U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content







AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window






Chapter 8

Table of Contents

Chapter 8

Bacterial Diseases

Colonel Dan C. Cavanaugh, MSC, USA (Ret.), Colonel Francis C. Cadigan, Jr., MC, USA, Major James E. Williams, MSC, USA, Colonel John D. Marshall, Jr., MSC, USA, Colonel William L. Moore, Jr., MC, USA, Carl R. Guiton, M.D., and Brigadier General Andre J. Ognibene, MC, USA

Section I. Plague

Colonel Dan C. Cavanaugh, MSC, USA (Ret.), Colonel Francis C. Cadigan, Jr., MC, USA, Major James E. Williams, MSC, USA, and Colonel John D. Marshall, Jr., MSC, USA


Plague is an ancient disease, innumerable epidemics of which have resulted in a total mortality perhaps not exceeded by any other infectious disease. The persistence of plague in natural foci throughout the world and a succession of outbreaks during the past two decades present clear evidence that the technological advances of the 20th century are still inadequate to eradicate it.

W. J. Simpson (1905, p.37) remarked that plague is brought in the train of armies or of commerce. The relationship of plague to warfare is not remarkable, as the devastation and disruption common to military operations often create favorable conditions for the proliferation of rats and fleas in areas where human beings are concentrated under unsanitary conditions. The history of plague is given extensive treatment in the works of various authors (Beasley 1969; Caten and Kartman 1968; Hirst 1953; MD-PM7; Pollitzer 1954,1960; Simpson 1905; Wu Lien-teh 1911, 1926, 1959; Wu Lien-teh et al. 1936). Its impact on military operations is a matter of record.

The first plague epidemic which can be identified with some certainty was the Philistinian plague of 1320 B.C. in which the Philistines, returning from a war with the Israelites, were stricken with the disease; following this episode, plague became established in the Levant. Troops operating in the endemic region of Constantinople were considered by Procopius to be involved in the first


plague pandemic, the Justinian plague, which occurred in the 6th century (Hirst 1953, p. 10).    

The second great pandemic, which resulted in the deaths of over 25 percent of the population of the Western World, was first noted in the city of Caffa on the Black Sea in 1346 (Hirst 1953, p.11). The Tartar armies beseiging the city experienced a devastating plague outbreak, and the victims' bodies were catapulted into the city, whereupon the disease appeared within its walls. Individuals fleeing the area in ships carried to Europe the "seeds of infection" that initiated a sequence of epidemics now known as the Black Death.

In the 18th century, the disease reappeared as a military problem. Frederick the Great, in 1745, was forced to conceal from his troops that the "foul fever" devastating the ranks was plague. Troops of Catherine the Great, return ing from operations in the Balkans in 1769, initiated a destructive plague epidemic which ravaged Moscow for 2 years. French military operations in Egypt were impeded by plague epidemics, one of which caused them to abandon an attack on Alexandria in 1798 (Hirst 1953, p.77).   

The third and most recent pandemic, 1894-1920, began when Chinese troops were deployed in remote Yunnan Province, an endemic area, to suppress a Muslim rebellion. Military traffic to and from the area resulted in epidemics which quickly involved coastal cities. Aided by modern transport, plague was introduced into nearly every country in Asia. In India alone, an estimated 12 million people died before the disease retreated to its permanent foci (Hirst 1953, p. 145).

The third pandemic and the period following it have been characterized by intensive research on the etiology, epidemiology, treatment, and control of plague. The plague bacillus was isolated and described independently by Kita-sato and Yersin in 1894. The association of the disease with rats and the incrimination of the flea as a vector were demonstrated by Liston, an officer of the Indian Medical Service and a member of the Indian Plague Commission from 1898 to 1914 (Liston 1903; Mollaret 1963, p.1177). The vital concepts of rat and flea control emerged from the studies of this commission.

Haffkine, working in Bombay during this time, developed the plague vaccine that bears his name. The Haffkine vaccine was shown to be effective in reducing both morbidity and mortality and became an important feature of control programs. However, studies on pneumonic plague in Manchuria demonstrated the requirement for other techniques to control this fulminating form of the disease, which did not involve rats or fleas but rather direct manto-man transmission from index cases infected by handling wild rodents. Changes in construction and fittings resulting in ratproof ships, together with the fumigation of premises and vessels with equipment developed by Clayton, were highly effective in reducing the territorial spread of the disease.

Although plague had subsided, World War II initiated renewed concern. A killed plague vaccine was developed and administered to U.S. troops deployed in endemic areas, and penicillin and the sulfonamides were studied to ascertain their value in treating the disease among indigenous people (MD-PM7, pp. 94-97). Fortunately, however, plague did not occur in U.S. troops in endem-


ic areas. DDT was developed and introduced during the war, and plague control became synonymous with flea control.

Following World War II, new concepts and technology contributed to a quantum leap in plague control. Improved bacteriological and serological techniques permitted the detection and delineation of plague foci in both com mensal and wild rodents (Cavanaugh et al. 1965, 1970). Insecticides and rodenticides were developed which reduced the threat presented by these foci to manageable proportions (Pollitzer 1954, pp. 529-52; 1960, pp. 367-70). Advances in immunology led to the introduction of promising new vaccines (Meyer 1970, 1971). Of prime importance, the sulfonamides and, in particular, streptomycin and tetracyclines proved effective in the treatment of plague in all its clinical forms (McCrumb, Larson, and Meyer 1953; McCrumb et al. 1953; Simeons and Chhatre 1946; Smadel et al. 1952).

Plague has never posed a serious problem to the U.S. military thanks to an understanding of the epidemiology of the disease. Effective control measures have been incorporated into preventive medicine programs as rapidly as they have been developed. The prudence of this policy was amply demonstrated in Vietnam, where fewer than a dozen cases of plague were diagnosed in American personnel serving in an area of intense epidemic activity (Cavanaugh, Dangerfield et al. 1968; Feeley and Kriz 1965; Marshall et al. 1967; OM-PHD).


Plague results from infection with the gram-negative bacterium Yersinia pesos [Pasteurella pestis]. Clinical forms of the disease include bubonic plague, almost always initiated by the bite of an infective flea; primary septicemic plague, which bypasses massive involvement of the lymph nodes; and pneumonic plague, primary or as a complication of bubonic plague. Hematogenous dissemination of the plague bacillus results in pulmonary involvement. Primary pneumonic plague infection is obtained either by handling infectious material or by direct interhuman transmission of Y. pestis in aerosols.

Fleas, the biological vectors of Y. pestis, following a blood meal on a bacteremic rodent, develop a fibrinoid mass that retains plague bacilli in the proventriculus, an organ resembling a portion of the human esophagus. Growth of these plague bacilli gradually occludes the proventriculus and prevents the passage of food material into the stomach, resulting in a "blocked" flea (Bacot and Martin 1914; Cavanaugh 1971). Blocked fleas bite, suck blood into the proventriculus, and then regurgitate it along with plague bacilli into the host, initiating plague infection. Estimates of the number of bacilli introduced by the flea's attempts to feed range from 11,000 to 24,000 organisms (Cavanaugh and Randall 1959).

Epidemiological evidence indicates that the majority of bubonic plague cases occur when ambient temperatures are under 28 0C, when, as shown by laboratory studies, Y. pestis organisms in the stomachs of blocked fleas are in a phagocytosis-sensitive condition (Bacot and Martin 1914; Cavanaugh and


Randall 1959). Therefore, the majority of plague bacilli introduced by fleabite are probably phagocytized.Phagocytosis-sensitive bacilli that are ingested by some mononuclear phagocytes, however, are not destroyed but reproduce within the cells, elaborating the factors that render them resistant to further phagocytosis and destruction by neutrophils at the time of their release from the infected cells (Cavanaugh and Randall 1959). Once established, the phagocytosis-resistant Y. pestis bacilli produce a rapidly progressing disease. Interhuman aerogenic transmission of phagocytosis-resistant Y. pestis explains the fulminating nature of epidemics of primary pneumonic plague.

Plague bacilli cultivated in vitro at temperatures under 28 0C are sensitive to phagocytosis and destruction by neutrophils. Bacilli incubated at 370C, however, produce at least three antigens which inhibit phagocytosis: the Fraction I antigen, which is the specific capsular antigen of Y. pestis, and the V and W antigens (Burrows 1955, 1957, 1960; Burrows and Bacon 1956). Fully virulent organisms have the capacity to produce these antigens, in vivo, in mammalian hosts (Burrows and Bacon 1954).

The plague bacillus contains several potent toxins or virulence factors other than those described above. A murine toxin, lethal for mice but not for rabbits, inhibits the respiration of heart mitochondria and induces alterations in the ST segments of electrocardiograms of susceptible rats inoculated with it (Rust et al. 1963). A lipopolysaccharide endotoxin of Y. pestis exhibits classical endotoxic biologic properties, in particular, the capacity to produce localized and generalized Shwartzman reactions in rabbits (Albizo and Surgalla 1970). A coagulase of Y. pestis produces firm clots in plasma when tests are conducted at temperatures under 28 0 C; above this temperature, fibrin apparently is destroyed rapidly by a fibrinolytic factor of Y. pestis (Cavanaugh 1971). Observations in patients indicate that several or all of the known virulence factors of Y. pestis may contribute individually or in concert to the pathophysiology of human plague.

Poland (1972, p.1143) states: "Progressive heart failure may develop in severe infections, and sudden death from cardiac failure has occurred even during convalescence." Winter and associates (1971, p. 383) noted that occasionally patients succumb following apparently adequate therapy. Y. pestis cannot be isolated from tissues of these patients at autopsy; it is believed that the potent toxins of the plague bacillus may cause death in a manner analogous to the Herxheimer reaction. Clinical and postmortem observations indicate that a generalized Shwartzman phenomenon, mediated by the lipopolysaccharide endotoxin of Y. pestis, plays a major role in the pathogenesis of plague (Finegold 1968; Butler 1972).

Many patients display evidence of disseminated intravascular coagulation, such as low platelet counts, prolonged partial thromboplastin times, positive ethanol gelatin tests, fibrin thrombi in purpuric lesions, and, in some in stances, acute cor pulmonary secondary to pulmonary thrombosis (Butler 1972, p. 274). At autopsy, fibrin thrombi found in the glomerular capillaries and elsewhere in the tissues of children have suggested that disseminated intravascular coagulation had occurred during the course of the disease (Finegold 1968; Finegold et al. 1968).



The risk of plague is greatest in concentrations of human beings living under unsanitary conditions in the proximity of large commensal or wild rodent populations infested with fleas that bite both man and rodents. If the climate is suitable and the plague bacillus is present or is introduced into the local rodent reservoir, an outbreak will ensue. The largest plague outbreaks have been associated with commensal Rattus species infested with the Oriental rat flea, Xenopsylla cheopis. This flea enjoys a cosmopolitan distribution, although the largest populations are found in tropical areas. It is wise to remember, however, that all fleas should be considered dangerous in plague endemic areas.

The basic epidemiology of fleaborne plague, which is essentially a disease of rodents, nearly always involves transmission of the plague bacillus from rat to rat and from rat to human being by the bite of an infected flea (Liston 1903, 1904, 1905; Lowson 1898, p. 247; Ogata 1897; Simond 1898). Interhuman infection or infection from fleas infesting bubonic plague patients is also possible (Swellengrebel 1967; Bahmanyar 1972; Pollitzer 1954, pp. 385-87).

Handling of infected rodents or exposure to droplets of infectious material produced by plague patients with pulmonary complications may lead to large, explosive epidemics of pneumonic plague (Wu Lien-tieh 1911, 1926; Pollitzer 1954, p.245). Approximately 5 percent of bubonic plague patients develop a potential for the aerogenic transmission of the plague bacillus (Poland 1972, pp.1141-48).

The epidemiology of primary pneumonic plague is not well understood. Most severe epidemics have occurred in areas where the climate is relatively cool. Primary plague pneumonia is rather rare in the Tropics, even in the presence of fulminating epidemics of bubonic plague, since high temperatures and humidities are unsuitable for the survival of Y. pestis in aerosol clouds. On the other hand, low relative humidity also is associated with the rapid death of plague bacilli in aerosols. Cool weather, moderate humidity, and close contact between susceptible individuals appear to be most favorable for epidemics of primary plague pneumonia (Burmeister, Tigertt, and Overholt 1962; Winter et al. 1971, pp. 377-87).

The observation that asymptomatic contacts may harbor virulent plague bacilli in their throats requires further evaluation to ascertain the role of such individuals in the epidemiology of plague pneumonia (Marshall, Quy, and Gib son 1967; Legters, Cottingham, and Hunter 1970). Future investigators also should consider the possibility that other respiratory infections superimposed on this "carrier state" might enhance the aerosolization of the plague bacillus and result in rapid dissemination.

Wild rodents are now considered to be the principal reservoir of plague, serving as sources of infection for both human beings and commensal rodents. Widespread infection in numerous species of small rodents and other wild animals in the United States and elsewhere throughout the world is maintained in natural or temporary foci. The plague bacillus is transmitted among rodents by various fleas infesting the animals (Macchiavello 1954; Pollitzer 1954, p. 337;


MAP 3.- Known and probable foci and areas of plague, 1969 (Source WHO Expert Committee

Pollitzer 1960, p. 357). Rodents may also become infected by digging burrows contaminated with infectious detritus (Mollaret 1963) or by cannibalism (Rust et al. 1972). Marmota bobak is a particularly notorious rodent, shown to be the initial source of infection in the great epidemics of pneumonic plague that occurred in Manchuria (Wu Lien-teh 1926; Wu Lien-teh et al. 1936).

Within a natural focus, the etiologic agent persists in wild rodents for considerable periods of time, and epizootics alternate with variable periods of quiescence. Temporary foci are territories that are seeded from permanent foci but, for unknown reasons, do not persist indefinitely. The epizootic process in each natural plague focus has its specific cyclic and periodic pattern, conditioned by the ecology of local rodent reservoirs and flea vectors. Each focus has distinct seasonal appearances of epizootic plague with characteristic peaks


Map 3. cont - on Plague, Fourth Tech. Rep. s. No. 447. Geneva: World Health Organization, 1970.)

of activity reflecting favorable local conditions (WHO-a).    

Epidemics of disease in humans are also distinctly seasonal. The initial outbreak, which may or may not succeed in establishing a permanent focus, may occur at any season of the year; when the infection has secured itself in the rodent population of an area, characteristic plague seasons develop. In general, bubonic plague is a disease of the cooler months in hot climates and of the warmer months in cooler climates. Major epidemics occur at temperatures between 15° and 27.5° C associated with vapor pressure deficits not in excess of 7.6 mm Hg (Greenwood 1911, 1913, 1935; Brooks 1917; Cavanaugh and Marshall 1972).

The incidence of bubonic plague is directly related to the prevalence of the flea vector. Both cold and hot, dry weather cause the disappearance of X.


cheopis. If hot weather is associated with a vapor pressure deficit in excess of 7.6 mm Hg, the lifespan of adult X. cheopis is reduced (Hudson et al. 1973). If the relative humidity falls below 65 percent, X. cheopis larvae cannot maintain a water balance suitable for maturation (Knulle 1967). Rat fleas are not effective vectors when temperatures exceed 27 0 C (Kartman 1969; Cavanaugh 1971).    

Data on the present distribution of known plague foci are shown on map 3. There is no evidence that new natural foci have been created in the recent past or that known natural foci have spontaneously become inactive or smaller in size, except for a few foci where intensive control measures have been employed. Many undetected foci may exist, particularly in Asia, Africa, and South America. Among known foci, those in Vietnam demonstrated the greatest activity in terms of human disease.

Rapid population growth and new development schemes are likely to bring more human beings into close contact with natural foci and create fresh problems in plague control, and military operations in such areas will probably create conditions favorable for the expansion of foci. Transportation of infected rodents and fleas from known foci may introduce the disease into plague-free areas. Although the ratproofing of ships has greatly reduced the risk of transporting infected rats and fleas to distant seaports, the new technique of shipping in containers poses a threat; transport of such containers by air may present a particular hazard.


Plague is a deadly disease which begins abruptly and can progress rapidly to death in a period of hours or of several days in 60 to 90 percent of patients unless therapy is initiated early (Poland 1971, 1972). Early recognition of the clinical syndrome is difficult because of the nonspecific nature of many of the early signs and symptoms, but delay of therapy while awaiting laboratory diagnosis might prove fatal because of the rapid progress of the disease.

The history of onset of symptoms is almost monotonously regular: virtually all patients have sudden onset of fever, chills, and headache (Butler 1972), usually followed within several hours by nausea and vomiting. The percentage of patients presenting with various manifestations follows:

Altered mentation - 26-38
Headache - 20-85
Chest pain - 13
Prostration or severe malaise - 75
Chills - 40
Vomiting - 25-49
Cough - 25
Abdominal pain - 18   

Within 6 to 8 hours after onset, patients become aware of pain at bubo sites if buboes exist. The sudden onset of symptoms, characteristic of septicemic diseases, is the clinical picture in 95 percent of cases, according to Mathis and Pons (Pollitzer 1954, p. 411). Simpson (1905, p. 262) states: "no disease except cholera manifests * * * so rapid a development of its symptoms and overwhelms or


prostrates the patient to the verge of death in so short a time." (Meningococcemia should probably also be excepted.)

Fever is variable, ranging from 99° to 106° F (37° to 41°  C). Low fevers are not necessarily a sign of milder disease since the fever may never reach 100° F (38° C) in septicemic plague. Fluctuating levels of fever are of little value in the prognosis, except that it is unfavorable when the pulse rises as the temperature falls or when a sudden fall is accompanied by signs of collapse (Simpson 1905, p. 268). With successful antibiotic therapy, the temperature often falls precipitously but is accompanied by a slowing of the pulse and obvious improvement in the general condition of the patient.

Few acute infections produce a leukocytosis of such magnitude in adults as does plague. Counts are frequently in the range of 20,000 to 25,000/mm3, and even 50,000/mm3 is not unusual. Exceptions occur, however, and counts as low as 4,500/mm3 have been recorded, even in the presence of proven bacteremia; some evidence suggests that semi-immune individuals tend to have lower levels of leukocytosis than do fully susceptible individuals. Later in the course of disease, abnormalities are noted in the chest X-ray and, occasionally, in the electrocardiogram, blood chemistry, and urinalysis. Careful evaluation frequently reveals that disseminated intravascular coagulation is present. Cultures of sputum or blood, and aspirates or biopsies of buboes, will confirm the diagnosis in approximately 72 hours, but to reduce mortality, therapy must be initiated much earlier on the basis of clinical suspicion. Abundant plague bacilli may be observed in stained smears of material aspirated from buboes, in peripheral blood, or in sputum (Simpson 1905). These procedures are recommended. However, the hazardous nature of these clinical materials must be emphasized.

Buboes usually are found only when the infection is acquired by fleabite or mechanical irritation. Tularemia may also be suggested by the presence of buboes; in serological studies of suspected cases of that disease, 4.5 percent could be diagnosed retrospectively as cases of plague (Sites, Poland, and Hudson 1972). Buboes usually are heralded by severe pain in the area where they will appear. The bubo itself will usually become palpable or visible within 24 hours. The pain is intense, and even a nearly comatose patient attempts to reduce pressure on the area and shield it from trauma. The lightest touch may elicit exquisite pain, which is characteristic, but occasional patients have only moderate tenderness. The position of the body usually gives an immediate clue as to the location of a bubo-for example, hip partially flexed for groin buboes or patient supine with arm held away from body for axillary buboes.

The most common location for a bubo is in the groin, involving inguinal or femoral nodes (fig. 46) or both. These lesions comprise over half of those seen. Axillary and cervical glands (fig. 47) are next in frequency of involvement, ac counting for another quarter to a third of cases. Virtually all lymph nodes of the body, including intrathoracic and abdominal nodes, are possible sites. Involvement of the latter may result in "acute surgical abdomen"; there are numerous records of patients operated on for this reason. As a bubo enlarges and suppurates, both pain and tenderness decrease. Swollen anterior cervical nodes and


FIGURE 46. - Left: Typical femoral bubo as observed in Vietnam. Right: Femoral Bubo showing drainage and early healing.


FIGURE 47. - Axillary bubo observed in acutely ill Vietnamese patient.

associated edema can cause sufficient pressure on the trachea to result in respiratory embarrassment or even death.    

Multiple buboes are not uncommon, although they may not be obvious, as in the case of iliac buboes secondary to inguinal buboes. Buboes may appear simultaneously at widely separated sites or may develop secondary to a primary bubo along the line of lymph drainage. Bilateral cervical buboes suggest pharyngeal and tonsillar involvement. Pharyngeal or tonsillar plague may produce classical buboes or may result in large swollen nodes similar to those seen in other severe bacterial infections, in which case misdiagnosis may occur. A pseudodiphtheritic membrane can also occur and cases may be misdiagnosed as diphtheria (Simpson 1905).

Bubonic plague is observed in all degrees of severity ranging from pestis minor or ambulatory plague, which may be self-limited, to severe disease in which a febrile and relatively toxic, though alert, patient may be moribund in a few hours. The gravity of the case appears to be correlated with the degree of altered mentation and restlessness, depression, anxiety, and distress "peste facies" or delirium and coma). Pneumonia and sepsis are grave manifestations. Observation of plague bacilli in smears of peripheral blood is a basis for a poor prognosis.

The most dreaded form of plague is that which involves the lungs. Pollitzer (1954, pp. 440-41) distinguishes three forms: one involving development of wellmarked pneumonic foci with consolidation and all the signs of bacterial


FIGURE 48.-Bloody sputum in advanced pneumonic plague. (Courtesy, Capt. F. R. McCrumb, MC.)

pneumonia; a transitory form with slight pneumonitis; and a form in which congestion and edema are marked and, although bacilli are found in the lower respiratory tract, no consolidation occurs. These are obviously partitions in a continuous spectrum of involvement but they are useful in classification. The first, the most severe form, is classic pneumonic plague. Primary pneumonic plague begins in the same fashion that severe bubonic plague does except that bubo pain does not appear. About 24 hours after onset of symptoms, a productive cough begins and a mucopurulent sputum rapidly changes to sanguinopurulent (fig. 48).

Hematogenous dissemination of the plague bacillus, as of any bloodborne pathogen, may result in the infection of any organ, leading to numerous complications. In some instances, especially in individuals receiving suboptimal therapy, purulent meningitis may develop. Primary plague meningitis without antecedent disease has occurred. Petechiae and ecchymoses can occur in such numbers and locations as to mimic severe meningococcemia, and indeed the microscopic lesions (fig. 49) are almost indistinguishable. The pathogenesis is similar, probably involving a generalized Shwartzman reaction (Butler 1972, p. 274), and the prognosis is equally poor. Large ecchymosis or purpura, especially on the back, is a common feature of terminal pneumonic or septicemic cases. Purpura resulting in gangrene of the extremities has occurred in treated patients.*


1 Col. Dan C. Cavanaugh, MSC: Personal observation.


FIGURE 49. - Left Lesion in plague purpura, low power magnification. Right: Skin lesion in plague purpura, high power magnification. (Courtesy, Col. H. G. Dangerfield, MC.)


At present, the preferred antibiotic therapy of the WHO (World Health Organization) Expert Committee on Plague is tetracycline. For the first 48 hours, this should be in large doses (4 to 6 g daily). It is better to use tetracycline in combination with streptomycin (0.5 g every 4 hours for 2 days, then 0.5 g every 6 hours until patient has improved). Chloramphenicol (50 to 75 mg/kg daily to a total dose of 20 to 25 g) can be substituted for tetracycline if indicated. The committee recommends sulfonamides not be used if any of the above antibiotics are available. This regimen should be followed for at least 10 days, and then throat and blood cultures should be repeated (WHO-a). Legters, Cottingham, and Hunter (1970, p. 651) have shown that, despite clinical cure and demonstrated sensitivity of the organisms to the drug used, bacterial cultures of the throat at 10 days may still be positive. Two reasons for therapy until cultures are negative are the possibility of a risk of spread from an asymptomatic carrier, and good evidence that plague meningitis not only occurs after "adequate" sulfonamide therapy for bubonic plague but may occur more frequently than in untreated cases (Landsborough and Tunnell 1947). There are not sufficient data to make the latter statement about meningitis following the antibiotic regimens suggested above, but the occasional persistence of Y. pestis in the throat certainly suggests the possibility of a similar situation.    

Isolation procedures are probably unnecessary in bubonic cases without pulmonary involvement, except for care in handling blood and discharges from the buboes. Discretion is indicated, however, since early pulmonary involvement or throat infection may not be recognized. Except in epidemic situations, bubonic plague patients should be kept in a separate ward or room. On the other hand, rigidly enforced, strict isolation must be practiced for patients with pulmonary involvement until throat and sputum samples are free of Y. pestis. Many secondary cases of pneumonic plague have occurred in nursing and medical staff, despite precautions. Protection of the staff is not limited to gowns, masks, and gloves but should include a safeguard for the eyes; Pollitzer (1954, pp. 431-34) lists seven reported cases in which the eye seemed to have been the portal of entry. Eyeglasses may be sufficient, but a shield is preferable.

Personnel may become infected via broken skin, mucous membranes, or the respiratory tract through accidents in the laboratory or when taking specimens. Manipulations with pipettes or syringes are particularly hazardous. There is a particular requirement to keep laboratory animals free from ectoparasites, especially fleas. Adequate identification of Y. pestis can be carried out at the clinical laboratory level, but this is not recommended because of the hazard involved. Suspicious cultures or specimens should be forwarded as quickly as possible to laboratories that are completely equipped to accomplish timely identification. Immediate reports to public health authorities are mandatory.


The laboratory diagnosis of plague infection is made by the isolation of Y. pestis from buboes, blood, or sputum. In addition, epidemiological studies often require the isolation of Y. pestis from rodent tissue or from fleas.


FIGURE 50.-Plague bacilli in clinical specimen of peripheral blood, Wayson stain. (Oil immersion.)

Smears of clinical material can be prepared under almost any circumstance, and the value of this procedure cannot be overemphasized. The plague bacillus can be demonstrated, often in great numbers, in specimens collected from patients. Smears of material aspirated from buboes, peripheral blood, and sputum are prepared, dried, fixed with alcohol (absolute methanol for 5 minutes), and then stained. While the Gram's stain is adequate to show the presence of gramnegative bacilli, Giemsa or Wayson stains are recommended for the demonstration of classical bipolar coccobacilli in the specimen (Poland 1972, p. 1141) (fig. 50). The organisms show pleomorphism.

In the laboratory, nutrient agar, blood agar and desoxycholate agar plates (particularly valuable when dealing with sputum specimens), and broth media should be routinely inoculated. Highly contaminated specimens, as well as most specimens of rodent tissues and fleas, should be given a preliminary passage through experimental animals before cultural isolation is attempted.

Following is a summary of the characteristics of Y. pestis which aid in the laboratory diagnosis of plague.

Microscopic Morphology

Yersinia pestis is a bipolar staining coccobacillus, 0.5 to 0.7 by 1.5 to 1.75 microns. Organisms usually occur singly or in pairs and are extremely pleomorphic. Y. pestis is gram-negative, nonacid-fast, and nonspore-forming. It is non-


motile in semisolid media. A capsule can be demonstrated in wet india ink preparations observed by dark field or by fluorescent antibody tests (Pollitzer 1954) when bacilli are incubated at 37 0 C. The demonstration of a capsule with fluorescent antibody tests (Moody and Winter 1959) is presumptively diagnostic, but absolute identification requires the cultural isolation of Y. pestis. Certain nonencapsulated variants of Y. pesos isolated from patients (Winter, Cherry, and Moody 1960) and some surface antigens of the closely associated Y. pseudo tuberculosis (Quan et al. 1965) present problems and indicate that fluorescent antibody tests should be interpreted with some caution.

Animal Pathogenicity

The organism is virulent for white mice, white rats, guinea pigs, and other rodents by the cutaneous, subcutaneous (recommended), intraperitoneal, and intranasal routes. White mice should not be used for tests with sputum. Post mortem examination of mice usually shows buboes and marked splenic enlargement with other nonspecific signs of generalized infection. The infection in rats and guinea pigs produces changes quite characteristic of plague: subcutaneous congestion, buboes in corresponding regional lymph nodes, enlargement of the spleen and liver with necrotic nodules, and pneumonic foci in the lungs (Baltazard et al. 1956). Y. pestis is observed in abundance in direct smears from all tissues. Certain strains of Y. pestis isolated from patients in Brazil are virulent for white mice but not for guinea pigs (Burrows and Gillett 1971).


Metabolism is aerobic or facultatively anaerobic. Optimal temperature is 28  0 C, with a range of 0 0 to 43 0 C; elaboration of the specific Fraction I capsular antigen requires incubation at 37 0 C. In broth, 24 hours at 37 0 C produces moderate growth with little or no turbidity. Floccular or granular deposit swirls up and disperses evenly, although not always completely, with shaking. Blood cultures should be subcultured on solid media after 48 hours. On agar media, 24 hours at 37 0 C produces very small 0.1- to 0.2-mm mucoid colonies, usually visible only in the initial streak. At 48 hours, colonies are considerably larger, 1 to 2 mm. With prolonged incubation, colonies continue to grow and assume a beaten copper surface. The center of the colonies is raised, and the periphery is flat with an umbonate edge. Older colonies often assume a "fried egg" configuration. Growth on desoxycholate agar does not appear until the second day, when small, reddish, pinpoint colonies are observed. Gelatin stab shows good filiform growth, confluent at top and discrete below, extending to the bottom of the stab line. No liquefaction occurs in 7 days at 22 0 C (Pollitzer 1954).

Biochemical Characteristics

No hemolysins are produced, although certain strains occasionally produce some "greening" of blood media. The organism is catalase positive and H2S


negative, and indole is not produced. Y. pestis is coagulase specific, rapidly coagulating citrated rabbit and guinea pig plasma in tests conducted at temperatures less than 27 0 C. Fibrin clots are rapidly destroyed by Y. pestis incubated at temperatures in excess of 28 0 C. Y. pestis is oxidase negative, and variable results with nitrate reduction tests are used to differentiate variants. The organism is urease negative (one exception has been noted) and methyl red positive; methylene blue is not reduced (Baltazard et al. 1956, pp. 457-509). The Voges-Proskauer reaction is negative, and litmus milk remains unchanged (Pollitzer 1954).

Sugar Fermentation

Acid, but no gas, is produced from glucose, maltose, mannitol, and salicin. No fermentation of lactose, sucrose, rhamnose, and melibiose occurs. Variable results with glycerol are used to differentiate races.


The plague bacillus is lysed by a bacteriophage at temperatures ranging from 21 0 through 37 0 C. The lysis is specific for Y. pestis at 20 0 C; at temperatures in excess of 25 0 C, occasional strains of Y. pseudo tuberculosis and Escherichia coli may demonstrate lysis in the bacteriophage tests (Cavanaugh and Quan 1953). Media containing blood should not be used in bacteriophage tests; some strains of Y. pestis are not lysed by bacteriophage when tested on blood agar plates. Plates of nutrient agar should first be seeded with Y. pestis before the application of individual drops of a bacteriophage suspension to the seeded area. Distinct plaques usually are visible in areas of otherwise confluent growth within 18 to 24 hours. Occasional variants of Y. pesos do not present a picture of complete lysis; scattered colonies of growth occur within the plaque.

Serological Identification

Demonstration of specific Fraction I antigen by gel diffusion precipitin techniques is diagnostic (Lawton, Fukui, and Surgalla 1960). Complement-fixing and hemagglutinating antibodies to the specific Fraction I antigen may be present in the serums of convalescent plague patients (Cavanaugh et al. 1970; Chen and Meyer 1954; Chen, Quan, and Meyer 1952). If available, specific Fraction I antiserums may be used in microslide agglutination tests to provide a rapid, presumptive identification of Y. pestis using selective colonies picked from blood agar as antigen.


Plague is essentially an anthropozoonosis, as the principal source of disease for man is contact with infected rodents and their flea ectoparasites. The attainment of a standard of living that allows the construction of ratproof buildings,


thus eliminating rat harborage, is the ideal means of long-term plague control. Unfortunately, this solution is not yet possible in many developing countries where urban plague exists.

Before World War II, the most important plague control measures were reducing the rat population and fumigating premises against fleas. A more rapid and direct method of flea control became feasible in the early 1940's with the introduction of chemicals such as DDT. The U.S. Army was the first to use DDT to control plague (in the African theater, 1944). A rational control program, however, should deal with both rodents and fleas. Specific actions for any area depend upon many factors, but in general, the control of bubonic plague involves interrupting the rodent-flea-man transmission cycle with pesticides and sanitation. By contrast, in pneumonic plague the man-to-man cycle of transmission by the respiratory route must be interrupted; therefore, casefinding, treatment of cases, and isolation and immediate treatment of contacts on the appearance of fever or other symptoms are the salient features of pneumonic plague control.

An epidemiological survey should be conducted to ascertain the source of the infection. Rodents and fleas should be collected and examined if possible. Serum surveys are particularly valuable in detecting plague foci in rodent populations (Cavanaugh et al.1970,1965; Rust 1971). Maps and graphs containing information derived from these investigations are valuable, but statistics of plague prevalence at district or state levels may be misleading. Study at the village level, block by block or house by house, may show that the disease is endemic throughout an entire area or that sharp, serious outbreaks are occurring in quite limited foci. The predominant clinical form of plague observed in the focus under consideration often is a valuable indication of possible sources of infection and thus of necessary preventive measures. For example, epidemics of bubonic plague usually are associated with commensal rats and fleas while epidemics of pneumonic plague often are associated with wild rodents or with individuals leaving wild rodent foci before the onset of clinical disease. A preliminary estimation of the number of patients involved and the total land mass or surface area of premises requiring control is logistically useful. 

In controlling outbreaks of bubonic plague, the first and most important step is the selection of an effective insecticide. The spectacular success of insecticides such as DDT was short-lived; less than 10 years after their introduction, it became evident that fleas in many regions had developed resistance to them. Resistance to DDT may be anticipated where the pesticide has been employed for malaria eradication. The choice of insecticide is dictated by the results of susceptibility tests of the fleas in the area under consideration, but in emergency situations, an insecticide of known efficacy which has never been utilized in the area should be selected.

Insecticides are applied as dusts. Thorough coverage of rodent-frequented sites, particularly harborages and runways, is necessary. In houses, dust should be applied to the bottom of all walls and on floors for a distance of 15 to 30 cm from the walls. Where the wall-roof junction of the dwelling is open, it should be applied along the top of the walls and along rafters where rat runways are evident. Other areas which may provide food or shelter for rodents, both indoors


and outdoors, should be dusted. Retreatment may be required after a period of time, especially if nonresidual insecticides are used. The frequency of treatment cycles depends upon the insecticide and the conditions to which it has been subjected. 

Rodents pick up insecticide dust on their feet, carry it back to their nests, and distribute it over their bodies through constant preening. Where the general application of dust may be hazardous, bait box techniques may be employed. Rodents entering locked bait boxes must pass through insecticide dust deposited in an outer foyer of the box to reach bait placed in an inner chamber. Both insecticides and rodenticides can be used (Kartman 1958, 1960).

Plague may persist in a rodent population in the absence of fleas by oral transmission, through cannibalism. Oral infection may play a significant role in the epidemiology of local plague epidemics by providing a mechanism for the persistence of the disease through interepidemic periods and may also account for sporadic cases of bubonic or pneumonic plague. Whenever possible, a coordinated program of rat abatement and flea control is recommended (Rust et al. 1972). Flea control, however, must be achieved before any rat poisoning activity, to prevent the appearance of large numbers of hostless rat fleas which may attack people and increase the prevalence of human plague.   

In the control of wild rodent plague, the same considerations apply. The results of large-scale programs to eradicate or control the infection have been discouraging. Such programs demand large investments in personnel, equipment, and pesticides and, as a rule, while effective for short periods of time, do not achieve long-term control. Treated areas are soon repopulated with both rats and fleas and frequently demonstrate renewed enzootic or epizootic activity. Routine surveillance of such foci should indicate when control efforts are required. In some foci, in critical areas, routine control programs should be instituted either to reduce disease in man or to prevent the conveyance of infected rats and fleas into urban areas. Typical areas needing continual attention and control are: notorious foci in which index cases of large pneumonic plague epidemics were infected; environs of ports, airports, or towns; military camps; sites of large-scale construction of new settlements, roads, or railroads in endemic foci; and areas where a definite threat to agricultural or other workers exists.

An evaluation of the control program is mandatory. The incidence of bubonic plague should show a dramatic decline following a successful program. The rat flea population should be greatly reduced or almost eliminated within 48 hours, and a corresponding reduction in the prevalence of the disease in rats, estimated by bacteriological and serological evidence, should occur. Seasonal trends in individual foci should be taken into consideration to avoid interpreting normal seasonal declines as indicating a successful control program.

Modern means of transport, especially containerized cargo, provide new opportunities for fast and widespread dissemination. Fumigation with hydrogen cyanide or methyl bromide is useful in controlling rodents or ectoparasites which might be transported in various goods from infected areas. Complete kills of adult X. cheopis can be achieved within 24 hours by placing dichlorvos resin


CHART 8.-Incidence of human plague in the Republic of Vietnam, 1906 to 1 September 1967 1

strips (one per 9 m3) in each container before locking and sealing it for shipment. Rodents can be controlled by bait boxes containing anticoagulant rodenticides. Aircraft present special problems. Rodents should be trapped, as poisoned rodents may die in an inaccessible area vital to the function of the aircraft. Parking aircraft in rodent-free or ratproof areas and loading only disinfested cargo are helpful. Flea control on aircraft is best achieved through the use of aerosol insecticides (WHO-b).    

Vaccination with one of the approved plague vaccines is recommended for individuals at risk of infection. The killed vaccines developed at the Haffkine Institute in Bombay and the USP vaccine developed at the George Williams Hooper Foundation of the University of California have proved to be effective in reducing the incidence of human plague; indirect evidence supports the conclusion that administration of the USP vaccine to troops in Vietnam was partly responsible for maintaining a remarkably low plague incidence there (Cavanaugh et al. 1974). Living, attenuated strains of Y. pestis used as vaccines in Java, Madagascar, and elsewhere have been credited with reducing plague morbidity and mortality. However, seed strains used in the preparation of these vaccines are subject to considerable variation when held in vitro, and vaccines


prepared from seed which is deficient in one or several antigens or virulence factors are likely to demonstrate diminished potency. Living plague vaccines prepared from seed having the requisite antigenic composition are also noted for reactogenicity (Meyer 1970).


Vietnam has been recognized as a plague-endemic region for three-quarters of a century (chart 8). The period 1898-1920 was characterized by several sharp epidemics, with sporadic cases occurring between them. The incidence of the disease then declined, although plague was considered to be endemic in the coastal region around the city of Phan Thiet. During World War II, a recrudescence was observed when French Indochina was occupied by the Imperial Japanese Forces. Medecin General Roberts, of the French Medical Services, then considered Lang Bian Plateau to be an endemic area. The disease apparently had moved along the coast to the north of Phan Thiet and branched off in the general area of Phan Rang toward Da Lat (Cavanaugh, Dangerfield, et al. 1968).   

Following the departure of the French forces from Vietnam as a result of the Indochina War, the political situation deteriorated. In all likelihood, public health also was neglected during this period, resulting in conditions favorable for the development of plague in territories surrounding previously established natural foci. After the cessation of monsoon rains in 1962, a disease characterized by fever, lymphadenopathy, and high mortality was reported near the cities of Nha Trang and Saigon. It soon occurred in the population of Nha Trang, and one American serviceman became infected in 1963. A suspicion that the disease was plague was confirmed when Yersinia pestis was isolated from the serviceman and several Vietnamese patients through bacteriological studies at 1'Institut Pasteur in Nha Trang (Cavanaugh, Dangerfield, et al. 1968; Feeley and Kriz 1965).

A committee consisting of personnel from the South Vietnam Ministry of Health, the United States and Vietnamese military, and USOM (U.S. Operations Mission) was formed to evaluate and deal with the plague emergency. Personnel and facilities of the 7th Medical Laboratory, Nha Trang, under the command of Maj. Eugene J. Feeley, MC, were used in diagnostic and field studies. Control measures, chiefly vector control with DDT, were initiated in and around the plague foci by the appropriate Vietnamese authorities. The U.S. military promptly began standard plague control measures, including vaccination with a killed plague vaccine.

Correspondence between concerned parties led to a decision to deploy in Vietnam a medical research team from WRAIR (Walter Reed Army Institute of Research). The mission assigned to the team by Col. (later Brig. Gen.) William D. Tigertt, MC, Commandant, WRAIR, was to establish a laboratory which could support various investigations. In October 1963, the initial contingent arrived in Saigon. Led by Lt. Col. (later Col.) Paul E. Teschan, MC, the team rapidly developed protocols for the study of plague, as well as cholera, diarrheal


diseases, dengue, and hepatitis. A Joint Committee for Pathological Research was established, which brought together the Vietnamese health authorities, AID (Agency for International Development), and the U.S. military services. The protocol for proposed plague studies was presented to and accepted by this committee (Cavanaugh, Do-Van-Quy, and Ky-Vinh-Thai 1964).   

In 1964, the only plague research laboratory in Southeast Asia was established. The joint facilities of 1'Institut Pasteur of Vietnam and the WRAIR team were used in both field and laboratory studies. Various other agencies, American and Vietnamese, civilian and military, supported the program, particularly the Army 20th Preventive Medicine Unit, Saigon, and the Navy Preventive Medicine Unit, Da Nang. At the plague research laboratory, the extent and severity of plague in Vietnam were documented. Studies pertaining to plague control were actively pursued, and information was transmitted to military and civilian health officials as rapidly as it was generated. The disease did not become a major problem for U.S. military forces.

Control efforts in the Vietnamese sector were less successful. Plague continued to appear among the Vietnamese population in the city of Nha Trang during 1963, when Y. pestis was isolated from 85 of 186 patients tested (46 percent), and 1964, when it was isolated from 208 of 398 patients tested (52 percent) (Cavanaugh, Dangerfield, et al. 1968). Sporadic cases of plague occurred in and around Saigon during the same period. Succeeding years presented a picture of steady progression of the disease to new territories (chart 9 and maps 4 and 5).    

Fortunately, broad spectrum antibiotics were available, and a majority of treated patients survived the infection (Cavanaugh et al. 1970; Cavanaugh, Dangerfield, et al. 1968; Marshall et al. 1967). The invaluable observations made by U.S. civilian and military physicians aiding the Vietnamese in the plague emergency provided a sound basis for the management of this fulminating disease (Cutting et al., pp. 1096-98; Burkle 1973; Butler 1972).

The predominant clinical form of the disease in Vietnam in the 1960's was fleaborne bubonic plague (Cavanaugh, Dangerfield, et al. 1968; Marshall et al. 1967; Nguyen-Van-Ai 1963). Brief, sporadic outbreaks of pulmonary plague were observed (Legters et al.; Pham-Trong, Tran-Quy-Nhu, and Marshall 1967). Virulent plague bacilli also were isolated, on occasion, from the throats of asymptomatic contacts of plague patients, providing evidence of the occurrence of asymptomatic pharyngeal plague (Marshall, Quy, and Gibson 1967). However, the epidemiological significance of the asymptomatic pharyngeal patient was, and is as yet, obscure, as no evidence of transmission of Y. pestis by such individuals has been obtained (Marshall, Quy, and Gibson 1967; Legters, Cottingham, and Hunter 1970, p. 651).

A localized outbreak of fulminant pulmonary infection was observed near Rach Gia in which the etiologic agent appeared to be either Klebsiella pneumoniae or Y. pestis, or perhaps both (Legters et al.). In another instance, pneu monic plague was diagnosed in an American AID employee. The individual had traveled throughout Vietnam as a member of a team vaccinating Vietnamese civilians with an attenuated, live plague vaccine. He had received multiple inoculations of both killed and living plague vaccines before the onset of symp-


CHART 9.-Major plague outbreaks in the Republic of Vietnam, 1962-67

toms. His case history was not classical, in that the onset of serious symptomatology took several days, and he had an uneventful recovery following the administration of specific therapy (Cohen and Stockard 1967).

The broad spectrum antibiotics were freely available throughout South Vietnam, but numerous species of bacteria, in particular the enteric pathogens, had developed resistance to the majority of these drugs (Vivona et


MAP 4.- Extension of the human plague epidemic in the Republic of Vietnam, 1962. (Cavanaugh, D. C.; Dangerfield, J. G.; Hunter, D. H.; Joy, R. J. T.; Marshall, J. D., Jr.; Quy, D. V.; Vivona, S.; and Winter, P. E. 1968. Some observations on the current plague outbreak in the Republic of Vietnam. Am. J. Pub. Health 58: 742-52.)


MAP 5.- Extension of the human plague epidemic in the Republic of Vietnam, 1967. (Cavanaugh, D. C.; Dangerfield, J. G.; Hunter, D. H.; Joy, R. J. T.; Marshall, J. D., Jr.; Quy, D. V.; Vivona, S.; and Winter, P. E. 1968. Some observations on the current plague outbreak in the Republic of Vietnam. Am. J. Pub. Health 58: 742-52.)


al. 1966). Because of fear that heavy and sustained antibiotic pressure would result in the selection of antibiotic resistant Y. pestis, all plague strains isolated in Vietnam were screened to detect resistance as rapidly as possible. However, aside from a few strains, antibiotic resistant plague bacilli were not a problem (Marshall et al. 1967a). Moreover, strains which exhibited some resistance to streptomycin remained sensitive to other antibiotics (McCrumb, Larson, and Meyer 1953). Some problems in evaluating the clinical response to streptomycin therapy occurred; several patients remaining febrile following treatment were found to be suffering from concomitant typhoid fever which was then effectively treated with chloramphenicol.

Information was gathered on animal reservoirs of plague in Vietnam. Tissue pools from 22,144 small mammals and pools of their flea ectoparasites were examined for infection with Y. pestis (Cavanaugh, Ryan, and Marshall 1969). In addition, the serums of many rodents were tested for specific antibody to the Fraction I antigen. In coastal urban areas, four mammals were determined to be important in the epidemiology of plague: Rattus exulans, R. norvegicus, R. rattus, and Suncus murinus were infected at a frequency of about four Y. pestis isolations per 1,000 animals tested. A number of isolations and seropositives suggested that sylvatic rodents as well as urban rodents were involved (Cavanaugh et al. 1970; Cavanaugh, Dangerfield, et al. 1968; Marshall et al. 1967a, 1967b; Cavanaugh, Ryan, and Marshall 1969; Cavanaugh, Hunter, et al. 1968; Marshall, Currie, and Quy 1968; Van Peenen et al. 1970).

The Oriental rat flea, Xenopsylla cheopis, the classic vector of plague bacilli, was the principal flea ectoparasite. Over 99 percent of the fleas collected were X. cheopis (Cavanaugh, Dangerfield, et al. 1968, p. 748; Cavanaugh, Ryan, and Marshall 1969; Stark 1971). Other fleas collected included Xenopsylla astia, X. vexabilis, Ctenocephalides felis felis, C. canis, C. fells orientis, Stivalius klossi, S. aporus subspecies, Leptopsylla segnis, and Macrostylophora species (Stark 1971).

Bacteriologic and serologic studies performed on specimens collected from trapped rodents indicated that enzootic plague was present in the R. norvegicus population of Saigon for several years. Scattered and sporadic cases of plague occurred from 1962 to 1967, excepting the first half of 1965, which was characterized by a rather sharp epidemic of bubonic plague. DDT was used in control efforts in 1965, but by the time supplies of this insecticide were available, the epidemic (and rodent epizootic) was already in decline. Insecticide treatments did not prevent the reappearance of enzootic plague in Saigon, suggesting that DDT was not effectively controlling fleas (Hunter and Dangerfield 1967). Both bacteriologic and serologic evidence demonstrated that extensive enzootic plague could be present in an urban locale without overt human epidemics (Cavanaugh et al. 1970).

In contrast to Saigon, which experienced an epidemic only once in 6 years, Nha Trang experienced plague annually. In Saigon, isolations of Y. pestis were obtained more frequently from fleas than from rodent tissues during the epidemic year, while in endemic years, the reverse was true. The isolation rate for Y. pestis from rats collected in Nha Trang was approximately 10 times


greater than the rate in Saigon, and the isolation rate from fleas was approximately 20 times greater. While flea indices in Saigon were somewhat depressed during the endemic years, flea populations in Nha Trang were always high from January to May. Temperature conditions in Nha Trang were also more favorable for the flea transmission of Y. pestis from rat to man than they were in Saigon. As conditions for flea transmission were not always favorable in Saigon, studies were undertaken to determine how plague could persist in the rodent population. Cannibalism among rodents appeared to be a reasonable possibility, and was confirmed by subsequent experimentation (Rust et al. 1972).   

DDT was initially used throughout Vietnam for flea control without apparent success. Aliquots of DDT from available stocks were submitted for analysis to Geigy, a Swiss chemical company, and found to be entirely as specified.*

When tested by WHO methods, the fleas collected in Saigon and Nha Trang were shown to be resistant to DDT but sensitive to other insecticides (Cavanaugh, Dangerfield, et al. 1968). These observations prompted a program to collect fleas in other plague foci and test them for insecticide resistance. Collecting trips often were possible only during periods when the flea population was at a seasonal low, so colonization of the new fleas that could be collected was required to provide sufficient numbers for the tests. A method for establishing flea colonies in the field was developed (Cavanaugh et al. 1972), and tests indicated that sensitivity to DDT varied from one focus to another.

These tests resulted in the selection of the organophosphate insecticide diazinon for use in the plague control programs of the U.S. Army in Vietnam. This insecticide was used in a program in which results could be fully evaluated, and prompt control was achieved (Legters, Cottingham, and Hunter 1970, pp. 640-41). One difficulty in relying upon a nonresidual insecticide for flea control may have been elucidated by this study. A recrudescence of the infection occurred 3 weeks after the application of diazinon, indicating that the insecticide had a rather short half-life. New flea populations might have become infected by feeding on rodents which were bacteremic long after infection by fleabite or which had been infected by some other mechanism. Reinfestation of the area by the immigration of infected rodents was also a possibility. Reapplications of diazinon were required to control this second outbreak. Areas controlled with organochlorine insecticides, noted for their residual effects, had not experienced such problems.

As in epidemics of bubonic plague in India (Rogers 1928), distinct plague seasons were noted in Vietnam. However, the seasons varied from locale to locale and appeared to be related to climate (Cavanaugh and Marshall 1972).

The prevalence of plague in Vietnamese civilians and murine typhus in Americans showed a close relationship to the abundance of X. cheopis, the vector for both diseases in the coastal lowlands. Analysis of the climate in foci


1 Col. Dan C. Cavanaugh, MSC: Personal communication to Dr. M. Giaquinto, 1965.


CHART 10.- Relationship between the occurrence of plague and various climatic factors in the coastal lowlands of the Republic of Vietnam, 1962-66

located in the coastal lowlands, where the majority of cases were reported, provided the composite shown in chart 10. The sixth month was the month of peak plague prevalence. Epidemics began at the end of the rainy seasons and later appeared to regress as temperatures exceeded 27.9°

C. Periods of low plague prevalence were marked by high temperatures, high vapor pressure deficits, or high precipitation. Possibly many fleas were drowned when exposed to heavy rainfall in poorly drained locales. Definition of plague seasons in the various foci was useful in forecasting periods of high risk and in coordinating requirements for control programs.    Plotting the results of bacteriological examinations was useful. House-byhouse or street-by-street plots showed that, in established foci, the disease occurred at random throughout the entire endemic districts, as, for example, in Nha Trang (Cavanaugh, Dangerfield, et al. 1968). In cities experiencing initial plague outbreaks, plotted data clearly showed pathways of dissemination and probable mechanisms of distribution. Such a pattern was observed in Hue, where canal boats apparently carried infected rats and fleas from one locale to another (Marshall et al. 1967). Retrospective epidemiological studies confirmed the dangerous potential of individuals infected with pneumonic plague who travel from one locale to another (Pham-Trong, Tran-Quy-Nhu, and Marshall


1967). The use of bacteriological data in this manner identified areas requiring immediate attention and areas requiring protection against the introduction of disease.    

Inquiry into circumstances surrounding index cases of plague in newly established foci produced limited though convincing evidence that the transportation of infected rats and their ectoparasites contributed to the spread of plague in Vietnam. In the months just preceding the initial outbreak in Qui Nhon, the aftereffects of a devastating typhoon required emergency food shipments from Saigon, where plague was present. The rice received from that city, showing gross evidence of rat infestation, was stored in a warehouse where plague-infected rats were collected within a few days. Soon, cases of plague were noted in surrounding areas. In Ban Me Thuot, the index case was a merchant, just returned from Nha Trang with a load of goods. In Dak To, suspicion was directed toward a specific air shipment of rice for indigenous troops that was dispatched from Nha Trang. These observations and the fact that inspectors of the NCDC (National Communicable Disease Center) Foreign Quarantine Program began finding R.exulans in an increasing number of ships and aircraft returning to the United States from Vietnam (Pratt 1967) resulted in cargo fumigation for military materiel to prevent the introduction of the disease into other areas, including the United States.

Concern that new episodes of plague abroad arose as a result of the Vietnamese epidemics soon prompted intensive study. New outbreaks of plague occurred in Java in 1968 (Velimirovic 1972) and in Yemen in 1969. In the United States, the plague bacillus was isolated from tree squirrels in Denver, Colo., in 1968 (Hudson et al. 1968) and from rat fleas in Tacoma, Wash., in 1971 (Hudson et al. 1973). However, plague bacilli isolated in Yemen proved to be the classical variant for the area. Dr. B. W. Hudson and his associates (1973), using acrylamide gel electrophoresis, were able to demonstrate that bacilli isolated in the Javanese focus and in the United States were in many instances unique and separate from Vietnamese strains. Furthermore, the epizootic that occurred in Denver was in tree squirrels only, with the exception of one human plague case. Other species of carnivores and rodents in the same area remained uninvolved. In Tacoma, all evidence indicated that commensal rodents in only one small area were infected. Preventive medicine personnel on military installations adjacent to both Denver and Tacoma cooperated fully in these investigations, collecting and examining rodents on military properties for evidence of disease, with entirely negative results. Thus, it can be concluded that the measures taken to prevent dangerous infestation of military cargo returning from Vietnam were successful.

American personnel lived and fought in Vietnam under conditions which brought them into intimate contact with plague-infested rodents and fleas. On several occasions, the plague bacillus was isolated from rats and fleas collected in the cantonment area of Cam Ranh Bay, where numerous cases of murine typhus were encountered. While plague was diagnosed in fewer than a dozen Americans throughout Vietnam and documented only in eight (Cavanaugh et al. 1974), obviously exposure to plague occurred on numerous occasions. For exam-


ple, examination of acute and convalescent serums from a number of hospitalized murine typhus patients demonstrated rising titers to specific Y. pestis antigens in 7 of 58 patients. The infrequent occurrence of clinical plague in Americans in Vietnam inspired confidence in the killed plague vaccine used; there is, indeed, considerable evidence that killed plague vaccines are highly effective protection against bubonic plague (Bartelloni, Marshall, and Cavanaugh 1973; Meyer, Smith, et al. 1974; Marshall, Bartelloni, et al. 1974; Marshall, Cavanaugh, et al. 1974; Cavanaugh et al. 1974).

While the American authorities used a killed vaccine, the Vietnamese authorities employed the live, attenuated Y. pestis EV strain in their control programs. This vaccine, as developed and used by the French scientist, Girard (1963), was of extreme value in Madagascar, where it was credited with a great reduction in plague morbidity. Similar results were not obtained in Vietnam. Logistical considerations, including production and delivery on the scale necessary for use within a limited time frame, a requirement for careful refrigerated handling during shipment, and the coordination of vaccine campaigns, all made the mass utilization of the EV vaccine difficult. Facilities for the lyophilization of these vaccines probably would have obviated these difficulties; equipment could have been purchased and Vietnamese trained for the purpose. Unfortunately, numerous technical problems that could not be resolved on an emergency basis occurred, but a freeze-drying facility was completed just before the departure of the WRAIR team from Vietnam.

Subsequent laboratory studies revealed that several variants of the EV strain differed in their capacity to evoke a satisfactory immunogenic response in man and laboratory animals (Meyer 1971). Strains with the greatest immunogenic potential were also exceedingly reactogenic and, indeed, potentially dangerous (Meyer 1971; Meyer, Cavanaugh, et al. 1974).

Observations that were made in the course of the Vietnamese experience are still stimulating studies throughout the world. While the United States experienced little difficulty in suppressing the disease in American personnel, the picture in the Vietnamese population was far different. The many problems involved in plague control, which is very expensive, were aggravated by the war and civil unrest, by early failures of insecticides to limit the territorial spread of the disease, and by the need to develop policy on a day-to-day basis as facts concerning the complexity of the local situation emerged. The United States was indeed fortunate in having the disciplined, highly trained organization of Army Medical Department specialists capable of quick response in a wartime environment.


In future plague emergencies, knowledge of the resistance of local fleas to various insecticides will be valuable, as will information on where adequate stocks of pesticide may be procured. Information on effective rat abatement procedures is also indicated; rat control might be of particular importance in a wartime situation such as existed in Saigon. Research on methods of rat con-


trol, other than improvements in local socioeconomic conditions, should be encouraged. Studies on the efficacy of various vaccines in preventing the pneumonic form of the disease should receive support. While the results on specific therapy for the plague patient are gratifying, the situation for those who procrastinate in seeking treatment is often hopeless. Studies on the pathophysiology of disease should continue to seek solutions to the problems encountered in dealing with these patients. Finally, surveillance, to alert authorities to the development of potentially hazardous enzootic or epizootic plague, should be continued in areas where the disease has occurred on previous occasions.

Section II. Melioidosis

Colonel William L. Moore, Jr., MC, USA

Melioidosis is an infectious disease of man and animals, endemic in tropical zones, particularly throughout Southeast Asia. It is caused by the gramnegative organism, Pseudomonas pseudomallei, a natural saprophyte which can be isolated from market produce, soil, rivers, ponds, drainage ditches, and rice paddies throughout endemic zones but is found more commonly in areas of human habitation (Chambon 1955; Strauss, Jason, and Mariappan 1967; Strauss, Groves, et al. 1969; Joubert and Phung Van Dan 1958; Fournier and Chambon 1958, pp. 7-24).


In 1911, while looking for cases of human glanders in Rangoon, Burma, a British medical officer, Capt. A. Whitmore, and Assistant Surgeon C. S. Krishnaswami (1912) recovered a previously unrecognized organism from post mortem specimens obtained from a morphine addict who had died 10 days after the onset of an infectious illness. Despite the apparent clinical, pathological, and bacteriological similarities of his disease to glanders, the epidemiological history made that an unlikely diagnosis. Subsequent bacteriological studies led to the recognition of a gram-negative, motile organism which had the morphological characteristics of Pseudomonas mallei but differed in several important respects (Whitmore 1912). During the ensuing 10 months, there were 37 more fatal cases from which the organism, now taxonomically classified as Pseudomonas pseudomallei, was isolated. In one case, the diagnosis was suspected before death, and the organism was recovered by guinea pig inoculation of the patient's blood. In only 1 of the initial 38 cases was the patient observed throughout the entire course of his illness (Whitmore and Krishnaswami 1912). In 1914, Fletcher observed the occurrence of fatal glanderslike illness in laboratory animals at the Institute for Medical Research in Kuala Lumpur, Malaya (Stanton and Fletcher 1921). The laboratory epizootic continued and it was not until 1917 that Stanton and Fletcher (1925) recognized


that the causative organism was identical to that which had been described by Whitmore and Krishnaswami.    

Until 1925, the organism had been isolated and identified from animal and human sources only in Rangoon, Kuala Lumpur, and Singapore (Stanton and Fletcher 1925). That year Pons and Advier (1927) observed a case in a young pregnant Vietnamese woman near Saigon. Subsequently, human, animal, and environmental isolates have been reported from a number of tropical and subtropical areas, predominantly from Southeast Asia and the Western Pacific, but also in the Western Hemisphere. Krishnaswami (1917) claimed to have seen more than 200 cases between 1910 and 1915, but only 83 case reports could be found for a literature review in 1932 (Stanton and Fletcher).

Melioidosis has been, and remains, a disease of significance to military medicine. First, it was discovered by military physicians. Second, the occurrence of approximately 100 cases among the French Expeditionary Forces in the Indochina War between 1948 and 1954 led to the conclusion that it was a disease of some importance to military operations in an endemic area (Rubin, Alexander, and Yager 1963). Third, there is some justification for the view that the movement of personnel and supplies during and after World War II was responsible for dissemination of the organism from its original endemic home or, at the very least, that military interest in the disease resulted in its increased recognition throughout the world (Fournier and Chambon 1958). Finally, melioidosis became increasingly important to U.S. military forces operating in the endemic area over the past three decades.

Cox and Arbogast (1945) reported the first case in an American soldier in Burma. Sporadic cases occurred thereafter in military personnel, but the disease remained of interest primarily to epidemiologists and laboratory investigators. Rubin et al. prophetically labeled it a "military medical problem" in 1963, and the deployment of large numbers of troops to South Vietnam beginning in 1965 led to the fulfillment of that prophecy. The Melioidosis Registry of the U.S. Army Office of the Surgeon General (OTSG-MR) contains reports of 343 cases, among which there were 15 deaths directly caused by the disease and 21 others in which it was a contributing factor or secondary diagnosis. Anecdotal information from physicians assigned to Vietnam during 1968 and 1969 leads one to conclude that a significant number of diagnosed cases were not reported and that there must have been a number of undiagnosed cases as well.

Of what significance is a disease occurring in so few of the 2.5 million persons at risk (Clayton, Lisella, and Martin 1973) and resulting in so few deaths? First, despite the clairvoyance of Rubin et al., the disease was not anticipated by medical personnel on duty in Vietnam, and the occurrence of a number of fulminant, rapidly fatal cases in 1966 (Weber et al. 1969) was demoralizing to those who were already overwhelmed by diseases with which U.S. physicians generally lacked familiarity. Second, 1 in 10 of these cases was initially diagnosed as pulmonary tuberculosis, fostering unecessary epidemiological concern and adding to the frustration of the medical officers involved. Third, at least 28 of these cases occurred in the continental United States or Hawaii following duty in Vietnam (OTSG-MR), causing some concern and leading to the disease's being called


a "medical time bomb." Two of these patients became ill while serving tours of duty in Germany, recalling the experience reported by European physicians after repatriation of French forces in 1954 (Jackson, Moore, and Sanford 1972). On the basis of serologic studies described below, it can be estimated that between 25,000 and 225,000 U.S. soldiers had subclinical infection with P. pseudomallei. There is adequate documentation that the disease can recur months or years after apparent cure or can first occur after many years of latency. Intercurrent illness, injury, or stress appear to be important factors in the latter circumstance (Jackson, Moore, and Sanford 1972, p. 271; Clayton, Lisella, and Martin 1973, p. 24).


Despite extensive study of the disease, the true incidence of melioidosis has not been established. Prevatt and Hunt (1957) were able to find only 300 cases in their review of the world's literature. If one includes all reports, substantiated or not, the number of recognized cases after more than 60 years does not exceed 1,000.

Despite the paucity of recognized cases, serological surveys indicate that mild or inapparent infection is not uncommon, suggesting that contact with the organism occurs frequently. Significant titers have been found in 6 to 20 percent of indigenous personnel in Vietnam, Thailand, and Malaysia (Brygoo 1953; Nigg 1963; Strauss, Alexander, et al. 1969). Similarly, 1.2 percent of Europeans living in Vietnam (Brygoo 1953) and from 1.1 percent (Spotnitz, Rudnitzky, and Rambaud 1967) to 2 percent'` of healthy or nonwounded U.S. Army troops returning to the United States after spending 6 to 12 months in Vietnam had significant HA (hemagglutination) titers.

In a prospective serological study, Legters and coworkers** collected paired serums from 553 Special Forces personnel assigned to temporary duty in Vietnam for 6-month periods from 1961 to 1963. Of 97 individuals who had ex perienced one or more febrile illnesses, 2 demonstrated a fourfold or greater rise in HA/titer for melioidosis, representing infection in 0.36 percent of the population at risk. Sporadic cases occurred between 1960 and 1965, and following the troop buildup beginning in April 1965, the number increased sharply (table 27).

By August 1967, the 9th Medical Laboratory had begun to do HA tests for melioidosis on virtually all serum specimens submitted from a variety of sources. By January 1968, 66 cases had been identified by high or rising HA titers (ML9-AR). These represented approximately 9 percent of serum specimens submitted for an FUO (fever of undetermined origin) evaluation. Additional cases were identified from specimens submitted for cold agglutination titers from patients with suspected atypical pneumonia or tuberculosis. Unfortunately, demographic data on these patients are incomplete, and followup information is


*Sanders, C. V.; Moore, W. L.; and Sanford, J. P. Unpublished observation, 1969.

**  Legters, L. J.; Buescher, E.; and Coppedge, R. L. Unpublished observations, personal communications, 1963.


TABLE 27.-Meliodosis in Vietnam, 1965-71

available on very few of them. Of the 66 cases, 25 (38 percent) came from the 25th Infantry Division (Johnson 1968). Nine cases of melioidosis in the 25th Infantry Division in 1966 were reported in detail (Weber et al. 1969). In 13 of the cases from this division reported through September 1968, the geographic location had been established (Johnson 1968). In one instance, the patient had been in Tay Ninh Base Camp for his entire tour. Significant HA titers (1:40 or higher) were found in 5.2 percent of 38 men from his unit. Most of the cases were from the Cu Chi area, and P. pseudomallei was cultured from several soil and water samples taken there (ML9-AR; Johnson 1968).

Positive serologies were noted in 0 to 9 percent of patients in various FUO studies (Deller and Russell 1967; Reiley and Russell 1969; Deaton 1969; Colwell et al. 1969; Kishimoto et al. 1971; ML9-AR, p.43). In early 1969, a survey in the III CTZ (Corps Tactical Zone) disclosed 5 to 8 percent positive titers in tactical units (USARV-MB, p.17). In later studies of troops returned to the United States, the percentage of significant titers was found to be related to the presence, and perhaps extent, of trauma in the patients. Seven (3.5 percent) of 200 patients with wounds of all types had HA titers of 1:40 or greater (Kishimoto et al. 1971) compared with 18 percent of patients with open orthopedic wounds' and 32 percent of burn patients. **

To determine if incidence was related to season, location, duty assignment, and length of time in Vietnam, a continuing FUO serologic study was conducted from the latter half of 1969 through early 1970 by the 9th Medical Laboratory in conjunction with a number of medical facilities throughout the country. Of 366 serological confirmations, 19 (5 percent) demonstrated probable or confirmed melioidosis (USARV-CHR). There were two positives in September and October, eight in November, one in December, two in January, none in February,


*Maj. Creed D. Smith, Chief, Microbiology, 6th U.S. Army Area Laboratory, Fort Baker, Calif.; Personal communication.
**  Maj. Larry N. Dotin, MC, U.S. Army Institute of Surgical Research, Fort Sam Houston, Tex.: Personal communication.


three in March, and one in April. The seasonal occurrence of cases from retrospective review of the Registry reports is shown in table 27. Unfortunately, data on a large number of other variables which might have influenced these findings are not available for review.

Cowley found that 33 percent of cases had occurred in helicopter crewmen (Howe, Sampath, and Spotnitz 1971). In a subsequent study involving U.S. marines, no correlation could be established between elevated HA titers and any of the variables involved (Clayton, Lisella, and Martin 1973). While approximately 10 percent of the earlier cases had occurred in burn patients and 32 percent of such patients had significant HA titers,* by 1970 no cases were seen in the Institute of Surgical Research Burn Center at Fort Sam Houston and by 1971 there were no longer any positive HA titers found.** In all, 44 (12.8 percent) of the cases reported in the Registry occurred in burn patients (OTSG-MR).

Immunologically demonstrated exposure to the organism is far more common than clinical disease. Furthermore, there is an increased prevalence of disease and of asymptomatic seroconversion in patients with penetrating injury and burns, particularly when these wounds are contaminated with dirt and stagnant water from the environment.   

The epidemiology of melioidosis has been investigated extensively. The organism is widely distributed around the world between 20 0 N. and 20 0 S. latitudes. Human, animal, and environmental isolates of P. pseudomallei have been reported from Burma, Malaysia, Singapore, Vietnam, Thailand, Cambodia, Laos, India, Celebes, Java, Borneo, Indonesia, Ceylon, New Guinea, Papua, Australia, the Philippines, Guam, Turkey, Panama, Ecuador, Africa, Madagascar, Germany, France, Italy, Aruba, England, and the United States (Redfearn, Palleroni, and Stanier 1966; Rubin, Alexander, and Yager 1963). However, the patients whose cases occurred in the United States and Europe, except the one in Turkey (Ertug 1961), acquired the infection while residing or traveling in areas of known endemicity.

Of the 38 cases reported by Whitmore and Krishnaswami (1912), 28 occurred in chronic morphine addicts, leading to the appellation, "morphine injectors septicemia" (Krishnaswami 1917), although evidence for a cause-and-effect association was circumstantial. Stanton and Fletcher (1925) believed that the laboratory epizootic in Kuala Lumpur resulted from fecal contamination of animal feed by infected rats, but subsequent studies have shown that rats rarely harbor the organism. Chambon (1955) and a number of others (Strauss, Jason, and Mariappan 1967; Joubert and Phung Van Dan 1958) have shown conclusively that the organism is a saprophyte in soil and water. There are numerous clinical examples of man acquiring the organism through contamination of wounds by dirt, mud, and water; this mode of transmission has also been demonstrated in the laboratory (Le Moine, Hasle, and Nguyen-Duc-Khoi 1937). Pulmonary infection has been seen following aspiration of contaminated water (Johnson 1968, p. 3) and experimental airborne infection has been demonstrated in hamsters


* Maj. Larry N. Dotin, MC: Personal communication.

** Robert L. Lindberg, Chief, Microbiology, U.S. Army Institute of Surgical Research, Fort Sam Houston, Tex.: Personal communication.


(Rosebury 1947, pp. 151-58). Mosquitoes and fleas have also been shown to spread the organism under experimental conditions (Blanc and Baltazard 1941a, b).

In the absence of trauma or known instances of ingestion, inhalation, or aspiration, there is little to suggest the specific means by which many patients become infected. In 209 of 343 case reports in the Melioidosis Registry (OTSG MR), there is no evidence of trauma or other clear-cut exposure. No age, sex, or racial group is exempt from melioidosis. There is scanty evidence that chronic debility or preexisting disease (that is, diabetes mellitus) or intercurrent surgery predisposes one to the development of clinical melioidosis. Virtually all of the nontrauma-associated cases in U.S. Army personnel occurred in individuals who were otherwise healthy.

The distribution of cases in Vietnam may offer some epidemiological clue but, unfortunately, the available data do not permit firm conclusions. First, the disease was virtually unknown in the 9th and 4th Infantry Divisions, while there was an abundance of cases in the 25th Infantry Division, the 11th Armored Cavalry, and the 1st Cavalry Division (Airmobile). Second, at least 190 cases were seen in only nine medical facilities, while 99 others either had no medical unit designated or occurred in small numbers in numerous other field hospitals, evacuation hospitals, or surgical hospitals in Vietnam. The distribution of melioidosis among Army personnel, by facility, from 1965 to 1971 was as follows (OTSG-MR):

Evacuation HospitalsField Hospitals

12th - 533d - 21
24th - 248th - 5
67th - 14Other military, in Vietnam - 99
85th - 12Military, in continental United States or Hawaii - 28
91st - 10Military in Europe - 2
93d - 38Civilian or Veterans Administration in United States - 7
95th - 13Other facilities - 17

Man-to-man transmission has been conclusively demonstrated only once; McCormick et al. (1975) recently reported a case of venereal transmission by a Vietnam returnee with prostatic infection. Prostatic melioidosis infection had been reported previously (Sollier and Boutareau 1937). Some epidemiological confusion has resulted from an account of a neonatal case in which the only apparent source of infection was the father, recently returned from Vietnam, who, however, had no opportunity for contact with the infant between onset of illness and the child's death (Osteraas et al. 1971).

The incubation period for melioidosis can be as short as 1 day (Johnson 1968) or as long as 26 years (Mays and Ricketts 1975). Clearly, although much is known, a number of interesting questions remain about the epidemiology of this often confusing disease.


Melioidosis is caused by the slender, motile, gram-negative bacillus


FIGURE 51.-Typical culture appearance of Pseudomonas pseudomallei on blood agar plate, 48 hours. Note typical rugose, umbonate morphology. (Courtesy, Col. John J. Deller, Jr., MC.)

Pseudomonas pseudomallei. It has previously been classified in several genera, including Malleomyces, Loefflerella, Pfeifferella, and Actinobacillus. It stains somewhat poorly with Gram's stain, showing irregular mottling and bipolar densities which are more prominent with methylene blue, Giemsa's stain, or Leishman's stain. In early culture, colonies may be indistinguishable from some strains of Pseudomonas aeruginosa and must be kept for 48 to 96 hours before developing the characteristic opaque, white, umbonate, rugose morphology (fig. 51). While the organism grows well at 25°C, 37°C, or 42°C, incubation at room temperature favors the development of the characteristic colonies. The organism causes alpha hemolysis in blood agar, a heavy, scummy surface pellicle formation in broth culture, and a musty, wet-earth odor in mature culture. On MacConkey's agar, the colonies are small, pink to red, and have a faintly visible metallic sheen. On EMB (eosin-methylene blue) agar, the colonies are lavender with a metallic sheen, and some investigators have used this feature as a means of early identification. The biochemical and serological confirmation of the identity of the organism has been described in detail (Moore, Hedberg, and Lindberg 1970). 

Knowledge of the pathogenesis of melioidosis is derived from clinical and laboratory studies (Le Moine, Hasle, and Nguyen-Due-Khoi 1937; Rosebury 1947). The incubation period is apparently related to a number of factors. Exerimentally, highly susceptible animals - guinea pigs, hamsters, or rabbits-injected subcutaneously or intraperitoneally with large doses of virulent organisms may succumb within 24 hours. Animals exposed via scarification of skin, inhalation, or ingestion usually manifest symptoms within 2 to 7 days. A monkey which was infected by aerosol exposure to 130 organisms showed signs of illness only after 2 months and, when killed after 68 days, clearly had


melioidosis (Nigg and Johnston 1961). 

Human disease has occurred within 5 to 10 days following burns (Flemma et al. 1969), within 1 day following aspiration of contaminated water (Johnson 1968), and within 3 days following aerosol exposure during a laboratory accident (Green and Tuffnell 1968). Prolonged latency after exposure to an endemic environment for from 5 months (Flemma et al. 1969) to as long as 26 years has been reported (Mays and Ricketts 1975). Latent infection has become manifest spontaneously (Spotnitz, Rudnitzky, and Rambaud 1967; Beamer et al. 1954; McDowell and Varney 1947) or in association with intercurrent medical or surgical illness (Jackson, Moore, and Sanford 1972; Alain, Saint-Etienne, and Reynes 1949). Apparently, the incubation period is variable, depending on the magnitude and mode of exposure and a variety of environmental and hostsusceptibility factors (Dannenberg and Scott 1958a, b). It seems safe to assume in most cases of clinical illness the incubation period in health individuals is from 2 days to 3 weeks.

The disease mimics plague and tularemia in two major respects: septicemic dissemination frequently occurs in the course of primary pneumonia, and pneumonia, usually fatal, may develop after hematogenous spread from a localized infection or in the course of a primary septicemia. One can only speculate that the pathogenetic events in acute, systemic disease, in the absence of a primary infection, involve intracellular or extracellular hematogenous dissemination of the organism acquired from the environment through lungs, intestine, or skin. In recrudescent or late-occurring disease, the organism likely has lain dormant in intracellular sites, presumably in the reticuloendothelial system, but this has not been demonstrated clinically or experimentally. The factors which lead to active local or systemic infection in these cases are related to the complex interrelationships of humoral and cellular immune mechanisms and other poorly defined homeostatic systems. 

In the acute and subacute forms of the disease, widely disseminated abscesses from 1 mm to 3 cm in diameter are seen. They are buff or yellow-gray and may be firm or rubbery or show central caseous necrosis with or without early cavitation. Larger lesions are hemorrhagic at the periphery and contain a mixed cellular infiltrate including polymorphonuclear leukocytes, mononuclear cells, and atypical multinucleated giant cells. There is a marked degree of karyorrhexis of involved cells, suggesting that a severe form of cytotoxicity results from a toxin or toxins produced by the organism. The lesions in chronic disease are larger and often cavitary, containing mononuclear cells, plasma cells, and granulation tissue.


The clinical manifestations of melioidosis are sufficiently variable to preclude accurate diagnosis on the basis of symptoms, physical findings, and laboratory information short of bacteriological or serological confirmation. It was initially recognized as a severe, almost invariably fatal septicemic disease with involvement of the lungs and, often, widespread visceral dissemination


(Whitmore and Krishnaswami 1912; Whitmore 1912; Knapp 1915). Subsequent clinical and experimental investigations have disclosed its protean clinical spectrum of acute, subacute, chronic, and subclinical or asymptomatic infection. 

Characteristically, the onset is dramatically abrupt, and the predominant symptoms are those of acute pulmonary infection. In the 38 cases originally described by Whitmore (1912; Whitmore and Krishnaswami 1912), most patients clearly presented the dominant picture of acute pulmonary infection, and most of the remainder had obvious lung involvement. Numerous other reports reflect the prevalence of pulmonary presentation. Most patients present with fever, chills, malaise, and a cough which usually produces purulent or bloody sputum, sometimes after the first 1 to 2 days of illness. The temperature usually varies between 99 0 F and 104 0 F, occasionally higher. Dyspnea, often moderate, may be out of proportion to the paucity of physical and radiographic findings in early disease. Anorexia, vomiting, and diarrhea are encountered frequently (Weber et al. 1969). Physical findings include fever, moderate tachypnea and tachycardia (consistent with fever), conjunctival suffusion, pharyngeal enanthema, rales, rhonchi, occasionally signs of consolidation, and, infrequently, a pleural friction rub. Pleural effusion and empyema are distinctly rare. The spleen and liver are not palpable in the absence of dissemination. Laboratory findings include normal to elevated leukocyte counts, rarely exceeding 20,000Imm3; polymorphonuclear cells comprise 85 to 90 percent of the total count. Mild, normochromic, normocytic anemia has been seen as early as the first week of illness. The organism has been seen in sputum smears in approximately 80 percent of the reported instances in which this test has been done. Radiographically, the most characteristic pattern is one of irregular nodular densities 4 to 10 mm in diameter scattered throughout the lung (James, Dixon, and Johnson 1967). Unilateral single or multiple lobe irregular confluence proceeding to consolidation is a less frequent finding. Cavitation occurs in some patients surviving beyond 1 week of illness.

The acute septicemic form of the disease is exemplified by cases 1 to 4 of Weber et al. (1969) and the five cases of Flemma et al. (1969). The first two cases began with cutaneous lesions which progressed rapidly and relentlessly through phases of cellulitis, lymphangitis, distal metastatic abscesses in skin and viscera, and diffuse involvement of the lungs. The third patient progressed from early symptoms of gastroenteritis to rapidly fatal diffuse pneumonia. The fourth had a very nondescript initial febrile illness which terminated in a fulminant pneumonia. All had high fevers, up to 106 0 F, and demonstrated evidence of toxicity with malaise and depressed mentation. Two of the four had normal total peripheral white blood cell counts and two had substantial leukocytosis. In none was there an associated history or evidence of trauma, while the five patients of Flemma et al. had suffered significant burns. The distressing features of these cases which are commonly noted in acute severe disease are lack of specific early signs and symptoms, rapidity and extent of progression, failure to respond to appropriate antibiotic administration, and a high case fatality rate.

Features of the acute pulmonary form of the disease are seen in case 5 of Weber et al. Characteristic are the sudden onset, with fever, chest pain (often pleuritic), productive cough, and a positive chest roentgenogram which, in this


FIGURE 52.-Histopathologic material from a lung showing inflammatory cell infiltration in the area of an abscess. (Courtesy, Col. John J. Deller, Jr. MC.)

case as in many others, appears similar to a lung abscess (fig. 52) or cavitary tuberculosis. The white blood count may be normal or elevated. Response to therapy is more predictable and the case fatality rate, although estimated to be about 10 percent, is considerably lower than that of the acute systemic form of the disease.

The subacute or chronic pulmonary form of the disease is represented by the cases of Sweet, Wilson, and Chandler (1968). The first patient, a previously healthy, 23-year-old soldier, had a 2-month illness characterized by a dry non productive cough, left pleuritic chest pain, intermittent fever, anorexia, malaise, weight loss, lymphadenopathy, and abnormal lymphocytes in a peripheral blood smear. His initial chest X-ray demonstrated a large, thin-walled cavity in the left upper lobe, and he developed a fever of 105 0 F accompanied by a modest leukocytosis without developing other evidence of acute disease (figs. 53 and 54). He was treated for tuberculosis and appeared to show some improvement but responded briskly after the diagnosis of pulmonary melioidosis was established and therapy with chloramphenicol, novobiocin, and streptomycin was begun. Case 2 had a remarkably similar course, as did two patients seen by the author who were diagnosed and treated at the pulmonary and infectious disease services at Brooke Army Medical Center. Spotnitz, Rudnitzky, and Rambaud (1967) reported two patients with pneumonitis who were clinically well and had a cavitary lung lesion discovered at the time of a routine chest X-ray.

Chronic melioidosis has been described by McDowell and Varney (1947),


FIGURE 53. - Typical X-ray of cavitary melioidosis. Note the resemblance to tuberculosis. (Courtesy, Col. John J. Deller, Jr., MC.)

FIGURE 54. - Extensive pulmonary melioidosis of left upper lobe with widespread infiltration and multiple small cavities.  Note similarity to pulmonarytuberculosis. (Courtesy, Col. John J. Deller, Jr., MC.)


Salisbury and Likos (1970), and Mays and Ricketts (1975). Chronic forms involve lungs (Everett and Nelson 1975), bone (Sutherland and Dahlstrom 1968; Borchardt, Stansifer, and Albano 1966), skin, and a variety of soft tissues (McDowell and Varney 1947).    

Less common forms include localized acute lymphadenitis (Peck and Zwanenburg 1947), myocardial abscess (Baumann and Morita 1967) (fig. 55), meningoencephalitis (Schaeffer and Grant 1968) (fig. 56), and soft tissue or muscle abscess (fig. 57).


Confirmation of the diagnosis of melioidosis depends upon recovery of the organism from clinical specimens or a fourfold or greater rise in HA or CF (complement fixation) antibody titer associated with an episode of clinically ap parent disease. Review of available data from the 343 Registry cases indicates that the diagnosis was based on serologic evidence in 24 cases, and that it was presumptive (dependent on a single high HA or CF titer) in 38 others who had either FUO (10 cases) or a clinical illness compatible with one of the forms of melioidosis. Cases not reported in the Registry include probable or presumptive diagnosis based on serologic tests done by the 9th Medical Laboratory and reported in a variety of sources. 

Recovery of the organism from sputum samples in suspected cases is enhanced by digestion with 1 percent pancreatin for a half hour at 37'C, the pH adjusted to 7.5 with phosphate buffered saline, and addition of crystalline penicillin G and polymyxin B at a final concentration of 400 Ag of each antibiotic per ml of specimen. This is then plated on NAGCV (nutrient agar [Difco]; glycerol 3 percent; crystal violet [1:200,000]; pH 6.7 to 7,0) medium and subsequently inoculated into hamsters. Blood cultures are collected in the usual manner, in standard broth media, and subcultured to NAGCV or standard laboratory solid media.* Pus, body fluids, and tissues may be handled as is sputum or inoculated directly onto solid media (blood agar plate, EMB, NAGCV, or MacConkey's) or in broth. The organism has been obtained from sputum, pus, blood, urine, feces, body fluids, tissue, and cerebrospinal fluid.

Agglutination tests with patient serums were used in serologic diagnosis in the past but are no longer considered helpful. Precipitin reactions have also been demonstrated but give false positive results with normal serums (Cravitz and Miller 1950). 

The complement fixation test is most specific although less sensitive than others and, when the titer is elevated, suggests recent disease, since significant titers (1:8 or greater) do not persist for more than 2 years (Cook 1962). IHA (in direct hemagglutination) is convenient, relatively easy to perform, specific at titers of 1:40 or greater, and is the serologic test most often used. The titer is usually elevated in acute disease. An elevated titer in a single serum specimen in relation to a suspected case has less diagnostic significance than demonstration


1 Department of Bacteriology and Immunology, SEATO Medical Research Laboratory, Bangkok, Thailand: Laboratory procedures, 1968.


FIGURE 55.-Inflammatory cell infiltration causing microabscess formation in the myocardium. (Courtesy, Col. John J. Deller, Jr., MC.)

FIGURE 56.-Typical microabscesses in the brain. (Courtesy, Col. John J. Deller, Jr., MC.)


FIGURE 57.-Inflammatory cell infiltrate in the pectoralis muscle causing abscess formation (Courtesy, Col. John J. Deller, Jr., MC.)

of complement fixing titer because indirect hemagglutinin may persist for many years after an immunogenetically significant encounter with the organism, whether or not clinical disease has occurred.   

In view of the probable occurrence of false positive HA tests, Nigg (1963) recommended the combined use of complement fixation and hemagglutination tests in diagnostic and epidemiological studies. Paired serum samples are most helpful in establishing the diagnosis when cultures of appropriate clinical specimens are negative.


There is no effective means of active or passive immunization against melioidosis. Data are not available to support the notion that early appropriate antibiotic treatment of wounded or burned individuals potentially exposed to P. pseudomallei would reduce the incidence of clinical disease; unfortunately, there is no demonstrable correlation between cases of melioidosis and the widespread use of penicillin, tetracycline, kanamycin, gentamicin, and chloramphenicol in various combinations among such patients in Vietnam. Since the organism is more common in areas of human habitation throughout the endemic zone, the only effective means of prevention appears to be avoiding exposure to the environment and that is, obviously, often impossible.   

For a variety of reasons, the therapy of melioidosis remains a matter of mild


controversy. In vitro studies have shown that the organism is usually susceptible to clinically obtainable concentrations of sulfonamides, kanamycin, tetracycline, chloramphenicol, and novobiocin (Eickhoff et al. 1970), and synergism between novobiocin and tetracycline has been shown (Calabi 1973). Because of the high case fatality ratio associated with the acute septicemic and overwhelming pulmonary form of the disease, some authors have recommended combined therapy with chloramphenicol, 12 g per day; kanamycin, 4 g per day; and novobiocin, 6 g per day (Cooper 1967). The effectiveness of this regimen has not been demonstrated, and there have been several instances of severe bone marrow depression, ototoxicity, nephrotoxicity, and hepatotoxicity associated with this combination of drugs. In mild to moderate disease, tetracycline alone at 3 g per day or in combination with sulfisoxazole in standard or modestly increased doses is effective without significantly increasing the likelihood of toxic reactions (Cooper 1968; Spotnitz 1968) (figs. 58-60).   

Studies in Vietnam in 1969 (chart 11) demonstrated significant sensitivity to tetracycline at low concentration, further emphasizing the importance of this drug as the mainstay of the therapeutic effort. Patients who succumb to acute disease have had positive postmortem cultures despite several days of therapy with antibiotics to which the organism appears to be susceptible in vitro. The response to antibiotic therapy is not so rapid as in many other bacterial infections, and relapse may occur after an asymptomatic interval of weeks to many months (Jackson, Moore, and Sanford 1972). The occasional occurrence of relapse has led some physicians to consider treatment for 3 to 6 months even for uncomplicated disease.


The experience with melioidosis resulting from military operations in Vietnam has not resulted in any major advances but has brought attention to observations made earlier.

First, early in its course, the disease is frequently mistaken for pulmonary tuberculosis, and when melioidosis is not considered in the initial differential diagnosis there may be considerable unnecessary delay in proper evaluation. Early diagnosis frequently depends upon a high index of suspicion and requires that, in addition to appropriate serological tests, the laboratory be advised to observe the cultures for a minimum of 4 days. Otherwise, the organism may be discarded before the characteristic morphological appearance develops.

Second, although in vitro antimicrobial susceptibility of the organism has shown no major changes since the discovery of chloramphenicol, in vivo response to drugs is variable and at times unpredictable. Therapy with appropriate an tibiotics for less than 3 weeks has been accompanied by an unacceptable relapse rate; in the most extreme example, active disease has recurred after more than a year of continuous drug therapy (Jackson, Moore, and Sanford 1972). In fulminant cases, positive cultures have been obtained from postmortem specimens even after several days of therapy to which the organism shows in vitro susceptibility. On the other hand, high-dose multiple-drug therapy for fulminant


FIGURE 58. - Initial X-ray in a case of pulmonary melioidosis, untreated on 25 August 1969.(Courtesy, Col. John J. Deller, Jr., MC.)

FIGURE 59. - X-RAY showing sequential improvement in a case of pulmonary melioidosis, partially resolved on tetracycline therapy on 11 September 1969. (Courtesy, Col. J. Deller, Jr., MC.)


FIGURE 60.-Final X-ray showing improvement in a case of pulmonary melioidosis, near complete resolution following a course of tetracycline therapy, 25 September 1969. (Courtesy, Col. John J. Deller, Jr., MC.)

disease-specifically chloramphenicol, kanamycin, and novobiocin in combination-has not been shown to have increased effectiveness and the potential toxicity of this combination precludes its use.   

Third, the experience in Vietnam points to the need for a central repository or registry for the maintenance of records and data on diseases of military or potential military importance, particularly for diseases that are not widely known or understood. It further emphasizes the importance of area medical intelligence and global epidemiology to military activities.


CHART 11.-Drug sensitivity of sixty-one Pseudomonas pseudomallei isolates studied in Vietnam, 1969

Section III. Tuberculosis

Carl R. Guiton, M.D., and Colonel O'Neill Barrett, Jr., MC, USA (Ret.)


Tuberculosis has been a problem since antiquity and for centuries was a principal cause of death in men of military age. Records of hospital admissions and medical discharges from military service for tuberculosis have been maintained by the U.S. Army since the Civil War. During that conflict, there were 13,499 tuberculosis admissions and 5,286 deaths from the disease among white soldiers. The mean annual rate of discharge for tuberculosis was 8.6 per 1,000 in white troops and 3.1 per 1,000 in black troops. However, in neither the Civil War nor the Spanish-American War was the disease frequent enough to prompt any unusual comment in the analyses recording the medical aspects of military operations.


Unless otherwise noted, material under "History and Military Significance" in this section is derived from Medical Department, United States Army. Infectious diseases. Internal Medicine in World War II, vol. 2. Washington: Government Printing Office, 1963, pp.329-407.


During World War I, persons with tuberculosis were detected and excluded from military service almost entirely on the basis of the physical examination since roentgenology was in its infancy and screening skin testing resources were not available. There were 22,812 disability separations because of tuberculosis during the war, or 5.52 per 1,000 average strength per annum. The disease was the leading cause of disability separation, accounting for 11.1 percent of the total (MD-WW15, pp. 166-82,1022). Furthermore, the full magnitude of the problem did not become evident until several years after the war. Goldberg (1941) calculated that the approximate expenditure by the Veterans Administration for service-connected tuberculosis from the close of World War I through 1940 was $1,186,000,000. The number of hospitalized tuberculosis beneficiaries peaked in 1922, at 44,591 (Wolford 1944).

At the beginning of World War II, the Office of the Surgeon General recognized that drastic revision of the physical standards in the existing Mobilization Regulations was necessary because of technical developments in tuberculosis control. In April 1939, a chest X-ray was required before applicants could be commissioned. As early as 1940, routine screening chest X-rays for all inductees were considered, but they were not made a mandatory part of all physical examinations at induction stations until 3 June 1941. While approximately 10 million men had chest X-ray examinations, about 1 million men were inducted without them (MD-PS, pp. 30-33).

The average incidence rate of tuberculosis for World War II from 7 December 1941 to 14 August 1945 was 1.2 per 1,000 per annun (MD-IM2, p. 334). Tuberculosis accounted for 1.9 percent of all discharges for disability from disease between 1942 and 1945, ranking 13th on the list (HOA-46, pp. 22-23). Among Americans who had been prisoners of war, the rates were higher. Prisoners from the European theater had an incidence five to seven times that of the Army in general (MD-IM2, p. 346). Figures on those returned from the Pacific area were more difficult to obtain, but a special study of repatriated prisoners at West Coast debarkation hospitals, directed by The Surgeon General, showed that 2.7 percent of 3,742 individuals studied with chest X-rays had evidence of active pulmonary tuberculosis (Morgan, Wright, and Van Ravenswaay 1946).

As a means of standardizing therapy and of studying the disease in more detail, three specialty centers for the management of tuberculosis were established: Fitzsimons General Hospital in Denver, Colo., Bruns General Hospital in Santa Fe, N. Mex., and Moore General Hospital in Swannanoa, N.C. Fitzsimons Hospital has remained the Army center for tuberculosis.

While specific incidence rates are not available for the Korean conflict, tuberculosis continued to be a problem for the Army, as reflected by the approximately 600 admissions per year to Fitzsimons General Hospital during that period (HOA-52, p. 11; HOA-53, p. 11; HOA-54, p. 25).


Rightful concern was expressed about the exposure of American troops in


Vietnam to a population with a high tuberculosis infection rate. The conflict in Vietnam placed an estimated 500,000 American military personnel annually in varying degrees of contact with a highly infected population. While detailed statistics were initially lacking, several U.S. studies done there highlight the seriousness of the problem. In 1968, a chest X-ray survey by Siegler et al. of Vietnamese civilians showed that 31.7 percent over the age of 15 had definite radiologic evidence of active pulmonary tuberculosis (Cowley 1970). Another study demonstrated that nearly 100 percent of the adult population was tuberculin skin test positive (Houk 1967). Poffenbarger (1972) performed tuberculosis skin tests on 631 children, ages 1 through 18, and showed an increasing number of positive reactions with age. Of the 17- to 18-year-old group, 47.5 percent reacted. Furthermore, tuberculin-negative children retested 9 months later showed increasing conversion rates, with a 17-percent chance annually of acquiring infection.

Among American troops, approximately 95 percent had had no previous exposure to tuberculosis and were tuberculin-negative upon arrival in Vietnam (Stead and Bates 1969; Edwards 1969). Data from the 20th Preventive Medicine Unit indicated that only 6.2 percent of 901 first-time personnel were tuberculinpositive on entering the country, whereas 13.7 percent of 190 personnel with a previous tour in Vietnam were positive (PMU-20). In the first-tour group, breakdown by race showed that 3.2 percent of whites, 7.4 percent of blacks, 9.1 percent of Orientals, and 15.6 percent of Spanish-surnamed persons were positive; a similar racial distribution was noted in the group with previous tours in Vietnam (table 28).

In a review of 334 patients evacuated from Vietnam because of injury, who had had negative tuberculin tests immediately before arrival in Vietnam, 11 (or 3.3 percent) showed conversion. The average length of stay in Vietnam for these individuals was only 7 months. The author speculated that the conversion rate would have been higher with a full 1-year tour (Cowley 1970).

This suggestion was confirmed in a larger study done by Sowell, Russell, and Ionata (1973). This group studied U.S. Army enlisted men rotating to the United States from field units within the III Corps Tactical Zone, including the 1st Air Cavalry Division, 25th Infantry Division, 20th Engineer Brigade, and 11th Armored Cavalry Regiment. Among the 711 individuals studied, 35 (4.9 percent) demonstrated conversion. There was no difference in conversion based on age, rank, or MOS (military occupational specialty). However, a significant difference was found between whites and blacks. Whites had a 3.4-percent conversion rate (21 of 620) compared to 17.1 percent (12 of 70) in blacks. Overall distribution of troop strength for the U.S. Army in Vietnam by race was 87 percent white, 11 percent black, and 2 percent other.

These data are similar to those obtained in a U.S. Navy study at St. Albans Hospital in New York, which revealed a 4.7-percent conversion rate among Vietnam evacuees (Elliot, Miller, and Sachs 1970). The rates in both studies are higher than the 2.5 percent per annum conversion rate in Europe and the 1.0 percent per annum rate for Army personnel in the United States (Cowley 1970).

While the rate of tuberculin skin test conversion in U.S. servicemen in Viet-


TABLE 28.-Incidence of positive tuberculin skin test in personnel on first tour in Vietnam and personnel who had previous tours, 1970

nam seems reasonably well established, the incidence rate for the development of active pulmonary tuberculosis is not known, although apparently it was small. Data from the USARV (U.S. Army, Vietnam) medical consultant for 6 months in 1969 reveal that only 28 patients were evacuated from Vietnam because of active pulmonary disease.*

Early data suggested an increased incidence of primary drug-resistant tuberculosis in soldiers returning from Vietnam (Cowley and Briney 1970). This conclusion, however, was based on the discovery of only five cases of drug resistant disease among servicemen recently returned from Vietnam and on the supposition that there was an increased incidence of drug resistance in countries where administration of antituberculosis drugs is uncontrolled. There were no data on the incidence of drug-resistant strains in the local population in Vietnam. Primary drug resistance in developed countries has been reported to be 4 to 6 percent in the United States, 3.8 percent in Great Britain, 4.9 percent in Canada, and 9.8 percent in France (Fox 1968; Canetti 1965).

A Veterans Administration study of 3,183 strains of Mycobacterium tuberculosis from veterans in the United States showed no change in the susceptibility of the strains to streptomycin or isoniazid during the period 1962-69 (Hob by, Johnson, and Boytar-Papirnyik 1970). These findings are in accord with those of the Bureau of Tuberculosis, Department of Health, City of New York (Chaves 1970), and of the USPHS (U.S. Public Health Service).

Dantzker, Steinborg, and Kmiecik (1972) reported on the only large study dealing specifically with the problem of drug-resistant strains of tuberculosis in troops returned from Vietnam. They reviewed the records of 501 consecutive active-duty male tuberculosis patients who were discharged from the Army during 1967-70. The patients were divided into two groups: 201 who had had a tour of duty in Vietnam within 3 years of admission, and 300 who did not. Culture-proven disease was present in 127 patients in the Vietnam group and 154 in the non-Vietnam group. The percentage of resistant strains was 9.5 to 7.8, respectively, indicating no statistical evidence of increased drug resistance in the Vietnam group.


* Andre J. Ognibene: Unpublished data on hospital discharges of patients evacuated to Japan from Vietnam, 1969.



The clinical course of tuberculosis was apparently no different in American servicemen in Vietnam than it was in patients in the United States. Extrapulmonary forms and pleural effusion were uncommon.* Cavitary disease of the lung due to melioidosis caused confusion in early diagnoses at times, but this disorder was easily distinguished from tuberculosis with appropriate culture techniques (Weber et al. 1969).

Treatment of tuberculosis with chemotherapy at the present time is highly successful because of the variety of drugs which are effective against the organism. Primary or major drugs include isoniazid, para-aminosalicylic acid, ethambutol, streptomycin, and rifampin. Currently isoniazid is used in conjunction with either ethambutol or rifampin. Triple-drug therapy is recommended for extrapulmonary and far advanced pulmonary disease. The duration of treatment in Vietnam was 18 months for minimal pulmonary tuberculosis and pleural effusion and at least 24 months for extensive disease, cavitary disease, and extrapulmonary forms of the disease. When tubercle bacilli are found to be resistant to the primary drugs or when toxic symptoms preclude their use, so-called "secondary" drugs are available, including ethionamide, pyrazinamide, kanamycin, and cycloserine. Ideally, drug combinations are selected on the basis of sensitivity studies, and a combination of two or more effective drugs, either primary or secondary, must be used (TB MED).


Despite the large numbers of troops who served in Vietnam and the potential threat posed by an indigenous population with a high incidence of active disease, tuberculosis was not a major military medical problem for U.S. forces stationed there.

Section IV. Gram-Negative Infection

Brigadier General Andre J. Ognibene, MC, USA

The occurrence of nosocomial infections in Vietnam in the early years of the war was accorded little interest. Hospital-acquired infection was not a primary problem in the early management of patients, and evacuation policy was such that the susceptible patient was moved out of country rapidly. The continuing interest of the surgical staff in primary sepsis is recorded in their attempts to identify wound flora and analyze the effects of surgery in relation to the reduction of


1 Carl R. Guiton: Unpublished data.

Except where otherwise noted, this section is derived from the personal experience of the author and from the following article by him (1971): Newer patterns of infection encountered in Vietnam. In Symposium on changing patterns of bacterial infections and antibiotic therapy, ed. H.C. Neu. Excerpta Medica International Congress Series, no. 228, pp. 121-27.


infection rates (Pruitt and Baker). Nosocomial infection became a problem for study when in-country hospitals stabilized.

The development of relatively fixed installations and of increasing surgical capability, resulting in longer retention of patients in hospital, was a distinct feature of the later years of the war. This development, coupled with a capability for adequate specimen transport and bacteriologic support by the 9th Medical Laboratory, allowed a more specific diagnostic and therapeutic approach to gram-negative infections in Vietnam. In 1969, documentation of nosocomial infection was clearly obtained.


Before 1968, confirmed isolations of Serratia marcescens in Vietnam were rare. However, in late 1968 and early 1969, the organism was cultured from wounds and from the sputum of 12 patients and the blood of 8 of these 12. Of those with positive blood cultures, five died (table 29).

Because of experience with this initial group of patients, 2,600 blood cultures were subsequently examined at the 9th Medical Laboratory (Mobile) in Long Binh, with particular attention to Enterobacteriaceae. Of the 321 positive blood cultures, 100 (31 percent) were confirmed as S. marcescens. These 100 isolates were obtained from 40 patients at three hospitals in the vicinity of the laboratory.

This striking increase in the incidence of S. marcescens in blood cultures in Vietnam prompted attempts to isolate the source of the organism. Hospital infection committees were established for the first time in combat hospitals. Routine culturing practices in wards and operating rooms were undertaken. Initially S. marcescens could not be cultured from respirators or operating rooms, but a significant number of other pathogens were isolated, including both Klebsiella and Pseudomonas species.

Results of bacteriologic studies of the antibiotic sensitivity of 100 S. marcescens isolates in Vietnam were as follows:

Kanamycin - 43 %
Chloramphenicol - 41%
Polymyxin B - 34%
Colistin - 25%
Cephalothin - 23%
Tetracycline - 12%
Streptomycin - 5%
Ampicillin - 5%

Of particular importance is the striking finding that half of the strains isolated were resistant to all antibiotics. Gentamicin had not yet been used extensively and still enjoyed a place in the treatment of S. marcescens infections. This drug, however, was not available in Vietnam until late in 1969 and was not used for testing sensitivity of Serratia isolates.

Serratia marcescens is a short, almost spherical gram-negative rod. About 17 percent of isolates produced the pigment prodigiosin at room temperature. Nonpigmented forms present in Vietnam have also amply demonstrated their pathogenicity, as noted in the United States (Sanders et al. 1970; Clayton and von Graevenitz 1966). In the past the organism was confused with organisms of


TABLE 29.-Serratia marcescens infection in eight patients in Vietnam

the chromobacterium group because of its pigment production. However, the pigment has been clearly identified as prodigiosin and is not related to the viola-


cein pigment of the chromobacteria. S. marcescens has been properly classified as a member of the family Enterobacteriaceae. Formerly it had been regarded as a saprophyte; it has now been incriminated as a cause of endocarditis, lung abscess, pneumonia, and a variety of urinary tract infections (Alexander, Reichenbach, and Merendino 1969; Cabrera 1969). Epidemics related to the contamination of hospital respirators have been reported (Sanders et al. 1970).

Additional review of the initial hospital surveys in Vietnam showed that the organism could not be found in jugs of saline used for irrigation and cleansing of wounds as previously reported by Cabrera (1969). It was probably carried in the air. Cross-contamination of patients using nebulizers was possible before the use of disposable reservoir nebulizers, manifold tubing, and mouthpieces on all equipment. Such disposable parts are of prime importance in a combat area where patients with severe injuries often require prolonged periods of assisted ventilation and are particularly prone to contamination with S. marcescens.

The possible virulence of this particular organism demanded frequent blood cultures and meticulous care of wounded patients on assisted ventilation. Basic requirements included a more discriminate use of broad spectrum antibiotics in itially, greater attention to infection control procedures, and avoiding contamination of ventilation equipment.

The internist was not often involved in the care of the patient with a superinfection following wounding. The appearance of sepsis with S marcescens, however, fostered an exchange in expertise between medical and sur gical staffs in the prevention and management of such infections. It became clear that infection control in the hospital was of primary importance in the early management of the wounded patient, especially when combat support hospitals became fixed installations. Provision for effective bacteriologic analysis is critical to medical support of combat operations and must be available early if the establishment of relatively fixed support hospitals is contemplated.


Unusual gram-negative infections were not limited to surgical patients in Vietnam. In 1968, two isolates of Chromobacterium violaceum (janthinum), a gram-negative rod which characteristically produces an indigo pigment on culture, were reported on by the 9th Medical Laboratory (ML9-AR, p. 29). No details of patient status, source of infection, or other clinical data were obtainable. Fatal infection by this organism was first reported from Malaya in 1927 (Sneath et al. 1953). Of the 16 known human cases in the world, 9 came from that country. The most recent cases have been described in the southeastern United States (Dauphinais and Robben 1968; Nunnally and Dunlop 1968).

Two patients died of C. violaceum (janthinum) infection in Vietnam in 1969. The first, a 21-year-old soldier, entered the hospital with chills, fever, and a temperature of 105 0 F. He had lost 15 pounds over 2 weeks and appeared moderately ill, although his physical examination was unremarkable. The white blood count was elevated. He developed sudden cyanosis followed by respiratory arrest and died 8 hours after admission. Chest X-rays showed bilateral


pulmonary infiltrates. At autopsy, the lungs were studded with multiple small yellow abscesses, and necrotizing pneumonia was present. C. violaceum (janthinum) was grown from the premortem blood and tracheal cultures as well as from both lungs at autopsy.

The second patient was a 19-year-old soldier who had marked vomiting, diarrhea, and abdominal cramps for 3 days. He was noted to be markedly jaundiced and was thought to have fulminant hepatitis. His temperature rose to 105° F, and he was hypotensive. Hepatomegaly was the sole abnormal physical finding. Although kanamycin and cephalothin were begun, he died on the first hospital day. All blood cultures grew C. violaceum (janthinum). At autopsy, there were abscesses in the liver, a thickened gallbladder, and evidence of a necrotizing infection in the biliary tree and the liver.

Septicemia with liver abscess is characteristic of the infection. A number of patients had skin lesions which disseminated up to 15 months later, resulting in death (Dauphinais and Robben 1968). A patient with a positive culture from a skin lesion was identified in Vietnam in 1969. Unfortunately, followup could not be achieved. 

The infection was first described in water buffalo from Malaya (Sneath et al. 1953). The organism displays a predilection for tropical and semitropical areas. Because of previous reports of its isolation from water supplies, samples of a number of water supplies in Vietnam were cultured and the organism was grown from the main water supply of a village in central Vietnam. Positive isolations were not obtained from the rectal swabs of 50 water buffaloes.

Studies of four isolates (from two autopsies, one skin lesion, and the water supply) showed that sensitivity existed to tetracycline and chloramphenicol, but there was resistance to cephalothin, kanamycin, ampicillin, and colistin. 

Because of the latent period, reported by some, between appearance of a skin lesion and later dissemination, one must speculate about the probability of importation of this disease from Vietnam to the United States by returning servicemen. In five cases reported from Florida and Louisiana, no mention was made of military service or other exposure in Southeast Asia. However, experience with melioidosis suggested that an indolent C. violaceum (janthinum) infection should be considered in the differential diagnosis of infectious disease in returning servicemen.


While the occurrence of unusual gram-negative infection was recognized, some gram-negative infections of worldwide distribution did not cause significant problems. For example, the development of meningococcal disease in U.S. troops was not a feature of the Vietnam war. However, in 1967, 15 strains of meningococci were recovered from North Vietnamese carriers in a prisoner-of-war camp after there had been seven cases and one death. These were later characterized as only nine strains-seven Type B, one Type C, and one Type A*


*Malcolm S. Artenstein: Personal communication.


(table 30). Meningococcal meningitis also occurred in Vietnamese troops, but their length of service and the extent of the problem were not clear.

TABLE 30.-Sensitivity characteristics of Neisseria meningitidis strains from Vietnam

Studies done at Fort Dix and Fort Benning in 1966 and 1968 (Goldschneider, Gotschlich, and Artenstein 1969b) demonstrated that 92 percent of tested recruits developed serum bactericidal activity against acquired strains of men ingococcus. On the basis of this information, it might be speculated that the American soldier arriving in Vietnam had been "seasoned" in basic training in relation to the meningococcus organism. 

Goldschneider and coworkers believed this to be a reasonable explanation for the universal observation that seasoned military personnel are much less susceptible to meningococcal disease than are basic recruits. Such seasoned troops have apparently been immunized during basic training by means of the meningococcal carrier state. This is corroborated by a study in which 67 to 86 percent of a susceptible population of military recruits become carriers of meningococci other than the prevalent disease-producing strains (Goldschneider, Gotschlich, and Artenstein 1969a). Such individuals subsequently developed bactericidal antibodies to the pathogenic organisms.

The recognition of this fact was imperative in the reeducation of the physician new to the Army and to Vietnam who placed meningitis high in the differential diagnosis of the comatose and febrile young soldier and neglected falciparum malaria and Japanese B encephalitis. This was especially true when the encephalitis patient demonstrated a predominance of polymorphonuclear leukocytes in the spinal fluid. The change in the order of probability of these diseases to fit the Vietnam experience was ultimately accomplished and improved the practice of internal medicine in the Vietnam conflict.


Alain, M.; Saint-Ètienne, J.; and Reynes, V. 1949. La mélioidose: Considerations étiologiques, cliniques et pathogeniques à propos de 28 cas. Med trop. 9: 119-42.

Albizo, J. M., and Surgalla, M. J. 1970. Isolation and biological characterization of Pasteurella pestis endotoxin. Infection and Immunity 2: 229-36.

Alexander, R. H.; Reichenbach, D. D.; and Merendino, K. A. 1969. Serratia marcescens endocarditis. A review of the literature and report of a case involving a homograft replacement of the aorticvalve. Arch. Surg. 98: 287-91.

Bacot, A. W., and Martin, C. J.1914. Observations on the mechanism of the transmission of plague by fleas. J. Hyg. (Plague Supp. 3) 13: 423-39.

Bahmanyar, M.1972. Human plague episode in the district of Khawlan, Yemen. Am. J. Trop. Med. 21: 123-28.

Baltazard, M; Davis, D. H. S.; Devignat, R.; Girard, G.; Gohar, M. A.; Kartman, L.; Meyer, K. F.; Parker, M.T.; Pollitzer, R.; Prince, F. M.; Quan, S. F.; and Wagle, P. 1956. Recommended laboratory methods for the diagnosis of plague. Bull. World Health Organ. 14: 485.

Bartelloni, P. J.; Marshall, J. D., Jr.; and Cavanaugh, D. C. 1973. Clinical and serological responses to plague vaccine. U.S.P. Mil. Med 138:720-22.

Baumann, B. B., and Morita, E. T. 1967. Systemic melioidosis presenting as myocardial infarct. Ann. Int. Med 67: 836-42.

Beamer, P. R.; Varney, P. L.; Brown, W. G.; and McDowell, F. 1954. Studies on Malleomyces pseudomallei isolated from melioidosis originating in the Western Hemisphere. Am. J. Clin. Path. 24: 1231-40.

Beasley, P. 1969. Human plague in the United States. J.A.M.A. 208: 1024-25.

Blanc, G., and Baltazard, M. 1941a. Transmission du bacille de Whitmore par le moustique Aedes (Stegomyia) aegypti Compt. rend 213: 670-72.

______ . 1941b. Transmission du bacilli de Whitmore par la puce du rat Xenopsylla cheopis. Compt. rend 213: 541-43.

Borchardt, K. A.; Stansifer, P.; and Albano, P. M. 1966. Osteomyelitis due to Pseudomonas pseudomallei. J.A.M.A. 196:660-62.

Brooks, R. St. J. 1917. The influence of saturation deficiency and of temperature on the course of epidemic plague. J. Hyg. (Plague Supp. 5) 15: 881-99.

Brygoo, E. R. 1953. Contribution à 1'étude des agglutinines naturalles pour le bacille de Whitmore. Bull. Soc. path. exot. 46: 347-53.

Burkle, F., Jr. 1973. Plague as seen in South Vietnamese children. A chronicle of observations and treatment under adverse conditions. Clin. Pediat. (Phila.) 12: 291-98.

Burmeister, R. W.; Tigertt, W. D.; and Overholt, E. L. 1962. Laboratory-acquired pneumonic plague. Report of a case and a review of previous cases. Ann. Int. Med 56: 796.

Burrows, T. W. 1955. The basis of virulence for mice of Pasteurella pestis. In Mechanisms of microbial pathogenicity, ed. J. W. Howie and A. J. O'Hea, pp. 152-75. Cambridge: University Press.

______. 1957. Virulence of Pasteurella pestis. Nature, London 179: 1246-47.

______. 1960. Virulence determinants in Pasteurella pestis and Past. pseudotuberculosis. In Proc. Syrup., 10-14 Jan. 1959, Haffkine Inst., Bombay, pp. 14-17.

Burrows, T. W., and Bacon, G. A. 1954. The basis of virulence in Pasteurella pestis. Comparative behaviour of virulent and avirulent strains in vivo. Brit. J. Exper. Path. 35: 134-43.

______. 1956. The basis of virulence in Pasteurella pestis: The development of resistance to phagocytosis in vitro. Brit. J. Exper. Path. 37: 286-99.

Burrows, T. W., and Gillett, W. A. 1971. Host specificity of Brazilian strains of Pasteurella pestis. Nature, London 229: 51-2.

Butler, T. 1972. A clinical study of bubonic plague. Observations of the 1970 Vietnam epidemic with emphasis on coagulation studies, skin histology and electrocardiograms. Am. J. Med. 53: 268-76.

Cabrera, H. A.1969. An outbreak of Serratia marcescens and its control. Arch. Int. Med. 123:650-55.

Calabi, 0. 1973. Bactericidal synergism of novobiocin and tetracycline against Pseudomonas pseudomallei J. Med. Microbiol. 6: 293-306.

Canetti, G. 1965. Present aspects of bacterial resistance in tuberculosis. Am. Rev. Resp. Dis. 92: 687-703.

Caten, J. L., and Kartman, L. 1968. Human plague in the United States, 1900-1966. JA.M.A. 205: 333-36.

Cavanaugh, D. C.1971. Specific effect of temperature upon transmission of the plague bacillus by the oriental rat flea, Xenopsylla cheopis. Am. J. Trop. Med 20: 264-73.

Cavanaugh, D. C., and Marshall, J. D., Jr. 1972. The influence of climate on the seasonal prevalence of plague in the Republic of Vietnam. J. Wildlife Dis. 8: 85-94.

<>Cavanaugh, D. C., and Quan, S. F. 1953. Rapid identification of Pasteurella pestis using specific bacteriophage lyophilized on strips of filter paper. A preliminary report. Am. J. Clin. Path. 23: 619-20.

Cavanaugh, D. C., and Randall, R. 1959. The role of multiplication of Pasteurella pesos in mononuclear phagocytes in the pathogenesis of flea-borne plague. J. Immunol. 83: 348-63.
Cavanaugh, D. C.; Dangerfield, H. G.; Hunter, D. M.; Joy, R. J. T.; Marshall, J. D., Jr.; Quy, D. V.; Vivona, S.; and Winter, P. E. 1968. Some observations on the current plague outbreak in the Republic of Vietnam. Am. J. Pub. Health 58: 742-52.

Cavanaugh, D. C.; Deoras, P. J.; Hunter, D. H.; Marshall, J. D., Jr.; Do-Van-Quy; Rust, J. H., Jr.; Purnaveja, S.; and Winter, P. E. 1970. Some observations on the necessity for serological testing of rodent sera for Pasteurella pesos antibody in a plague control programme. Bull. World Health Organ. 42: 451-59.

Cavanaugh, D. C.; Do-Van-Quy; and Ky-Vinh-Thai. 1964. A cooperative study of plague in the Republic of Vietnam. Research proposal submitted to Ministry of Health, Republic of Vietnam, 10 Oct. 64.

Cavanaugh, D. C.; Elisberg, B. L.; Llewellyn, C. H.; Marshall, J. D., Jr.; Rust, J. H., Jr.; Williams, J. E.; and Meyer, K. F. 1974. Plague immunization. V. Indirect evidence for the efficacy of plague vaccine. J. Infect. Dis. (Supp.) 129: 37-40.

Cavanaugh, D. C.; Hunter, D. H.; Nguyen-Van-Ba; Tran-Cong Dung; Ryan, P. F.; and Marshall, J. D.; Jr. 1968. Ecology of plague in Vietnam. III. Sylvatic plague; Bandicota indica, a transitional species. Tr. Roy. Soc. Trop. Med and Hyg. 62: 456.

Cavanaugh, D. C.; Ryan, P. F.; and Marshall, J. D., Jr. 1969. The role of commensal rodents and their ectoparasites in the ecology and transmission of plague in Southeast Asia. Bull. Wildlife Dis. Assoc. 5: 187-94.

Cavanaugh, D. C.; Stark, H. E.; Marshall, J. D., Jr.; and Rust, J. H., Jr. 1972. A simple method for rearing fleas for insecticide testing in the field. J. Med Ent. 9: 113-14.

Cavanaugh, D. C.; Thorpe, B. D.; Bushman, J. B.; Nicholes, P. S.; and Rust, J. H., Jr. 1965. Detection of an enzootic plague focus by serological methods. Bull. World Health Organ. 32: 197-203.
Chambon, L.1955. Isolement du bacille de Whitmore a partir du milieu exterieur. Ann. Inst. Pasteur 89: 229-35.

Chaves, A. D. 1970. Isoniazid-resistant strain of M. tuberculosis. Am. J. Pub. Health 60: 612-13.
Chen, T. H., and Meyer, K. F.1954. Studies on immunization against plague. VII. A hemagglutination test with the protein fraction of Pasteurella pestis: A serologic comparison of virulent and avirulent strains with observations on the structure of the bacterial cells and its relationship to infection and immunity. J. Immunol. 72: 282-98.

Chen, T. H.; Quan, S. F.; and Meyer, K. F. 1952. Studies on immunization against plague. II. The complement-fixation test. J. Immunol. 68: 147-58.

Clayton, A. J.; Lisella, R. S.; and Martin, D. G.1973. Melioidosis: A serological survey in military personnel. Mil. Med 138: 24-26.

Clayton, E., and von Graevenitz, S. 1966. Nonpigmented Serratia marcescens. JA.M.A. 197: 1059-64.

Cohen, R. J., and Stockard, J. L. 1967. Pneumonic plague in an untreated plague-vaccinated individual. J.A.M.A. 202:365-66.

Colwell, E. J.; Brown, J. D.; Russell, P. K.; Boone, S. C.; Legters, L. J.; and Catino, D.1969. Investigations on acute febrile illness in American servicemen in the Mekong Delta of Vietnam. Mil. Med. 134:1409-14.

Communicable diseases: Arthropodborne diseases other than malaria, Preventive Medicine in World War II. See MD-PM7.

Cook, 1.1962. A survey for antibodies to melioidosis in man and native animals. M. J. Australia 49(2): 627-28.

Cooper, E. B. 1967. Melioidosis. JA.M.A. 200: 452-53.

-------. 1968. Treatment of melioidosis (Letters to the Editor). JA.M.A. 204: 176.

Cowley, R. G. 1970. Implications of the Vietnam War for tuberculosis in the United States. Arch. Environ. Health 21: 479-80.

Cowley, R. G., and Briney, R. R. 1970. Primary drug-resistant tuberculosis in Vietnam veterans. Am. Rev. Resp. Dis. 101: 703-5.

Cox, C. D., and Arbogast, J. L. 1945. Melioidosis. Am. J. Clin. Path. 15: 567-70.

Cravitz, L., and Miller, W. R. 1950. Immunologic studies with Malleomyces mallei and Malleomyces pseudomallei. I. Serological relationships between M. mallei and M. pseudomallei J. Infect Dis. 86: 46-51.

Cutting, R. T. et al. Plague in Vietnamese civilians-clinical and laboratory study. Military Medical Research Program SEA, U.S. Army Medical Research Team, Vietnam. Research in Biological and Medical Sciences Annual Progress Report, vol. II, 1 July 1969-30 June 1970.

Dannenberg, A. M., Jr., and Scott, E. M. 1958a. Melioidosis: Pathogenesis and immunity in mice and hamsters. I. Studies with virulent strains of Malleomyces pseudomallei. J. Exper. Med. 107: 153-66.

______. 1958b. Melioidosis: Pathogenesis and immunity in mice and hamsters. II. Studies with avirulent strains of Malleomyces pseudomallei. Am. J. Path. 34: 1099-1121.

Dantzker, D. R.; Steinborg, H. N.; and Kmiecik, J. E. 1972. Primary drug-resistant tuberculosis in Vietnam veterans, 1967 to 1970, Am. Rev. Resp. Dis. 106: 273-74.

Dauphinais, R. M., and Robben, G. G. 1968. Fatal infection due to Chromobacterium violaceum. Am. J. Clin. Path. 50: 592-97.

Deaton, J. G. 1969. Febrile illness in the Tropics (Vietnam). MiL Med. 134: 1403-8.

Deller, J. J., Jr., and Russell, P. K. 1967. An analysis of fevers of unknown origin in American soldiers in Vietnam. Ann. Int. Med. 66: 1129-43.

Diagnosis and management of tuberculosis, Department of the Army Technical Bulletin (Medical) See TB MED.

Edwards, P. Q. !969. Significance of the tuberculin test today. Clin Notes Resp. Dis. 8: 3-12.
Eickhoff, T. C.; Bennett, J. V.; Hayes, P. S.; and Feeley, J. 1970. Pseudomonas pseudomallei Susceptibility to chemotherapeutic agents. J. Infect. Dis. 121: 95-102.

Elliot, R. C.; Miller, C. H.; and Sachs, J.1970. The incidence of tuberculosis among Marine Corps and Navy personnel in Vietnam. Am. Thorac. Soc. Annual Meeting M. Abstr. 3.

Ertug, C. 1961. Melioidosis. Dis. Chest 40: 693-97.

Everett, E. D., and Nelson, R. A.1975. Pulmonary melioidosis. Observations in thirty-nine cases. Am. Rev. Resp. Dis. 112: 331-40.

Feeley, E. J.; and Kriz, J. J. 1965. Plague meningitis in an American serviceman. JAMA. 191: 412-13.

Finegold, M. J. 1968. Pathogenesis of plague. A review of plague deaths in the United States during the last decade. Am. J. Med. 45: 549-54.

Finegold, M. J.; Petery, J. J.; Berendt, R. F.; and Adams, H. R. 1968. Studies on the pathogenesis of plague. Blood coagulation and tissue responses of Macaca mulatta following exposure to aerosols of Pasteurella pestis. Am. J. Path. 53: 99-114.

Flemma, R. J.; DiVincenti, F. C.; Dotin, L. N.; and Pruitt, B. A., Jr. 1969. Pulmonary melioidosis. A diagnostic dilemma and increasing threat. Ann Thoracic Surg. 7: 491-99.

Fournier, J., and Chambon, L. 1958. La mélioidose: Maladie d'actualité et le bacille de Whitmore (Malleomyces pseudo-mallei). Paris: Editions Medicales Flammarion.

Fox, W.1968. Changing concepts in the chemotherapy of pulmonary tuberculosis (The John Barnwell lecture). Am. Rev. Resp. Dis. 97: 767-90.

Girard, G. 1963. L'immunité dans 1'infection pesteuse. Acquisitions apportées par 30 anees de travaux sur la souche de "Pasteurella pesos EV." Biol mid. 52: 631-731.

Goldberg, B. 1941. War and tuberculosis (Presidential Address). Dis. Chest 7: 322-25.
Goldschneider, I.; Gotschlich, E.C.; and Artenstein, M.S. 1969a. Human immunity to the meningococcus. I. The role of humoral antibodies. J. Exper. Med. 129: 1307-26.

______. 1969b. Human immunity to the meningococcus. II. Development of natural immunity. J. Exper. Med 129:1327-48.

Green, R. N., and Tuffnell, P. G. 1968. Laboratory acquired melioidosis. Am. J. Med 44: 599-605.
Greenwood, M., Jr. 1911. Statistical investigations of plague in the Punjab. Third Report: On some factors which influence the prevalence of plague. Sixth report on plague investigations in India. J. Hyg. (Plague Supp. 1) 11: 62-156.

______. 1913. The factors that determine the rise, spread, and degree of severity of epidemic diseases. Internat. Cong. Med, London 18: 49-72.

______. 1935. Epidemics and crowd-diseases. An introduction to the study of epidemiology, pp. 289-309. London: Williams & Norgate.

Health of the Army. See HOA.

Hirst, L. F. 1953. The conquest of plague: A study of the evolution of epidemiology. Oxford: Clarendon Press.

HOA-Office of the Surgeon General, War Department/United States Army. Health of the Army, 31 Aug. 1946, May 1952, May 1953, and May 1954. Copies at Uniformed Services University of the Health Sciences.

Hobby, G. L.; Johnson, P. M.; and Boytar-Papirnyik, V. 1970. Primary drug resistance: A continuing study of drug resistance in tuberculosis in a veteran population within the United States. VII. September 1965 to September 1969. Am. Rev. Resp. Dis. 102: 347-55.

Houk, V. N. 1967. Problems in global medicine. Pulmonary diseases. New York J. Med. 67: 2461. Howe, C.; Sampath, A.; and Spotnitz, M. 1971. The Pseudomallei group: A review. J. Infect. Dis. 124: 598-606.

Hudson, B. W.; Goldenberg, M. I.; McCluskie, J. D.; Larson, H. E.; McGuire, C. D.; Barnes, A. M.; and Poland, J. D. 1971. Serological and bacteriological investigations of an outbreak of plague in an urban tree squirrel population. Am. J. Trop. Med. 20: 255-63.

Hudson, B. W.; Quan, T. J.; Sites, V. R.; and Marshall, J. D.1973. An electrophoretic and bacteriologic study of Yersinia pestis isolates from central Java, Asia, and the Western Hemisphere. Am. J. Trop. Med. 22:642-53.

Hunter, D. H., and Dangerfield, H. G. 1967. Plague in Vietnam. USARVM. Bull. (USARV Pam 40-1), Jan.-Feb., pp. 34-38. Copy in Joint Medical Library, Office of the Surgeons General. Infectious diseases, Internal Medicine in World War II. See MD-IM2.

Jackson, A. E.; Moore, W. L., Jr.; and Sanford, J. P. 1972. Recrudescent melioidosis associated with diabetic ketoacidosis. Arch. Int. Med. 130: 268-71.

James, A. E.; Dixon, G. D.; and Johnson, H. F. 1967. Melioidosis: A correlation of the radiologic and pathologic findings. Radiology 89: 230-35.

Johnson, Capt. Curtis I., MC, Preventive Medicine Officer, 25th Infantry Division. 1968. Melioidosis: A brief review and comments on the problem in the 25th Infantry Division. Report, 6 Sept. 68. Joubert, L., and Phung Van Dan. 1958. Epidemiologie et prophylaxie de la melioidose, zoonose tropicale. Rev. Elev. Med. Vet. Pays Trop. 11: 23-29.

Kartman, L. 1958. An insecticide-bait-box method for the control of sylvatic plague vectors. J. Hyg. 56: 455-65.

______. 1960. Further observations on an insecticide-bait-box method for the control of sylvatic plague vectors; effect of prolonged field exposure to DDT powder. J. Hyg. 58: 119-24.

______. 1969. Effect of differences in ambient temperature upon the fate of Pasteurella pesos in Xenopsylla cheopis. Tr. Roy. Soc. Trop. Med. & Hyg. 63: 71-75.

Kishimoto, R. A.; Brown, G. L.; Blair, E. B.; and Wenkheimer, D. 1971. Melioidosis: Serologic studies on U.S. Army personnel returning from Southeast Asia. Mil. Med. 136: 694-88.

Knapp, H. H. G. 1915. Morphine injector's septicemia ("Whitmore's Disease"). Indian M. Gaz. 1: 287-88.

Knülle, W. 1967. Physiological properties and biological implications of the water vapour sorption mechanism in larvae of the oriental rat flea, Xenopsylla cheopis (Roths). J. Insect Physiol. 13: 333-57.

Krishnaswami, C. S. 1917. Morphia injector's septicemia. Indian M. Gaz. 52: 296-99. Laboratory procedures: Melioidosis. See SEATO.

Landsborough, D., and Tunnell, N. 1947. Observations on plague meningitis. Brit. M. J. 1: 4-7. Lawton, W. D.; Fukui, G. M.; and Surgalla, M. J. 1960. Studies on the antigens of Pasteurella pestis and Pasteurella pseudo tuberculosis. J. Immunol. 84: 475-79.

Legters, L. J.; Cottingham, A. J., Jr.; and Hunter, D. H. 1970. Clinical and epidemiologic notes on a defined outbreak of plague in Vietnam. Am. J. Trop. Med. 19: 639-52.

Legters, L. J.; Hunter, D. H.; Proctor, R. F.; and Conrad, F.G. Clinical and epidemiological notes on an outbreak of pneumonia in Rach Gia, Kien Giang Province, IV Corps Tactical Zone, April-May 1967. In Annual Progress Report, U.S. Army Medical Research Team (WRAIR) Vietnam and Institute Pasteur of Vietnam, 1 Sept. 1966-31 Aug. 1967, pp. 444-62.

Le Moine; Hasle; and Nguyen-Due-Khoi.1937. Mélioidose à forme septicemique suraiguë consécutive à un violent traumatisme et souillure des plaies par boue de riziàre. Bull. Soc. Med Chir. Indochine, pp. 662-64.

Liston, W. G. 1903. Some observations on fleas and some facts which would appear to associate these insects with the spread of plague. Tr. Bombay Med. & Phys. Soc. 7: 8-22.

______. 1904. Plague, rats, and fleas. J. Bombay Nat. Hist. Soc. 16: 253-74.

______. 1905. Plague, rats, and fleas. Indian M. Gaz. 40: 43-49.

Lowson, J. A. 1898. Presentation at the Hong Kong Medical Association, Dec. 1898. Quoted by R. Green (1951) in The Institute for Medical Research, 1900-1950, Federation of Malaya. Jubilee vol. 25. Kuala Lumpur: Government Press.

Macchiavello, A. 1954. Reservoirs and vectors of plague. J. Trop. Med 57: 3-8.

Marshall, J. D., Jr.; Bartelloni, P. J.; Cavanaugh, D. C.; Kadull, P. J.; and Meyer, K. F. 1974. Plague immunization. II. Relation of adverse clinical reactions to multiple immunizations with killed vaccine. J. Infect. Dis. (Supp.) 129: 19-25.

Marshall, J. D., Jr.; Cavanaugh, D. C.; Bartelloni, P. J.; and Meyer, K. F. 1974. Plague immunization. III. Serologic response to multiple inoculations of vaccine. J. Infect. Dis. (supp.) 129: 26-29.

Marshall, J. D., Jr.; Currie, J. A.; and Quy, D. V. 1968. Serological survey of small mammals of Vietnam for antibody against "Pasteurella pestis "and "Pasteurella pseudotuberculosis. "In Internat. Symp Pseudotuberculosis, Paris, 1967. Symp. Series Immunobiol. Standard, vol. 9, pp. 309-18. Basel/New York: Karger.

Marshall, J. D., Jr.; Joy, R. J. T.; Ai, N. V.; Quy, D. V.; Stockard, J. L.; and Gibson, F. L. 1967. Plague in Vietnam 1965-1966. Am. J. Epidemiol 86: 603-16.

Marshall, J. D., Jr.; Quy, D. V.; and Gibson, F. L. 1967. Asymptomatic pharyngeal plague infection in Vietnam. Am. J. Trop. Med 16:175-77.

Marshall, J. D.; Quy, D. V.; Gibson, F. L.; Dung, T. C.; and Cavanaugh, D. C. 1967a. Ecology of plague in Vietnam: Commensal rodents and their fleas. Mil. Med. 132: 896-903.

_______. 1967b. Ecology of plague in Vietnam. I. Role of Suncus murinus. Proc. Soc. Exper. Biol. & Med. 124: 1083-86.

Mays, E. E., and Ricketts, E. A. 1975. Melioidosis: Recrudescence associated with bronchogenic carcinoma twenty-six years following initial geographic exposure. Chest 68: 261-62.

McCormick, J. B.; Sexton, D. J.; McMurray, J. G.; Carey, E.; Hayes, P.; and Feldman, R. A. 1975. Human-to-human transmission of Pseudomonas pseudomallei. Ann. Int. Med. 83: 512-13. McCrumb, F. R., Jr.; Larson, A.; and Meyer, K. F.1953. The chemotherapy of experimental plague in the primate host. J. Infect. Dis. 92: 273-87.

McCrumb, F. R., Jr.; Mercier, S.; Robic, J.; Bouillat, M.; Smadel, J. E.; Woodward, T. E.; and Goodner, K.1953. Chloramphenicol and terramycin in the treatment of pneumonic plague. Am. J. Med. 14: 284-93.

McDowell, F., and Varney, P. L. 1947. Melioidosis: Report of first case from the Western Hemisphere. JA.M.A. 134: 361-62.

MD-IM2-Medical Department, U.S. Army. 1963. Infectious diseases. Internal Medicine in World War II, vol. II. Washington: Government Printing Office.

MD-PM7-Medical Department, U.S. Army. 1964. Communicable diseases: Arthropodborne diseases other than malaria. Preventive Medicine in World War II, vol. VII, pp. 79-100. Washington: Government Printing Office.

MD-PS-Medical Department, U.S. Army. 1967. Physical standards in World War II. Washington: Government Printing Office.

MD-WW15-Medical Department, U.S. Army. 1925. Statistics, part 2. The Medical Department of the U.S. Army in the World War, vol. XV. Washington: Government Printing Office.

Melioidosis Registry. See OTSG-MR.

Meyer, K. F. 1970. Effectiveness of live or killed plague vaccines in man. Bull World Health Organ. 42: 653-66.

______.1971. The clinical and immunological response of man to P. pesos vaccine. Proceedings of the Symposium on Bacterial Vaccines, Yugoslav Acad. Sc. Arts, Zagreb, pp. 299-312.

Meyer, K. F.; Cavanaugh, D. C.; Bartelloni, P. J.; and Marshall, J. D., Jr. 1974. Plague immunization. I. Past and present trends. J. Infect. Dis. (Supp.) 129: 13-18.

Meyer, K. F.; Smith, G.; Foster, L. E.; Marshall, J. D., Jr.; and Cavanaugh, D. C. 1974. Plague immunization. IV. Clinical reactions and serologic response to inoculations of Haffkine and freezedried plague vaccine. J. Infect. Dis. (Supp.) 129: 30-36.

ML9-AR-9th Medical Laboratory. Activities Reports, 1967 and 1968. On file at U.S. Army Center of Military History.

Mollaret, H. H. 1963. Conservation expérimentale de la peste dans le sol. Bull, Soc. path. exot. 56: 1168-82.

Moody, M. D., and Winter, C. C. 1959. Rapid identification of Pasteurella pestis with fluorescent antibody. III. Staining Pasteurella pestis in tissue impression smears. J. Infect. Dis. 104: 288-94. Moore, W. L., Jr.; Hedberg, C. L.; and Lindberg, R. L. 1970.

Melioidosis. In Tice's Practice of Medicine, vol. 3, chap. 39. Hagerstown: Harper & Row Publishers, Inc.

Morgan, H. J.; Wright, I. S.; and Van Ravenswaay, A. 1946. Health of repatriated prisoners of war from the Far East. JA.M.A. 130: 995-99.

Nguyen-Van-Ai; Vanderkove, M.; Nguyen-Van-Ba; Louis, J.; and Do-Van-Quy. 1963. Situation of the plague in South Vietnam. Outline of the epidemiology of the plague in South Vietnam during the last 8 years. Rapport Annuel sur le Fonctionnement Technique, 1963. Saigon: Institute Pasteur.
Nigg, C. 1963. Serologic studies on subclinical melioidosis. J. Immunol, 91: 18-28.

Nigg, C., and Johnston, M. M.1961. Complement fixation test in experimental clinical and subclinical melioidosis. J. Bact. 82:159-68.

9th Medical Laboratory. See ML9-AR.

Nunnally, R. M., and Dunlop, W. H. 1968. Fatal septicemia due to Chromobacterium janthinum. J. Louisiana M. Soc. 120: 278-80.

Ogata, M. 1897. Ueber die Pestepidemie in Formosa. Zentralbl. Bakt. 21: 769-77.

Ognibene, A. J. 1971. Newer patterns of infection encountered in Vietnam. In Symposium on changing patterns of bacterial infections and antibiotic therapy, ed. H.C. Neu. Excerpta Medica International Congress Series, no. 228, pp. 121-27.

OM-PHD-U.S. Operations Mission, Public Health Division. Brief study on a plague outbreak, August-September 1962, Saigon, Vietnam. Report, undated.

Osteraas, G. R.; Hardman, J. M.; Bass, J. W.; and Wilson, C. 1971. Neonatal melioidosis. Am. J. Dis. Child 122: 446-48.

OTSG-MR-Biostatistics Agency, Office of the Surgeon General, Department of the Army. Melioidosis Registry.

Peck, C. R., and Zwanenburg, T.1947. A case of melioidosis presenting as an abscess in the neck. Brit. M. J. 1: 337-38.

Pham-Trong; Tran-Quy-Nhu; and Marshall, J. D., Jr. 1967. A mixed pneumonic bubonic plague outbreak in Vietnam. Mil, Med 132: 93-97.

Physical standards in World War II. See MD-PS.

PMU-20-20th Preventive Medicine Unit, USARV. Annual Report, 1970.

Poffenbarger, P. L. 1972. Tuberculosis in South Vietnam. Am. J. Trop. Med 21: 226-33.

Poland, J. D. 1971. Diagnosis and therapy of plague. Proc. Plague Symp., Saigon, Oct. 71.

_______. 1972. Plague. In Infectious diseases. A guide to the understanding and management of infectious processes, ed. P. D. Hoeprich. New York: Harper & Row.

Pollitzer, R. 1954. Plague. Geneva: World Health Organization.

______ . 1960. A review of recent literature on plague. Bull, World Health Organ. 23: 313-400.

Pons, R., and Advier, M. 1927. Melioidosis in Cochin China. J. Hyg. 26: 28-30.

Pratt, H. D. 1967. Plague in Vietnam -a possible threat to the United States. Vector Control Briefs. No. 20, Nov. 67, pp. 12-13. Public Health Service, National Communicable Disease Center, U.S. Department of Health, Education, and Welfare.

Prevatt, A. L., and Hunt, J. S. 1957. Chronic systemic melioidosis. Am. J. Med 23: 810-23.

Pruitt, B., and Baker, H. Bacterial population as related to treatment and subsequent infection in war wounds. Annual Progress Report, U.S. Army Medical Research Team (WRAIR) Vietnam and Institute Pasteur of Vietnam, 1 Sept. 1967-30 June 1968, pp. 398-405.

Quan, S. F.; Knapp, W.; Goldenberg, M. L; Hudson, B. W.; Lawton, W. D.; Chen, T. H.; and Kartman, L. 1965. Isolation of a strain of Pasteurella pseudo tuberculosis from Alaska identified as Pasteurella pestis: An immunofluorescent false positive. Am. J. Trop. Med 14: 424-32.

Redfearn, M. S.; Palleroni, N. J.; and Stanier, R. Y. 1966. A comparative study of Pseudomonas pseudomallei and Bacillus mallei. J. Gen. MicrobioL 43: 293-313.

Reiley, C. G., and Russell, P. K. 1969. Observations on fevers of unknown origin in the Republic of Vietnam. Mil. Med 134: 36-42.

Rogers, L. 1928. The yearly variations in plague in India in relation to climate Forecasting epidemics. Proc. Roy. Soc. London S. B. 103: 42-72.

Rosebury, T. 1947. Experimental air-borne infection. Microbiological Monographs. Baltimore: Williams & Wilkins.

Rubin, H. L.; Alexander, A. D.; and Yager, R. H.1963. Melioidosis-a military medical problem? Mil. Med. 128: 538-42.

Rust, J. H., Jr.; Cavanaugh, D. C.; Kadis, S.; and Ajl, S. J.1963. Plague toxin: its effect in vitro and in vivo. Science 142: 408-9.

Rust, J. H., Jr.; Harrison, D. N.; Marshall, J. D., Jr.; and Cavanaugh, D.C. 1972. Susceptibility of rodents to oral plague infection: A mechanism for the persistence of plague in inter-epidemic periods. J. Wildlife Dis. 8: 127-33.

Rust, J. H., Jr.; Miller, B. E.; Bahmanyar, M.; Marshall, J. D., Jr.; Purnaveja, S.; Cavanaugh, D. C.; and U Saw Tin Hla. 1971. The role of domestic animals in the epidemiology of plague. II. Antibody to Yersinia pestis in sera of dogs and cats. J. Infect. Dis. 124: 527-37.

Salisbury, W. A., and Likos, J. J. 1970. Pseudomonas pseudomallei in a case of chronic melioidosis. Am. J. Clin. Path. 54:602-6.

Sanders, C. V., Jr.; Luby, J. P.; Johanson, W. G., Jr.; Barnett, J. A.; and Sanford, J. P. 1970. Serratia marcescens infections from inhalation therapy medications: Nosocomial outbreak. Ann. Int. Med. 73:15-21.

Schaeffer, H. C., and Grant, R. 0., Jr. 1968. Melioidosis presenting as meningoencephalitis. USARV M. Bull. (USARV Pam 40-7), Jan.-Feb., pp. 62-64. Copy in Joint Medical Library, Office of the Surgeons Genera],

Simeons, A. T. W., and Chhatre, K. D. 1946. One thousand cases of bubonic plague treated in an emergency plague hospital. Indian M. Gaz. 81: 235-38.

Simond, P. L. 1898. La propagation de la peste. Ann. Inst. Pasteur 12: 625-87.

Simpson, W. J. 1905. A treatise on plague dealing with the historica4 epidemiologica4 clinica4 therapeutic and preventive aspects of the disease. Cambridge: University Press.

Sites, V. R.; Poland, J. D.; and Hudson, B. W. 1972. Bubonic plague misdiagnosed as tularemia. J.A.M.A. 222:1642-43.

Smadel, J. E.; Woodward, T. E.; Amies, C. R.; and Goodner, K. 1952. Antibiotics in the treatment of bubonic and pneumonic plague in man. Ann. New York Acad. Sc. 55: 1275-84.

Sneath, P. H. A.; Singh, R. B.; Whelan, J. P. F.; and Edwards, D. 1953. Fatal infection by Chromobacterium violaceum. Lancet 2: 276-77.

Sollier, L. F., and Boutareau, C. 1937. Un cas de melioidose a manifestations cliniques prostatourinaires. Bull. Soc. Med Chir. Indochine 15: 8-10.

Sowell, J. M.; Russell, R. M.; and Ionata, V. J. 1973. Tuberculin tine test conversion among United States Army enlisted personnel in Vietnam. Mil. Med 138: 96-98.

Spotnitz, M. 1968. Treatment of melioidosis (Letters to the Editor). J.A.M.A. 204: 176.

Spotnitz, M.; Rudnitzky, J.; and Rambaud, J. J.1967. Melioidosis pneumonitis. Analysis of nine cases of a benign form of melioidosis. J.A.M.A. 202: 950-54.

Stanton, A. T., and Fletcher, W.1921. Melioidosis, a new disease of the Tropics. Tr. Fourth Cong. Far East Assoc. Trop Med 2: 196-98.

______. 1925. Melioidosis, a disease of rodents communicable to man. Lancet 1: 10-13.

______. 1932. Melioidosis: Studies from the Institute for Medical Research, Federated Malay States. Bull. No. 21. London: John Bale, Sons, & Danielson.

Stark, H. E. 1971. Some observations on fleas of small mammals associated with man in Vietnam and their relation to plague. Proc. Plague Symp., Saigon, Oct. 71.

Statistics, Medical Department of the United States Army in the World War. See MD-WW15.

Stead, W. W., and Bates, J. H. 1969. Primary tuberculosis from the Far East. Transmission by a veteran to two civilians. Ann. Int. Med 70: 707-11.

Strauss, J. M.; Alexander, A. D.; Rapmund, G.; Gan, E.; and Dorsey, A. E. 1969. Melioidosis in Malaysia. III. Antibodies to Pseudomonas pseudomallei in the human population. Am. J. Trop. Med. 18:703-7.

Strauss, J. M.; Groves, M. G.; Mariappan, M.; and Ellison, D. W. 1969. Melioidosis in Malaysia. II. Distribution of Pseudomonas pseudomallei in soil and surface water. Am. J. Trop. Med 18:698-702.

Strauss, J. M.; Jason, S.; and Mariappan, M. 1967. Pseudomonas pseudomallei in soil and surface water of Sabah, Malaysia. M.J. Malaya 22: 31-32.

Sutherland, J. D., and Dahlstrom, D. D. 1968. Melioidosis: Roentgen manifestations in a long bone. Radiology 91: 519.

Sweet, R. S.; Wilson, E. S., Jr.; and Chandler, B. F. 1968. Melioidosis manifested by cavitary lung disease. Am. J. RoentgenoL 103: 543-47.

Swellengrebel, N. H. 1967. Interhumane transmissie bij builenpest. NederL tijdschr. geneesk. 111: 647-48.

TB MED-Department of the Army. 1972. The diagnosis and management of tuberculosis. Technical Bulletin (Medical) 236, 3 Feb. 72.

20th Preventive Medicine Unit, USARV. See PMU-20.

USARV-CHR--USARV surgeon. Monthly Command Health Reports to USARV commander, Sept. 1969-Apr. 1970. On file at U.S. Army Center of Military History.

USARV-MB- USARV M. Bull, (USARV Pam 40-13), Jan.-Feb. 1969. Copy in Joint Medical Library, Office of the Surgeons General.

USARV Medical Bulletin. See USARV-MB.

USARV monthly Command Health Reports. See USARV-CHR. U.S. Operations Mission, Public Health Division. See OM-PHD. USPHS-U.S. Public Health Service Cooperative Investigation. 1964. Prevalence of drug resistance in previously treated patients. Am. Rev. Resp. Dis. 89: 327-36.

U.S. Public Health Service Cooperative Investigation. See USPHS.

Van Peenen, P. F. D.; Marshall, J. D., Jr.; Cavanaugh, D. C.; and Rust, J. H., Jr. 1970. Mammals of South Vietnam. II. Disease implications. Mil. Med. 135: 391-97.

Velimirovic, B. 1972. Plague in South-East Asia. A brief historical summary and present geographical distribution. Tr. Roy. Soc. Trop. Med. & Hyg. 66: 479-504.

Vivona, S.; Trinh-Thi Minh Ha; Gibson, F. L.; and Cavanaugh, D. C. 1966. Antibiotic sensitivities of Enterobacteriaceae isolated in Vietnam. Mil, Med. 131: 68-67.

Weber, D. R.; Douglass, L. E.; Brundage, W. G.; and Stallkamp, T. C. 1969. Acute varieties of melioidosis occurring in U.S. soldiers in Vietnam. Am. J. Med. 46: 234-44.

Whitmore, A. 1912. On the bacteriology of an infective disease occurring in Rangoon. Brit. M. J. 2: 1306-8.

Whitmore, A., and Krishnaswami, C. S. 1912. An account of the discovery of a hitherto undescribed infective disease occurring among the population of Rangoon. Indian M. Gaz. 47: 262-67.

WHO-a-World Health Organization. 1970. WHO Expert Committee on Plague, Fourth Report. Tech. Rep. s. No. 447. Geneva: World Health Organization.

WHO-b-World Health Organization. 1970. Insecticide resistance and vector control. Seventeenth report of the WHO Expert Committee on Insecticides. Tech. Rep. s. No. 443. Geneva: World Health Organization.

Winter, C. C.; Cherry, W. B.; and Moody, M. D. 1960. An unusual strain of Pasteurella pestis isolated from a fatal human case of plague. Bull, World Health Organ. 23: 408-9.

Winter, P. E.; Marshall, J. D., Jr.; Rust, J. H., Jr.; and Cavanaugh, D. C.1971. Plague. In Management and treatment of tropical diseases, ed. B. G. Maegraith and H. M. Gilles, pp. 377-87. Oxford: Blackwell Scientific Publications.

Wolford, R. A.1944. The tuberculosis program of the Veterans Administration. Am. Rev. Tuberc. 50: 380-90.

World Health Organization. See WHO.

Wu Lien-teh. 1911. Views of Harbin taken during the plague epidemic, December 1910-March 1911. Shanghai: Commercial Press.

______. 1926. A treatise on pneumonic plague, pp. 334-41. Geneva: League of Nations.

______. 1959. Plague fighter. The autobiography of a modern Chinese physician. Cambridge: W. Heffer & Sons.

Wu Lien-teh; Chun, J. W. H.; Pollitzer, R.; and Wu, C. Y. 1936. Plague. A manual for medical and public health workers. Shanghai: National Quarantine Service, Mercury Press.