Clinical epidemiology and risk factors
Zoonotic transmission of B. mallei from solipeds to humans appears to be uncommon, even in cases of frequent and close contact with infected animals. However, rigorous examination regarding the rarity of transmission has not been conducted. Hypotheses to explain this low incidence include low organism concentrations at the infection sites or a high infectious dose required to cause disease in humans[9]. Despite the low incidence of animal to human transmission occupational exposure remains a key risk factor among veterinarians, veterinary students, farriers (hoof care workers), flayers (hide workers), transport workers, soldiers, slaughterhouse personnel, farmers, horse fanciers, and stable hands[9]. Subclinical infections in horses and mules also pose a hidden risk to humans, albeit a low risk. While infection by ingesting contaminated food and water has occurred, it does not appear to be a significant route of entry for human infections[18–20].
Human-to-human transmission is also rare. However, it may occur during occupational exposure in medical practice or at autopsies[3, 9, 19, 21]. Transmission has also occurred in home settings, where the care of glanders-infected individuals has led to the infection of other family members[21].
While laboratory workers have rarely been infected, their close contact with high concentrations of virulent organism puts them at a higher risk for infection. Eight cases of laboratory acquired glanders have been reported between 1943 and the present at US research centers, specifically Fort Detrick, Maryland. Seven of the eight Fort Detrick laboratory-acquired infections occurred prior to modern biological containment practices[22]. The first six cases occurred during the performance of normal laboratory procedures that could generate aerosols such as washing, centrifugation, and aerating B. mallei cultures. In contrast to zoonotic transmission, aerosols of B. mallei are highly infectious[22]. Unfortunately, the seventh case is not well described in published literature. The eighth case may have resulted from inconsistent use of personal protective equipment particularly latex gloves[3, 9, 23].
Routes of infection
Glanders is transmitted by direct invasion of abraded or lacerated skin; inhalation with deep lung deposition; and by bacterial invasion of the nasal, oral, and conjunctival mucous membranes. The occupational exposures described above most often occur through exposed skin, particularly the hands, arms, neck, and face[9]. B. mallei is not believed to penetrate normal intact skin, although wounds or penetrations during the likely exposure interval were not identified in many cases[9]. Indeed, most laboratory-acquired infections are not associated with injury or a recollection of injury[24]. In the case of the eighth case mentioned above, a break in the skin or a specific exposure-associated laboratory incident (needle stick or broken glassware) was not recalled or identified. However, this patient reported collecting a personal blood sample via a finger-stick for diabetic monitoring prior to entering the laboratory[9]. In light of the lack of glove use, the finger-stick site may have been the potential entry point of the bacteria. The clinical manifestation of unilateral axillary lymphadenopathy in this patient was consistent with percutaneous infection.
Six of the eight cases above were described in Howe and Miller’s report which detailed glanders cases occurring within one year among Fort Detrick laboratory personnel working with B. mallei (referred to as Malleomyces mallei in the report), including two notable cases[22]. Two workers (patients one and two) were present when a flask containing B. mallei was dropped and broken thus potentially creating aerosol droplets. Both patients were admitted to the hospital on the same day, approximately two weeks after the incident and were the only personnel present in the lab when the flask was dropped. The other four patients (patients three to six) were actively engaged in washing B. mallei growth from agar plates for the preparation of vaccines approximately 10–14 days prior to the onset of their symptoms. While all workers wore the required protective clothing at the time and used caution while performing their duties, the possibility that they inhaled an aerosol could not be eliminated. Patient four also failed to heat kill the B. mallei sample they were working with prior to making dilutions for turbidity measurements two weeks before symptom onset. Approximately two weeks before their admission to the hospital, patients five and six had engaged in procedures involving the aeration of cultures. They recalled that on at least two occasions, the containers had been opened immediately after the air current had been turned off, rather than after a period of delay long enough to avoid the escape of aerosolized organisms into the room. Additionally, these two patients were working with a strain of organism of higher virulence than the other four. These events strongly suggest aerosol route of infection by B. mallei can result in human glanders cases, especially in the laboratory setting[22].
Time between exposure and symptoms
The acute form of disease has a typical incubation period of 1–14 days[25], while the chronic form of the disease has an incubation period of up to 12 weeks. A localized infection typically follows within one to five days of exposure and may be characterized by swelling of the affected area and a weeping discharge. Acute pulmonary infections may require anywhere from 10–14 days of incubation before symptoms appear[25]. Septicemia may develop immediately after exposure or up to two weeks after initial infection. Pneumonic disease usually has a rapid onset and is almost uniformly lethal between 10 and 30 days in untreated cases[22]. Figure 1 summarizes the time between exposure and symptoms. An important feature of the eight cases that have occurred since 1943 is that at least half of the patients not only felt better but also experienced improved clinical signs for a period of time after the first wave of symptoms and prior to a second wave of symptoms. This period of temporary improvement has the potential to be misinterpreted as eradication of disease by patient and physician. Such temporary improvement should not limit recommended treatments.
Diagnostics
The definitive diagnosis of glanders requires isolation of the organism and positive identification. Radiology may reveal abcesses in multiple organs including lungs, liver, and spleen, but these abcesses are indistinguishable from those caused by other disease and specific diagnosis is required[26]. While there is no validated in vitro diagnostic test for glanders, experimental serological tests, such as agglutination and complement fixation tests[27–29], and PCR based tests[30] have been used on a limited and experimental basis in solipeds as an alternative to isolating the organism. An indirect hemagglutination assay (IHA) used to diagnose melioidosis in endemic regions can also be used to diagnosis glanders[31]. Briefly, the IHA uses Burkholderia antigens to coat sheep red blood cells prior to their incubation with serum from suspected melioidosis (or glanders) patients. In the resulting reaction the antigen-coated sheep RBCs agglutinate or form a pellet if the patient is seropositive for B. pseudomallei or B. mallei (i.e., possesses antibodies to these pathogens)[31, 32]. Since antibodies from glanders and melioidosis patients cross react with antigens present on a variety of Burkholderia species, the IHA assay is not a specific diagnostic test and other glanders-specific diagnostic tests are needed[31, 33]. The mallein test is used for veterinary diagnostic purposes and involves injecting purified protein derivatives intradermally to observe a delayed type hypersensitivity reaction, similar to the tuberculin test for tuberculosis[34]. However, this test is not used in human diagnostics.
Features of human clinical glanders
General symptoms
Many forms of glanders have been described, including chronic, disseminated, pulmonary, and septicemic. The variety of infections is largely explained by various routes of infection. Localized infections are generally regionally confined and typically characterized by foci of suppuration. The abscesses can ulcerate and drain for long periods of time. However, localized infections may disseminate, leading to pulmonary, septicemic, or multi-tissue infection[9]. The most common clinical features of the eight laboratory-acquired infections from Fort Detrick included (in order of most common occurrence) afternoon to evening low-grade fever, malaise, fatigue, headache, myalgias including backache, lymphadenopathy, and chest pain. Approximately half of the patients not only felt better but were also clinically better for a period of time after the first wave of disease symptoms. This period lasted from a few days to 2 months then patients developed clinical signs of infection.
Mucosal involvement
Involvement of the eye and conjunctiva in a B. mallei infection presents with excessive lacrimation and photophobia. Nasal involvement is characterized by inflammation and swelling of the nose, which is common following inhalation of B. mallei. This may be followed by copious nasal discharge. Additionally, infection may invade the nasal septum and bony tissues, causing fistulae and tissue destruction[9]. The face may swell and regional lymph nodes may become inflamed. Infection may also extend lower in the respiratory tract, resulting in bronchitis which can be accompanied by cough and mucopurulent sputum production. Constitutional signs and symptoms such as fever and chills typically occur within the first few days following infection. Additionally, these symptoms may persist through treatment and be severe. Common signs and symptoms can include, but are not limited to, fever or low-grade fever in the afternoon to evening; chills with or without rigors; and severe headache[9].
Cutaneous involvement
Cutaneous manifestations include papular lesions that may erupt anywhere on the body with a more chronic, indolent course of infection. B. mallei entry through an abrasion is typically followed by an inflammatory response, including pain and swelling. In these cases, a glanders node may appear as a single blister, gradually developing into an ulcer that may become hemorrhagic[18, 35]. A localized infection with a discharge typically develops at the entry site. Inflammation may extend along regional lymphatics and cause lymphangitis with numerous foci of suppuration along their course. The endotoxins present in B. mallei strains affect the smooth muscle of the lymphatics[9] by enhancing the irritation and inflammation seen in the lymphatics.
Pulmonary involvement
A pulmonary infection typically results in pneumonia, pulmonary abscess, pleuritis, and plural effusion. Signs and symptoms of pulmonary infection can include cough, dyspnea, chest pain, and mucoplurent sputum. Nonspecific signs and symptoms such as fatigue, fever (often exceeding 102°F), chills, headache, myalgias, lymphangitis, sore throat, pleuritic chest pain, cough, tachypnea, dyspnea, discharge, and gastrointestinal signs often accompany respiratory infections[9]. Many symptoms may take up to 2 to 3 weeks to develop. Nonspecific signs, such as dizziness, rigors, myalgia, nausea, night sweats, severe headache, tachycardia, weight loss, and mucosal eruptions are also usually present and may indicate a disseminated infection[9].
Dissemination of infection
Dissemination from local cutaneous or mucosal infection result in septicemia and the colonization of internal organs such as the spleen, liver, and lungs with the development of abscesses[9]. Ultrasonography and computed tomography may reveal multiple, small discrete abscesses in both the liver and the kidney[23]. These infections are typically associated with septic shock and high mortality.