Plague in Madagascar— Maybe Closer, Maybe Soon?

abstract & commentary

Synopsis: Recent reports from both the highlands and port cities of Madagascar point to an ongoing epidemic of urban and sylvatic plague in Madagascar. Bubonic plague may be difficult to recognize clinically and progression to pneumonic plague not only increases mortality but also transmission to unwary contacts.

Source: Ratsitorahina M, et al. Epidemiological and diagnostic aspects of the outbreak of pneumonic plague in Madagascar. Lancet 2000;355:111-113.

This report describes an outbreak of pneumon-ic plague in Madagascar, which occurred during 1997, when an index patient with secondary pulmonary disease infected a traditional healer and his family. Upward of 2000 cases of plague had been reported annually in Madagascar recently, but only a small percentage had been pneumonic. Most had been bubonic, but in October 1997 eight patients with suspected pneumonic plague were transferred to a provincial hospital for treatment with streptomycin.

Two methods were used to detect F1 antigen: an immunocapture enzyme-linked immunoassay (ELISA) and a dipstick assay from the Naval Medical Research Center in Bethesda, Maryland. The latter is a 15-minute one-step assay, which proved to be quite specific and rapid for F1 antigen detection (so-called fraction 1 glycoprotein surface antigen detection).

The index case was a woodcutter from a district 90 km north of the capital city of Antananarivo who initially presented with both fever and tender axillary adenitis. He consulted the traditional healer in his village as his course progressed over one day to clinical pneumonic plague, including chest pain, blood-stained sputum, and cough. He died the following morning but only after the healer had incised the patient’s abdomen and sucked out some blood as part of the healing process. The healer became ill in three days and died within seven days; his wife and son followed soon after. Further exposures to the healer and at funeral ceremonies resulted in 18 cases, eight of whom died. The rate of infection in this remote, previously unexposed population was 8.4%, which represented a fairly good estimate of the risk of spreading pneumonic plague in these remote villages, having not seen plague for 50 years. Not all patients with pneumonic plague are expected to be bacteremic, hence, 10 patients, both treated and untreated, were tested for Yersinia pestis in sputum by culture and the organisms could be isolated from only two, whereas the F1 dipstick detected antigen in nine and the F1 ELISA detected antigen in eight.

This epidemic occurred in the remote central highlands of Madagascar, where plague is endemic, transmitted by oriental fleas living upon rats and shrews, the animal reservoirs. The use of F1 antigen assays as a mainstay for the diagnosis of suspected cases of pneumonic plague and for early chemoprophylaxis of infected contacts was evident in this outbreak.

Comment by Frank J. Bia, MD, MPH

No sooner had this report appeared in the Lancet early this year when Madagascar appeared in the popular press in a lead article written for the Science section of the New York Times.1 The wreck of the flagship, Adventure, which had belonged to the infamous 17th century privateer, William Kidd, had just been located in waters just off Madagascar’s east coast on the shores of Isle Ste Marie. This tropical isle and its inlets of sandy beaches was a perfect place for ships to be hauled ashore and turned on their sides to have barnacles scraped from the hull, a process known as careening—essential maintenance if pirate ships were to travel as fast as possible over open waters. Kidd took advantage of this anchorage when he turned his back on a royal commission from the British crown to fight piracy, and was transformed into a pirate himself. Ultimately, William Kidd was hanged twice in London during 1701 (the rope broke on the first try).

Yet one more reason for adventure travel to Madagascar—a destination with great allure. As a result, travel medicine consultants must become more aware of it. Forestry students, divers, archeologists, and bird watchers are among those who must be advised about the dangers posed by both urban and sylvatic plague in Madagascar. Yet, it wasn’t always so.

Plague did not become established in Madagascar until the last great pandemic of 1894, which began in Hong Kong and spread rapidly over five continents. Y. pestis arrived in Madagascar during this third pandemic on ships from India, which appeared in November 1898. By 1921, plague had reached the high plateau regions, causing an outbreak of pneumonic plague in Antananarivo, then disappearing from seaports and becoming established at altitudes above 800 meters throughout this region.2

Silent until 1994, plague reappeared in epidemic form in neighboring regions such as Mozambique, Malawi, and India. Nor has this organism proven to be a particularly stable one, genetically speaking. In less than a century, the original strain that had been introduced into Madagascar has undergone chromosomal rearrangements, leading to the emergence of new ribotypes, based upon analysis of ribosomal RNA genes. It is not known what the selective advantage of such new variants may be and how they may relate to certain geographic environments, but Madagascar now has the distinction of harboring a particularly dangerous strain of Y. pestis.3

In 1995 a clinical isolate of this organism was recovered in Madagascar from a 16-year-old boy with an inguinal bubo, high fever, delirium, and prostration. The multi-drug-resistant organism, Y. pestis 17/95, carried a plasmid, which could be transferred by conjugation to other isolates of this organism. It carried resistance to antibiotics commonly used to treat plague such as streptomycin and the tetracyclines. In addition, the organism produced a beta-lactamase, mediating resistance to ampicillin and an acetyltransferase, causing resistance to chloramphenicol. It was not sensitive to sulfonamides, thus not allowing for synergy, but it retained sensitivity to trimethoprim, which may have accounted for this fortunate boy’s recovery. This strain also retained sensitivity to the quinolones, cephalosporins, and other aminoglycosides. Madagascar’s plague surveillance has been extensive, yet no such isolates had been identified between 1926 and 1995.

The appearance of multi-drug-resistant plague is an ominous event that could lead to an unpleasant emergency patient presentation in your office setting, as was recently reported by Dr. Martin Wolfe at his Travelers Medical Service in Washington, DC.4 He treated a 47-year-old woman who had been working as a mammalogist in the La Paz District of Bolivia and had been placed on ampicillin in Bolivia for severe headache, chills, fever, myalgias, and swelling in her right axilla but without relief. Her axillary swelling increased to the point where she could no longer move her right arm effectively and Wolfe saw her on arrival in Washington at his office. He found her to be febrile with a 2.5-cm fluctuant lymph node in her right axilla. Immediate hospitalization and aspiration of the node revealed gram-negative bipolar organisms, which were evident on a Wayson stain. Y. pestis grew from cultures of the node aspirate and she recovered following a 10-day course of streptomycin, until now the standard drug for treatment of all isolates of this organism. The patient had been skinning rice rats as part of her work in Bolivia and then crushing their fleas with her fingers, the likely source of her infection. This case represented the first recognized imported plague into the United States since 1926—and surely not the last, as this emerging disease continues to reappear, but now with potentially altered antibiotic sensitivity patterns, as was the case in Madagascar.

Plague had never entirely disappeared from Madagascar after its introduction by steamboats from India in 1898, but only 20-50 cases per year were reported until 1989.5 (See Figure.) Since then, a steady rise in the number of suspected plague cases has been reported, now reaching 800 to 1500 per year. No longer limited to the highland regions, plague has also reappeared in the northwest coastal town of Majunga. The introduction of both F1 antibody and F1 antigen immunodiagnostic tests has increased the number of confirmed cases two- to threefold because of their greater sensitivity. Perhaps analogous to the recent West Nile virus outbreak in New York City that was preceded by increased deaths among the city’s crow population, every human outbreak of plague in Madagascar has been preceded by large numbers of rat deaths. Shrews are also infected with the vector flea, X. cheopis, and may be the reservoir for maintenance of plague between epidemics.

Travelers to endemic regions of Madagascar and other geographic foci of plague activity should be instructed regarding the presence of epizootic plague,6 avoidance of sick or dead animals, and use of repellents and insecticides, gloves, and protective clothing. Prophylactic treatment with tetracycline for seven days (2 g/d) can be given to persons with close exposure to patients with pneumonic plague or with high-risk animal exposure. Doxycycline may be more efficacious than other tetracyclines, but there are no comparative evaluations among the tetracyclines. The role for the current killed whole-cell plague vaccine is limited and it does not fully protect against primary pneumonic plague. It requires three primary inoculations, and booster doses as frequently as every six months may be necessary.6 Should a patient arrive on your doorstep with suspected plague, respiratory isolation is appropriate if pneumonic plague is suspect. The newer immunodiagnostic tests for F1 antigen detection should be considered and no assumptions regarding sensitivity of the organism to previously first-line agents such as streptomycin or chloramphenicol should be made if future reports from Madagascar or other endemic regions indicate spread of multidrug resistance. (Dr. Bia is Professor of Medicine and Laboratory Medicine, Co-Director, Tropical Medicine and International Travelers’ Clinic, Yale University School of Medicine, New Haven, CT.)


1. Broad WJ. Seeking pirate treasure: Captain Kidd’s sunken ship. New York Times. February 6, 2000;F1, F6.

2. Guiyoule A, et al. Recent emergence of new variants of Yersinia pestis in Madagascar. J Clin Microbiol 1997; 35:2826-2833.

3. Galimand M, et al. Multidrug resistance in Yersinia pestis mediated by a transferable plasmid. N Engl J Med 1997;337:677-680.

4. Wolfe MS, et al. Bubonic plague—An imported case. Am J Trop Med Hyg 1999;61:227-228.

5. Chanteau S, et al. Plague, a reemerging disease in Madagascar. Emerging Infect Dis 1998;4:101-104.

6. Denis DT. Plague. In: Guerrant RL, Walker DH, Weller PF, eds. Tropical Infectious Diseases. Philadelphia, Pa: Churchill Livingstone; 1999:506-516.

Which of the following statements regarding human plague is true?

a. Pneumonic plague is generally not transmitted from person to person, but by septicemic spread from primary bubonic plague.

b. Most isolates of Y. pestis from Madagascar, India, and Mozambique are now resistant to the aminoglycosides.

c. Plague in Madagascar is no longer confined to the highland plateau, but has also appeared in coastal regions.

d. Resistant plague emerging from Madagascar would best be treated with a synergistic combination of folic acid antagonists consisting of sulfonamides and trimethoprim.

e. Fluoroquinolones will have no future role in the treatment of pneumonic plague.