph node-bearing area (bubo), which may be preceded by lymphadenopathy. Acute buboes are painful but lack fluctuation. They are often associated with erythema and edema of the overlying skin. The inguinal region is the most frequently involved site (the word "bubo" derives from the Greek word for "groin"); axillary or cervical regions may also be involved (picture 1). Axillary, cervical, or epitrochlear buboes may occur in patients with bubonic plague associated with cat exposure [9].

In the absence of treatment, the initial bubonic stage may be followed by disseminated infection (sepsis) in approximately 50 percent of untreated cases, which can lead to complications, such as pneumonia (secondary pneumonic plague) and meningitis [12]. Bacteremic patients may develop signs of septic shock [1].

Septicemic plague — Septicemic plague that occurs without a preceding bubo accounts for 10 to 20 percent of cases [1]. The septicemic form of plague may be particularly difficult to diagnose in a timely manner if characteristic clinical clues, such as a bubo, are not present. Patients with septicemic plague are febrile and extremely ill but may not have localizing signs or symptoms. Gastrointestinal symptoms, including nausea, vomiting, diarrhea and abdominal pain, may be observed. Hypotension, disseminated intravascular coagulation, and multiorgan failure develop in the later stages of the illness.

Pneumonic plague — Pneumonic plague can be primary or secondary. Primary pneumonic plague can be acquired by inhalation of respiratory secretions or aerosolized droplets from infected animals or humans, or by laboratory exposure [2-4,11,13]. Secondary pneumonic plague is more common and arises via hematogenous spread of bacteria from a bubo or other source.

Primary pneumonic plague has a short incubation period, ranging from a few hours to a few days. Affected patients typically present with the sudden onset of dyspnea, high fever, pleuritic chest pain, and cough that may be accompanied by characteristic bloody sputum. Pneumonic plague is rapidly fatal unless an appropriate antimicrobial agent is begun within the first day of illness [7].

Secondary pneumonic plague develops in approximately 10 percent of patients with plague in the United States, usually as a result of delayed treatment of bubonic infection. The manifestations are as described above for primary pneumonic plague.

Human-to-human transmission of plague via aerosols remains a source of controversy. In one study of an outbreak in Madagascar, 8 percent of 154 contacts became infected with Y. pestis [7]. However, in a study of cases of plague from an outbreak in Uganda, transmission was observed only in very close contacts (caregivers) of patients who are severely ill [14]. Treatment of patients and prophylaxis of contacts stopped the outbreak in Madagascar cited above. The capacity for spread via aerosolization makes Y. pestis a potential bioterrorism agent. (See 'Prevention' below and "Identifying and managing casualties of biological terrorism".)

Other manifestations — Meningitis can occur in conjunction with any of the three forms of plague, or it may occur as a "primary" meningeal infection, presumably the result of occult bacteremic seeding of the central nervous system. Symptoms tend to be similar to other bacterial meningitides. Cerebrospinal fluid examination typically reveals a low glucose concentration, increased protein concentration, and a neutrophilic pleocytosis. (See "Clinical features and diagnosis of acute bacterial meningitis in adults".)

Other, less common manifestations of plague include pharyngitis and tonsillitis associated with anterior cervical lymphadenitis, following ingestion of Y. pestis [3,11]. Asymptomatic transient pharyngeal carriage of Y. pestis has been described in healthy contacts of bubonic plague cases [15]. One small series described involvement of the urinary tract or gastrointestinal tract in 14 percent of cases [3].

Patients who develop plague without a bubo are difficult to diagnose, particularly when infections occur as sporadic cases in the absence of a recognized outbreak. Such patients may develop manifestations of multiple organ dysfunction and die prior to or in spite of a correct antemortem diagnosis.

Laboratory and imaging features — White blood cell counts may vary from 3000 to 100,000/microL, but white cell counts >20,000/microL and/or thrombocytopenia are present in approximately one-half of all cases. The combination of white cell counts >20,000/microL and thrombocytopenia is a useful diagnostic clue for the recognition of plague in endemic areas [16].

The radiographic appearance of pneumonic plague is not sufficiently specific to be diagnostic. Chest radiography may demonstrate bronchopneumonia, consolidation, cavities, or pleural effusions. Unlike other, more common bacterial pneumonias, hilar or mediastinal adenopathy is sometimes seen in patients with plague pneumonia, and this finding may provide one clue to the presence of this unusual pathogen [17]. The sputum may initially be clear, but it is more typically purulent, or hemorrhagic, and it usually contains gram-negative rods when Gram stains are done.

DIAGNOSIS

Clinical suspicion — A high index of suspicion must be maintained in order to make a timely diagnosis of plague. Patients with fever and painful lymphadenopathy should be questioned about travel to areas of endemic disease, including the Southwestern United States. In addition, animal or rodent vector contact within the preceding 10 days may provide clues that raise suspicion for a diagnosis of plague. The diagnosis of plague is established by isolation of the organism in culture or by serologic testing [1,2]. Occasionally, patients with plague may be coinfected with malaria, leading to diagnostic confusion in areas where malaria is endemic [18].

Three important diagnostic clues to the presence of plague include [7]:

●The presence of fever in a person with either known contact with dead rodents or residence or travel to a plague-endemic region

●The presence of fever, hypotension, and unexplained regional lymphadenitis

●The presence of clinical findings of pneumonia in association with hemoptysis and sputum containing gram-negative rods on Gram stain

Microbiologic diagnosis

Culture and staining — Yersinia grows well on common laboratory media. Microbiology personnel should be informed of any specimen suspected to harbor the organism so that they may exercise proper precautions to prevent laboratory acquisition of infection.

Blood cultures may be positive in 27 to 96 percent of patients, depending on the form of their clinical presentation [4,11]. Y. pestis can also be cultured from sputum, cerebrospinal fluid, or pus aspirated from buboes. In one study, cultures of bubo aspirates were positive in 10 of 13 cases [3]. Aspiration of the bubo after injection with saline may be necessary to obtain an adequate specimen.

Examination of a peripheral blood specimen with Wright-Giemsa stain demonstrates rod-shaped organisms in up to 40 percent of cases. Microscopy evaluation of a bubo aspirate may also demonstrate the organism. Wayson stain may demonstrate the typical bipolar staining, which resembles a "closed safety pin" (picture 2). Gram stain shows small gram-negative coccobacilli. Fluorescent antibody is also useful if available.

Automated bacterial identification systems may not be sufficiently accurate in identifying Y. pestis in a specimen culture, as illustrated by reports of misclassification as Pseudomonas luteola, Acinetobacter lwoffi, and Yersinia pseudotuberculosis [19,20]. Thus, in a patient with symptoms consistent with plague or microbiologic findings of an organism with bipolar staining, the possibility of Y. pestis should not be discounted if these other organisms are detected by such systems.

Serology — Serologic confirmation requires acute and convalescent serum, looking for at least a fourfold rise in antibody titers to the F-1 antigen of Y. pestis [2]. A single titer of >1:16 using the passive hemagglutination test is suggestive of the diagnosis [11].

Rapid tests (if available) — A rapid diagnostic test capable of detecting 0.5 ng/mL of the Y. pestis F1 antigen in the sputum or serum within 15 minutes has been developed [7,21]. In field testing in Madagascar, this assay had 100 percent sensitivity and specificity for Y. pestis, other Yersinia spp, and other bacteria; positive and negative predictive values were 91 and 87 percent, respectively [21].

Presumptive identification of Y. pestis can be made by polymerase chain reaction (PCR) or antigen-capture enzyme-linked immunosorbent assay [22,23]. PCR testing has been used to detect Y. pestis in teeth of skeletons that are hundreds of years old, and whole-genome sequencing of Y. pestis has been accomplished using specimens from victims of the Black Death who died in the Middle Ages [24].

An experimental method of rapid detection and simultaneous antibiotic susceptibility testing is based upon a genetically engineered "light-tagged" reporter phage that preferentially infects Y. pestis and, in doing so, creates a bioluminescent signal [25]. Because non-pestis strains of Yersinia and a few species of Enterobacteriaceae may also produce bioluminescence, albeit at an attenuated level, the potential for false-positive tests requires additional confirmatory tests, such as the Y. pestis F1 antigen. However, thus far, organisms likely to produce a plague-like illness, such as Francisella tularensis, Bacillus anthracis, and Klebsiella pneumoniae, do not appear to produce false-positive tests with this method. Thus, the use of reporter phages for detection of Y. pestis may have utility in patients with pneumonic plague or in circumstances in which bioterrorism is suspected.

TREATMENT

General principles

Importance of prompt therapy — Appropriate and timely antibiotic therapy markedly improves patient outcomes. Delayed antibiotic therapy of plague has been associated with higher mortality. As an example, in a systematic review of 762 published cases of plague, mortality rates were higher among those treated three or more days after symptom onset than those treated within two days: 24 versus 9 percent for bubonic plague, 33 versus 20 percent for pneumonic plague, and 56 versus 18 percent for septicemic plague [26].

Antibiotic resistance — Drug-resistant isolates of Y. pestis have been described but do not appear to be widely prevalent [27-29]. Nevertheless, modification of empiric therapy may be required to conform to local susceptibility patterns.

In 1995, an isolate from a patient in Madagascar contained a plasmid that conferred resistance to ampicillin, chloramphenicol, kanamycin, streptomycin, spectinomycin, sulfonamides, tetracycline, and minocycline [30]. The isolate remained sensitive to other aminoglycosides, quinolones, and trimethoprim. The plasmid probably originated among Enterobacteriaceae and was readily passed in vitro between Y. pestis strains. A second isolate of Y. pestis with plasmid-mediated high-level streptomycin resistance was subsequently described in another patient in Madagascar, but this strain was not simultaneously resistant to other common antimicrobial agents [31,32]. A third isolate with plasmid-mediated doxycycline resistance was isolated from a rat in Madagascar in 1998 [28]. These three antibiotic-resistance plasmids were unrelated and separated in space and time, suggesting the potential for acquired antibiotic resistance under natural conditions, possibly via horizontal conjugative genetic exchange in the midgut of fleas [33].

Preferred antibiotic regimens — The approach to regimen selection depends in part on the clinical syndrome and whether the infection is related to bioterrorism, as detailed below.

In 2021, the United States Centers for Disease Control and Prevention (CDC) published guidelines on the treatment of plague. Our approach is generally consistent with those guidelines [34].

Bubonic plague — For patients who present with known or suspected bubonic plague, we suggest treating with an aminoglycoside (streptomycin or gentamicin), a fluoroquinolone (levofloxacin, ciprofloxacin, or moxifloxacin), or doxycycline. These are all first-line agents suggested by the CDC [34]; doses are listed in the table (table 1). Of the first-line options, we generally favor an aminoglycoside, as clinical experience is greatest with this class, unless there are concerns for toxicity (eg, advanced age, chronic kidney disease, use of other nephrotoxic agents). Buboes that suppurate and become fluctuant also warrant incision and drainage.

Monotherapy with one of the agents listed above is typically sufficient for naturally acquired bubonic plague, as resistance is uncommon in this setting. If infection in the setting of a bioterrorist attack is suspected, we suggest use of a combination of agents from two of the above classes until susceptibility testing results are available, consistent with CDC recommendations, because of the possibility of engineered resistance [34].

Patients who initially have bubonic plague but develop pulmonary involvement or disseminated disease should be treated as having pneumonic or septicemic plague. (See 'Septicemic or pneumonic plague' below.)

Untreated bubonic plague has reportedly been associated with a mortality rate of 50 to 90 percent [1,7]. Mortality rates in patients with bubonic plague who are treated with antibiotics are considerably lower. As an example, in a 2020 systematic review of studies published since 1942 that reported aggregate data on plague outcomes with antibiotic therapy, the mortality rate was 14 percent among the approximately 2000 patients with bubonic plague [35].

Clinical experience is most extensive with aminoglycosides, in particular streptomycin [16,36,37]. However, limited aminoglycoside availability and the associated nephro- and ototoxicity often favor using alternative agents, specifically tetracyclines or fluoroquinolones.

Systematic clinical data informing the efficacy of specific antibiotics are limited, and there is no clear evidence that any one of the first-line agents or antibiotic classes is superior to the others. In one of the only randomized plague trials conducted, which included 65 patients, nearly all of whom had bubonic plague, cure or clinical improvement within seven days was similarly high with either gentamicin or doxycycline (94 versus 97 percent) and there were no reported relapses [38]. The three patients who died (two with gentamicin, one with doxycycline) all did so within the first day or two of therapy.

Other clinical evidence is limited to observational studies and case series [26,39,40]. In a 2020 systematic review of published individual plague cases since 1937 that included 476 cases of bubonic plague, mortality rates were 11 percent with aminoglycosides, 0 percent with fluoroquinolones, and 7 percent with tetracyclines [26]. Similarly, in a review of bubonic plague cases reported in the United States between 1942 and 2018, mortality rates were 8 percent with aminoglycosides, 4 percent with fluoroquinolones, and 4 percent with tetracyclines [40].

Septicemic or pneumonic plague — For patients who present with known or suspected septicemic or pneumonic plague, we suggest treating with an aminoglycoside (streptomycin or gentamicin) or a fluoroquinolone (levofloxacin, ciprofloxacin, or moxifloxacin). These are all first-line agents suggested by the CDC [34]; doses are listed in the table (table 1). Of the first-line options, we generally favor an aminoglycoside, as clinical experience is greatest with this class, unless there are concerns for toxicity (eg, advanced age, chronic kidney disease, use of other nephrotoxic agents). Levofloxacin or moxifloxacin may also be preferred in settings in which community-acquired pneumonia due to other pathogens has not been ruled out.

We do not routinely use more than one drug to treat naturally acquired septicemic or pneumonic plague, as resistance in this setting is uncommon. However, the CDC suggests using agents from two different classes for patients with severe disease until the patient has clinically improved [34]. If infection in the setting of a bioterrorist attack is suspected, we do suggest use of a combination of agents from two of the above classes until susceptibility testing results are available, consistent with CDC recommendations, because of the possibility of engineered resistance.

Untreated pneumonic plague has reportedly been associated with universal mortality [1,7], whereas the mortality rates reported for treated pneumonic and septicemic plague are much lower. As an example, in a 2020 systematic review of studies published since 1942 that reported aggregate data on plague outcomes with antibiotic therapy, the mortality rate was 31 percent among the 122 patients with pneumonic plague and 20 percent among the 10 patients with septicemic plague [35].

As with bubonic plague, clinical experience is most extensive with aminoglycosides, in particular streptomycin [16,36,37]. In general, clinical evidence is limited to observational studies and case series [26,39-42]. In a 2020 systematic review of published individual plague cases since 1937, mortality rates among the 158 patients with pneumonic plague were largely similar for aminoglycosides (19 percent), fluoroquinolones (20 percent), and tetracyclines (17 percent) when each was used as monotherapy [26].

Tetracyclines (eg, doxycycline) are not considered first-line for septicemic or pneumonic plague since some data among nonhuman primates suggested that they were less effective than other first-line options [43]. Nonhuman primate studies have also been used to support use of fluoroquinolones [44,45]. As an example, in a randomized trial of African green monkeys exposed to aerosolized Y. pestis, 16 of 17 treated with levofloxacin survived, whereas all of the seven control animals died [44].

Other infections and special populations — Other manifestations of plague less common than bubonic, pneumonic, and septicemic plague include the following:

●Pharyngeal and tonsillar plague – These are treated with the same approach as bubonic plague. (See 'Bubonic plague' above.)

●Meningitis – For initial treatment of patients with plague accompanied by signs of meningitis, the CDC recommends combination therapy with chloramphenicol and levofloxacin or moxifloxacin [34]. These agents are preferred because of higher penetration into the cerebrospinal fluid.

●Infection during pregnancy – The CDC recommends combination therapy with gentamicin and either ciprofloxacin or levofloxacin for pregnant individuals with plague because of the high potential for maternal morbidity, possibility of perinatal transmission, and potential for increased drug clearance during pregnancy. Administration of these agents during pregnancy may warrant higher-than-usual doses. Streptomycin is contraindicated during pregnancy.

Alternative options — Alternative options for plague treatment include tetracycline, trimethoprim-sulfamethoxazole, and chloramphenicol. These have been effective for plague in case series and observational studies but have some drawbacks compared with the preferred agents discussed above. Other agents from the same antibiotic classes as the preferred agents (eg, aminoglycosides [tobramycin, amikacin, plazomicin], fluoroquinolones [gemifloxacin, ofloxacin], tetracyclines [minocycline, eravacycline, omadacycline]) may also be reasonable alternatives, although clinical data with these are lacking.

●Tetracycline – Mortality rates of approximately 7 to 10 percent have been reported in patients with plague treated with tetracyclines [26]. Tetracycline, however, does not provide added benefit over doxycycline and is not as well tolerated. It is also not available in an intravenous formulation, so it is generally not used for initial treatment of pneumonic or septicemic plague. The dose is 500 mg orally every six hours for adults and 10 mg/kg (maximum 500 mg per dose) every six hours for children.

●Chloramphenicol – Mortality rates of approximately 11 to 20 percent have been reported in patients with plague treated with chloramphenicol [26]. It achieves excellent spinal fluid penetration and may be used for treatment of meningitis either alone or in combination with an aminoglycoside [46]. However, availability of chloramphenicol is limited in many settings, and it is associated with hematologic toxicity. The typical dose of chloramphenicol is 12.5 mg/kg intravenously every six hours [34]. A higher dose up to 25 mg/kg (max 1 g/dose) can be given for severe infections but should be reduced with clinical improvement.

●Trimethoprim-sulfamethoxazole – Mortality rates of approximately 21 to 23 percent have been reported in patients with plague treated with sulfonamides [26]. Responses may be delayed or incomplete compared with other agents [47]. The dose is 5 mg/kg of the trimethoprim component intravenously or orally every eight hours [34].

Penicillins, cephalosporins, and macrolides are not appropriate agents for treatment of plague.

Duration — The optimal duration of antimicrobial treatment for plague is uncertain; regimens are generally given for at least 7 to 14 days and for at least a few days after clinical signs and symptoms of infection have resolved. The CDC recommends at least 10 to 14 days of therapy [34].

Most of the limited data on plague treatment have evaluated antimicrobial regimens of 7- to 10-day duration [38,42]. Because they are bacteriostatic agents, it is reasonable to extend tetracycline or doxycycline courses to 10 to 14 days.

PREVENTION

Reducing exposure — Reducing exposure is the best preventive measure. In known endemic areas, handling sick or dead animals (including rodent carcasses) should be avoided (or if unavoidable, gloves should be worn). Close contact with individuals with suspected or diagnosed pneumonic plague should be also avoided. Other preventative measures include rodent control, flea control, and use of insect repellents. (See "Prevention of arthropod and insect bites: Repellents and other measures".)

International travelers to areas where plague is endemic or where an epidemic is ongoing are generally at low risk for infection, but they should take the same precautions about reducing potential exposure.

Infection control in health care settings — Patients suspected of having any form of plague should be placed on droplet precautions until pneumonia has been ruled out and sputum cultures are negative. For individuals with pneumonic plague, droplet precautions should be continued until antibiotic therapy has been given for at least 48 hours and there is evidence of clinical improvement. Airborne precautions are typically not necessary but are reasonable during aerosol-generating procedures in patients with pneumonic plague.

Otherwise, standard precautions are sufficient. Mask, eye protection, and a face shield are suggested for any procedure that could result in splashes of body fluids.

In crisis settings where personal protective equipment is not available, pre-exposure prophylaxis with antimicrobials may be reasonable for health care providers and first responders if supplies are sufficient.

Post-exposure prophylaxis — We suggest post-exposure prophylaxis for individuals with unprotected face-to-face contact (ie, within one to two meters) of patients with known or suspected pneumonic plague who have not received at least 48 hours of effective antimicrobial therapy [1]. Agents that can be used for post-exposure prophylaxis include the following:

●Doxycycline 100 mg orally twice daily (children: 2.2 mg/kg twice daily, maximum 100 mg per dose)

●Ciprofloxacin 500 to 750 mg orally twice daily (children: 15 mg/kg twice daily, maximum 750 mg per dose)

●Levofloxacin 500 to 750 mg orally once daily (children <50 kg: 8 mg/kg twice daily, maximum 250 mg per dose)

●Moxifloxacin 400 mg orally once daily (for children, moxifloxacin dose depends on age)

Post-exposure prophylaxis is generally given for seven days.

There are no data informing the benefit of post-exposure prophylaxis for plague. The rationale for administration is the high mortality rates associated with pneumonic and septicemic plague and the possibility of aerosol transmission. Antibiotic options are based on evidence on the treatment of plague.

Investigational vaccines — A killed whole-cell vaccine has been developed but is no longer commercially available in the United States. Data on the efficacy of the vaccine are limited; much of the experience has been with vaccination of military personnel deployed to areas endemic for plague, such as Vietnam [32]. The vaccination consists of a primary series of two injections one to three months apart, followed by a booster every six months for the duration of exposure [11].

The concern about plague as a bioterrorism agent has led to the development of a number of newer vaccines, some of which are undergoing clinical testing [1,48,49].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Plague".)

SUMMARY AND RECOMMENDATIONS

●Clinical syndromes – The three major syndromes associated with plague are bubonic plague (80 to 95 percent of cases), septicemic plague (10 to 20 percent of cases), and pneumonic plague (rare):

•Bubonic plague is characterized by sudden onset of fever, chills, weakness, and headache, followed by intense pain and swelling in a lymph node-bearing area (bubo). The acute bubo is exquisitely tender without fluctuation and often associated with erythema and edema (picture 1). (See 'Bubonic plague' above.)

•Septicemic plague manifests as fever and severe systemic illness without localizing signs or symptoms. Nausea, vomiting, diarrhea, and abdominal pain may occur. Hypotension, disseminated intravascular coagulation, and multiorgan failure develop in the later stages of the illness if untreated. (See 'Septicemic plague' above.)

•Pneumonic plague presents as a rapidly progressive pneumonia, with fever, chest pain, cough, and dyspnea. It can be primary (eg, acquired by inhalation of respiratory secretions or aerosolized droplets from infected animals or humans or by laboratory exposure) or secondary (eg, developing in the setting of bubonic or septicemic plague). (See 'Pneumonic plague' above.)

●Diagnosis – The diagnosis of plague is established by isolation of the organism in culture, by serologic testing or by rapid testing in appropriate settings. A high index of suspicion must be maintained for the timely diagnosis of plague. Patients with fever and painful lymphadenopathy should be questioned about travel to areas of endemic disease. (See 'Diagnosis' above.)

●Treatment – Antibiotic therapy should be administered promptly for plague. Plague is associated with mortality rates of 50 to 100 percent without treatment and 15 to 30 percent with treatment. Delay in antibiotic administration is associated with higher mortality. (See 'Importance of prompt therapy' above.)

First-line agents for plague include aminoglycosides (streptomycin or gentamicin), fluoroquinolones (levofloxacin, ciprofloxacin, moxifloxacin), and for bubonic plague, doxycycline. For most individuals with plague, we suggest streptomycin or gentamicin (Grade 2C). Clinical experience is greatest with aminoglycosides, but if there are concerns about toxicity (eg, advanced age, chronic kidney disease, use of other nephrotoxic agents), limited data suggest the other first-line agents are also effective. Doses are listed in the table (table 1). Duration of therapy ranges from 7 to 14 days. (See 'Preferred antibiotic regimens' above and 'Duration' above.)

Monotherapy is likely sufficient for most patients with naturally acquired plague because antibiotic resistance is uncommon. However, we suggest combination therapy with two of the agents listed above for treatment of plague when a bioterrorism event is suspected or documented because of the possibility of engineered resistance (Grade 2C). Combination therapy is also reasonable for those with severe septicemic or pneumonic plague. (See 'Preferred antibiotic regimens' above.)

●Infection control – Patients suspected of having any form of plague should be placed on droplet precautions until pneumonia has been ruled out, sputum cultures are negative, and at least 48 hours of effective antimicrobial therapy has been administered. (See 'Infection control in health care settings' above.)

●Post-exposure prophylaxis – We suggest post-exposure prophylaxis for individuals with unprotected face-to-face contact (ie, within two meters) of patients with known or suspected pneumonic plague who have not received at least 48 hours of effective antimicrobial therapy (Grade 2C). Appropriate agents for prophylaxis include doxycycline, levofloxacin, ciprofloxacin, and moxifloxacin. (See 'Prevention' above.)