A recent Institute of Medicine report, "Emerging Infections: Microbial Threats to Health in the United States" [2], called attention to the complacency that has developed regarding infectious diseases, highlighted the risk for importation of diseases that develop in remote parts of the world, and stressed the need to strengthen national and international infectious disease surveillance. The report cites the introduction of plague from other continents into Europe in the 14th century and into North America in the late 19th century. The recent epidemic of plague in India highlights the concerns expressed in this report.

In August 1994, an outbreak of bubonic plague was reported from the Beed district, a known plague-enzootic region [3] in Maharashtra State in western India. In late September, news came of an explosive epidemic of suspected primary pneumonic plague in the city of Surat in neighboring Gujarat State [4]. Hundreds of suspected cases and more than 50 deaths were reported from Surat. Press accounts described a mass exodus of hundreds of thousands of persons from this industrialized port city of nearly 2 million inhabitants [5]. By early October, more than 6300 suspected cases of plague had been reported from 12 Indian states, including Delhi, but only a few were considered laboratory confirmed, primarily by unvalidated serologic techniques [6].

Preliminary results of subsequent studies, done by the Indian government and an independent World Health Organization team, provided presumptive evidence of bubonic plague cases and enzootic plague activity in Maharashtra State (Gage KL, Chu MC, Centers for Disease Control and Prevention [CDC]. Personal communication). In Surat, reviews of clinical records and results of epidemiologic studies confirmed the reports of an outbreak there of acute respiratory illness characterized by fever, cough, hemoptysis, pulmonary infiltrates on radiographs, and a high fatality rate in the early course of the outbreak (Dennis DT, Orloski-Snider KA, CDC. Personal communication). Although some patients studied had antibodies to the plague bacillus, plague could not be established as the cause of this outbreak in the absence of the confirmed isolation of the causative agent. Although large numbers of suspected cases of plague were reported from New Delhi, Bombay, and Calcutta, no convincing evidence suggested transmission in any city but Surat. As of 25 October 1994, 693 suspected cases of plague with unvalidated serologic evidence of infection had been identified in six Indian states, and no new cases were being reported [7].

The initial reports of the outbreak caused considerable international concern about the risk for exportation of pneumonic plague from India. Official responses in various countries ranged from the enhancement of surveillance for ill travelers arriving at airports to the closure of borders, the embargo of flights to and from India, and the restriction of imports of Indian goods. Commerce between the United States and India, however, remained unrestricted during the epidemic. By enhancing surveillance at airports [8], providing written information about plague to all air travelers arriving from abroad, and promptly disseminating information on plague to clinicians and health department personnel, health officials in this country identified and evaluated for plague 13 travelers who had recently arrived from India with febrile illnesses and other conditions (CDC. Unpublished data) [7, 9, 10]. Two persons were found to have malaria, one had both dengue and malaria, and one had Salmonella bacteremia. No tourists in India are known to have contracted plague, and no patients with plague are known to have departed India during the crisis. The economic costs of emergency response systems implemented internationally during this crisis undoubtedly were substantial, and the resulting losses to the Indian tourism industry and other industries are expected to be staggering [5].

Plague is caused by infection with Yersinia pestis, a gram-negative bacillus carried by rodents and their fleas in parts of Asia, Africa, and the Americas [11, 12]. Most human plague is the bubonic form, transmitted by the bites of infected fleas; however, persons can also acquire plague by handling infected animals or by exposure to respiratory droplets from persons or animals with pneumonic plague. The incubation period for plague is 1 to 7 days (usually 2 to 4 days in primary pneumonic disease). Manifestations of the illness include acute onset of fever, chills, malaise, myalgias, and prostration, often with nausea. In particular, bubonic plague is characterized by painfully swollen regional lymph nodes (buboes) draining from the cutaneous inoculation site [11, 12]. Pneumonic plague is characterized by a productive cough, often with hemoptysis, and pulmonary infiltrates. Septicemic plague may result in endotoxic shock and disseminated intravascular coagulation without localized signs of infection. Plague meningitis is a less common presentation, and overlapping clinical presentations can occur.

Laboratory diagnosis of plague depends on cultural isolation of Y. pestis from tissues or body fluids, direct detection of antigens in tissues or fluids by fluorescent antibody staining, or detection of serum antibodies by passive hemagglutination assays [11, 12]. Visualization of characteristic bipolar bacilli in Giemsa- or Wayson-stained smears supports a diagnosis of suspected plague. The organism grows slowly on bacteriologic agar media and is infective for laboratory mice. Tests for plague are highly reliable when done by laboratory staff experienced with Y. pestis, but such expertise is usually limited to selected reference laboratories. All patients with suspected plague should be hospitalized and isolated, with precautions taken against transmission by respiratory droplets until pneumonia is ruled out or until specific antibiotic agents have been administered for at least 48 hours [12]. Specimens should be obtained promptly for laboratory diagnosis, chest roentgenograms should be done, and specific antibiotic therapy should be promptly initiated. Appropriate diagnostic specimens include sputum samples or tracheal aspirates for suspected cases of pneumonic plague; bubo aspirates for bubonic cases; and cerebrospinal fluid for meningitis cases. All specimens, including blood, should be smeared and examined with a Wayson or Giemsa stain, and all should be cultured. Acute- and convalescent-phase serum specimens should be obtained for antibody testing. Close communication between clinicians and the diagnostic laboratory is essential when a diagnosis of plague is being considered.

Streptomycin is the drug of choice for treating plague. Tetracycline is also highly effective, and gentamicin can be used when streptomycin is not readily available. Chloramphenicol is the preferred treatment for plague meningitis [11-14]. Prompt treatment can reduce overall plague mortality from 60% to 100% to less than 15%. Antibiotic prophylaxis is indicated for persons who have had face-to-face contact or who have occupied a closed space with a patient who has pneumonic plague. Recommended prophylactic drugs include tetracycline in adults and older children and sulfonamides in children 8 years of age and younger; chloramphenicol is also effective [14]. Prophylactic therapy should be continued for the duration of the potential exposure plus an additional 5 to 7 days. Routine travel to plague-endemic countries presents a low risk for infection with Y. pestis; therefore, antibiotic prophylaxis for plague is rarely indicated for travelers to these countries. Inactivated plague vaccine is apparently ineffective against primary pneumonic plague [15], and direct evidence for its effectiveness against bubonic plague in humans is unavailable [16]. It is potentially useful for persons who anticipate a visit to an enzootic plague focus (for example, field biologists studying plague), but it is not recommended for immediate protection during epidemics because a maximal antibody response requires the administration of multiple doses over several months [11, 12].

In 1992 (the most recent year for which complete data are available), human plague was reported in nine countries (Brazil, China, Madagascar, Mongolia, Myanmar, Peru, the United States, Vietnam, and Zaire) [17]. In addition, cases of bubonic plague have been recently reported in Malawi, Mozambique, and Zimbabwe. In India, large plague epidemics occurred during the first half of this century and resulted in millions of deaths; however, until the recent epidemic, the last laboratory-confirmed cases of human plague in India were reported in 1966 [17-19]. Sporadic cases of human plague occur annually in the western United States, particularly in rural and suburban regions of New Mexico, Arizona, and Colorado [20], after exposure to the infected fleas of wild rodents, contact with infected animal carcasses, or (rarely) droplet transmission from domestic cats with pneumonic plague. The extensive yet focal geographic distribution of Y. pestis is a reminder that when patients are evaluated for infectious diseases, the importance of a detailed travel history cannot be overemphasized.

The extreme measures taken by some persons and governments in response to the initial recent reports from India can largely be attributed to the widespread impression that pneumonic plague is not only deadly but also highly contagious in all circumstances. The latter impression, however, is not supported by the evidence. Person-to-person transmission of Y. pestis occurs by exposure to respiratory droplets from a patient with pneumonic plague who is coughing at close range (probably within 2 meters) [12]. Reported rates of secondary transmission of pneumonic disease have varied widely in different plague epidemics [15]. In epidemics with reportedly high secondary transmission rates, poverty and overcrowding in households was the rule. In contrast, in the plague-endemic areas of the western United States, no secondary plague cases resulting from person-to-person spread have been reported since 1925 [12], despite the occurrence of at least 37 pneumonic plague cases in the interim (including at least six primary pneumonic cases) (CDC. Unpublished data) [20]. In many of these cases, diagnosis and treatment were delayed, resulting in many potentially exposed persons with close contact and hundreds of persons with casual contact who received delayed or no prophylaxis. Thus, the risk for secondary plague transmission in the United States appears to be low. Clearly, the contagiousness of pneumonic plague deserves further study that includes a consideration of sociobehavioral, climatologic, and host-specific factors.

Important lessons can be learned from recent events in India. When an outbreak of uncertain cause occurs, appropriate specimen collection and diagnostic testing for potential causative agents is of paramount importance. With several effective antibiotics widely available with which to treat and prevent clinical plague, draconian measures in response to reports of plague outbreaks are unnecessary. Instead, a measured response based initially on a rapid but thorough assessment by a multidisciplinary team of experts (including clinicians, epidemiologists, entomologists, mammalogists, and microbiologists) is needed. The capacity for such a response depends directly on the quality of the public health infrastructure, including effective surveillance and laboratory systems, epidemiologic response capability, and vector and vertebrate host control programs [21]. Clinicians play an increasingly important role in recognizing illnesses and syndromes that require a public health response and alerting health departments to the need for prompt investigation [22]. Rational responses to future international public health crises depend on improving linkages between clinicians and public health professionals in all countries.

Although plague is of obvious historic importance [1] and continues to be a global threat, little research is currently being done on the subject. In many countries, including the United States, few clinicians, scientists, or laboratory or public health personnel are plague experts. In all plague-endemic countries, cost-efficient animal-based serosurveillance programs are needed to define and track the geographic distribution of Y. pestis, and extensive ecologic studies are needed to define and control enzootic and epizootic transmission cycles that put humans at risk [23]. The pathogenesis of plague and determinants of the virulence of Y. pestis should be studied anew using the most up-to-date laboratory methods. The potential for the emergence of resistant strains of Y. pestis related to widespread antibiotic use during pneumonic plague epidemics should be studied. Simpler and faster laboratory tests for plague are needed, and these should be made available for use by local or regional laboratories in developing countries in which plague is enzootic.

Largely because of the efficiency of modern air travel, microbial pathogens have shown an ever-increasing disrespect for international borders [24]. With the current decrease in political and trade barriers worldwide, this trend will certainly continue. Thus, it is imperative that the United States maintain multidisciplinary expertise in, and that clinicians remain alert for, "exotic" infectious diseases, including tropical diseases. The recent events in India underscore the need for expanding expertise and opportunities for training in arthropod-borne infectious diseases in this country and internationally [25] and for strengthening clinical, scientific, laboratory, and public health expertise in microbial diseases that are rarely recognized in the United States. The CDC has developed a plan to address these critical emerging infectious disease issues in collaboration with partners in state and local health departments, academia, clinical medicine, clinical laboratories, and international organizations [26]. The recent plague experience in India provides a clear example of the high price of ignoring global microbial threats [27].