Melioidosis

Melioidosis is an infectious disease caused by a Gram-negative bacterium, Burkholderia pseudomallei, found in soil and water. It is of public health importance in endemic areas, particularly in Thailand and northern Australia. It exists in acute and chronic forms. Signs and symptoms may include pain in chest, bones, or joints; cough; skin infections, lung nodules and pneumonia.

B. pseudomallei was previously classed as part of the Pseudomonas genus and until 1992, it was known as Pseudomonas pseudomallei. It is phylogenetically related closely to Burkholderia mallei which causes glanders, an infection primarily of horses, donkeys, and mules. The name melioidosis is derived from the Greek melis (μηλις) meaning "a distemper of asses" with the suffixes -oid meaning "similar to" and -osis meaning "a condition", that is, a condition similar to glanders.

Signs and symptoms

Acute melioidosis

In the subgroup of patients where an inoculating event was noted, the mean incubation period of acute melioidosis was 9 days (range 1â€"21 days). Patients with latent melioidosis may be symptom-free for decades; the longest period between presumed exposure and clinical presentation is 62 years. The potential for prolonged incubation was recognized in US servicemen involved in the Vietnam War, and was referred to as the "Vietnam time-bomb". A wide spectrum of severity exists; in chronic presentations, symptoms may last months, but fulminant infection, particularly associated with near-drowning, may present with severe symptoms over hours.

A patient with active melioidosis usually presents with fever. Pain or other symptoms may be suggestive of a clinical focus, which is found in around 75% of patients. Such symptoms include cough or pleuritic chest pain suggestive of pneumonia, bone or joint pain suggestive of osteomyelitis or septic arthritis, or cellulitis. Intra-abdominal infection (including liver and/or splenic abscesses, or prostatic abscesses) do not usually present with focal pain, and imaging of these organs using ultrasound or CT should be performed routinely. In one series of 214 patients, 27.6% had abscesses in the liver or spleen (95% confidence interval, 22.0% to 33.9%). B. pseudomallei abscesses may have a characteristic "honeycomb" or "swiss cheese" architecture (hypoechoic, multiseptate, multiloculate) on CT.

Regional variations in disease presentation are seen: parotid abscesses characteristically occur in Thai children, but this presentation has only been described once in Australia. Conversely, prostatic abscesses are found in up to 20% of Australian males, but are rarely described elsewhere. An encephalomyelitis syndrome is recognised in northern Australia.

Patients with melioidosis usually have risk factors for disease, such as diabetes, thalassemia, hazardous alcohol use, or renal disease, and frequently give a history of occupational or recreational exposure to mud or pooled surface water. However, otherwise healthy patients, including children, may also get melioidosis.

In up to 25% of patients, no focus of infection is found and the diagnosis is usually made on blood cultures or throat swab. Melioidosis is said to be able to affect any organ in the body except the heart valves (endocarditis). Although meningitis has been described secondary to ruptured brain abscesses, primary meningitis has not been described. Less common manifestations include intravascular infection, lymph node abscesses (1.2â€"2.2%), pyopericardium and myocarditis, mediastinal infection, and thyroid and scrotal abscesses and ocular infection.

Chronic melioidosis

Chronic melioidosis is usually defined by a duration of symptoms greater than two months and occurs in about 10% of patients. The clinical presentation of chronic melioidosis is protean and includes such presentations as chronic skin infections, chronic lung nodule, and pneumonia. In particular, chronic melioidosis closely mimics tuberculosis, and has sometimes been called "Vietnamese tuberculosis".

Diagnosis

A definitive diagnosis is made by culturing the organism from any clinical sample, because the organism is never part of the normal human flora.

A definite history of contact with soil may not be elicited, as melioidosis can be dormant for many years before manifesting. Attention should be paid to a history of travel to endemic areas in returned travellers. Some authors recommend considering possibility of melioidosis in every febrile patient with a history of traveling to and/or staying at endemic areas.

A complete screen (blood culture, sputum culture, urine culture, throat swab, and culture of any aspirated pus) should be performed on all patients with suspected melioidosis (culture on blood agar as well as Ashdown's medium). A definitive diagnosis is made by growing B. pseudomallei from any site. A throat swab is not sensitive, but is 100% specific if positive, and compares favourably with sputum culture. The sensitivity of urine culture is increased if a centrifuged specimen is cultured, and any bacterial growth should be reported (not just growth above 10⁴ organisms/ml which is the usual cutoff). Very occasionally, bone marrow culture may be positive in patients who have negative blood cultures for B. pseudomallei, but these are not usually recommended. A common error made by clinicians unfamiliar with melioidosis is to only send a specimen from the affected site (which is the usual procedure for most other infections) instead of sending a full screen.

Ashdown's medium, a selective medium containing gentamicin, may be required for cultures taken from nonsterile sites. Burkholderia cepacia medium may be a useful alternative selective medium in nonendemic areas, where Ashdown's is not available. A new medium derived from Ashdown, known as Francis medium, may help differentiate B. pseudomallei from B. cepacia and may help in the early diagnosis of melioidosis, but has not yet been extensively clinically validated.

Many commercial kits for identifying bacteria may misidentify B. pseudomallei (see Burkholderia pseudomallei for a more detailed discussion of these issues).

Also, a serological test for melioidosis (indirect haemagglutination) is available, but not commercially in most countries. A high background titre may reduce the positive predictive value of serological tests in endemic countries. A specific direct immunofluorescent test and latex agglutination, based on monoclonal antibodies, are used widely in Thailand, but are not available elsewhere. Cross-reactivity with B. thailandensis is almost complete. A commercial ELISA kit for melioidosis appears to perform well. but no ELISA test has yet been clinically validated as a diagnostic tool.

It is not possible to make the diagnosis on imaging studies alone (X-rays and scans), but imaging is routinely performed to assess the full extent of disease. Imaging of the abdomen using CT scans or ultrasound is recommended routinely, as abscesses may not be clinically apparent and may coexist with disease elsewhere. Australian authorities suggest imaging of the prostate specifically due to the high incidence of prostatic abscesses in northern Australian patients. A chest X-ray is also considered routine, with other investigations as clinically indicated. The presence of honeycomb abscesses in the liver are considered characteristic, but are not diagnostic.

The differential diagnosis is extensive; melioidosis may mimic many other infections, including tuberculosis.

Treatment

Current treatment

The treatment of melioidosis is divided into two stages, an intravenous high-intensity phase and an eradication phase to prevent recurrence.

Intravenous intensive phase

Intravenous ceftazidime is the current drug of choice for treatment of acute melioidosis. Meropenem, imipenem and the cefoperazone-sulbactam combination (Sulperazone) are also active. Intravenous amoxicillin-clavulanate (co-amoxiclav) may be used if none of the above four drugs is available, but it produces inferior outcomes. Intravenous antibiotics are given for a minimum of 10 to 14 days, and are not usually stopped until the patient's temperature has returned to normal for more than 48 hours. Even with appropriate antibiotic therapy, fevers often persist for weeks or months, and patients may continue to develop new lesions even while on appropriate treatment. The median fever clearance time in melioidosis is 10 days: and failure of the fever to clear is not a reason to alter treatment. It is not uncommon for patients to require parenteral treatment continuously for a month or more.

Intravenous meropenem is routinely used in Australia; outcomes appear to be good and meropenem is currently being tested with ceftazidime in a Thai clinical trial.

Theoretical reasons are given for believing mortality might be lower in patients treated with imipenem: first, less endotoxin is released by dying bacteria during imimipenem treatment, and the minimum inhibitory concentration (MIC) for imipenem is lower than for ceftazidime. However, no clinically relevant difference was found in mortality between imipenem and ceftazidime treatments. The MIC of meropenem is higher for B. pseudomallei than for many other organisms, and patients being haemofiltered will need more frequent or higher doses.

Moxifloxacin, cefepime, tigecycline, and ertapenem do not appear to be effective in vitro. Piperacillin-sulbactam, doripenem and biapenem appear to be effective in vitro, but no clinical experience exists on which to recommend their use.

Adjunctive treatment with granulocyte colony-stimulating factor or co-trimoxazole were not associated with decreased fatality rates in trials in Thailand.

Eradication phase

Following the treatment of the acute disease, eradication (or maintenance) treatment with co-trimoxazole and doxycycline is recommended to be used for 12 to 20 weeks to reduce the rate of recurrence. Chloramphenicol is no longer routinely recommended for this purpose. Co-amoxiclav is an alternative for those patients who are unable to take co-trimoxazole and doxycycline (e.g., pregnant women and children under the age of 12), but is not as effective. Single agent treatment with a fluoroquinolone (e.g., ciprofloxacin) or doxycycline for the oral maintenance phase is ineffective.

In Australia, co-trimoxazole is used on its own for eradication therapy, with relapse rates that are lower than those seen in Thailand; in vitro evidence also suggests co-trimoxazole and doxycycline are antagonistic, and co-trimoxazole on its own may be preferable. A randomised controlled trial (MERTH) to compare this with the current standard of co-trimoxazole and doxycycline started in 2006 and was due for completion in 2008. Studies reinforce the need for adequate follow-up and good adherence to the eradication phase of therapy. Dosing for co-trimoxazole is based on weight: (<40 kg: 160/800 mg every 12 hours; 40â€"60kg: 240/1200 mg every 12 hours, >60 kg: 320/1600 mg every 12 hours).

Surgical treatment

Surgical drainage is usually indicated for prostatic abscesses and septic arthritis, may be indicated for parotid abscesses, and is not usually indicated for hepatosplenic abscesses. In bacteraemic melioidosis unresponsive to intravenous antibiotic therapy, splenectomy has been attempted, but only anecdotal evidence supports this practice.

Historical treatment

Prior to 1989, the standard treatment for acute melioidosis was a three-drug combination of chloramphenicol, co-trimoxazole and doxycycline; this regimen is associated with a mortality rate of 80% and is no longer be used unless no other alternatives are available. All four drugs are bacteriostatic (they stop the bacterium from growing, but do not kill it) and the action of co-trimoxazole antagonizes both chloramphenicol and doxycycline.

Prognosis

Without access to appropriate antibiotics (principally ceftazidime or meropenem), the septicemic form of melioidosis exceeds 90% in mortality rate. With appropriate antibiotics, the mortality rate is about 10% for uncomplicated cases but up to 80% for cases with bacteraemia or severe sepsis. It seems certain that access to intensive care facilities is also important, and probably at least partially explains why total mortality is 20% in Northern Australia but 40% in Northeast Thailand. Response to appropriate antibiotic treatment is slow, with the average duration of fever following treatment being 5â€"9 days.

Recurrence occurs in 10 to 20% of patients, but with co-trimoxazole eradication therapy, this can be reduced to 4%. While molecular studies have established the mazjority of recurrences are due to the original infecting strain, a significant proportion of recurrences (perhaps up to a quarter) in endemic areas may be due to reinfection, particularly after two years. Risk factors include severity of disease (patients with positive blood cultures or multifocal disease have a higher risk of relapse), choice of antibiotic for eradication therapy (doxycycline monotherapy and fluoroquinolone therapy are not as effective), poor compliance with eradication therapy and duration of eradication therapy less than 8 weeks.

Prevention

Person-to-person transmission is exceedingly unusual; and patients with melioidosis should not be considered contagious. Lab workers should handle B. pseudomallei under BSL-3 isolation conditions, as laboratory-acquired melioidosis has been described.

In endemic areas, people (rice-paddy farmers in particular) are warned to avoid contact with soil, mud, and surface water where possible. Case clusters have been described following flooding and cyclones and probably relate to exposure. Other case clusters have related to contamination of drinking water supplies. Populations at risk include patients with diabetes mellitus, chronic renal failure, chronic lung disease, or an immune deficiency of any kind. The effectiveness of measures to reduce exposure to the causative organism have not been established. A vaccine is not yet available.

Postexposure prophylaxis

After exposure to B. pseudomallei (particularly following a laboratory accident) combined treatment with co-trimoxazole and doxycycline is recommended. Trovafloxacin and grepafloxacin have been shown to be effective in animal models.

Vaccination

No vaccines are licensed for the prevention of melioidosis.

Biological warfare potential

Interest in melioidosis has been expressed because it has the potential to be developed as a biological weapon. It is classed by the US Centers for Disease Control (CDC) as a category B agent. B. pseudomallei, like B. mallei which causes glanders, was studied by the U.S. as a potential biological warfare agent, but was never weaponized. The Soviet Union was reported to have also experimented with B. pseudomallei as a biological warfare agent.

Epidemiology

Melioidosis is endemic in parts of southeast Asia (including Thailand, Laos, Singapore, Brunei, Malaysia, Burma and Vietnam), China, Taiwan and northern Australia. Multiple cases have also been described in Hong Kong and Brunei India, and sporadic cases in Central and South America, the Middle East, the Pacific and several African countries. Although only one case of melioidosis has ever been reported in Bangladesh, at least five cases have been imported to the UK from that country. Recent news reports indicate B. pseudomallei has been isolated from soil in Bangladesh, but this remains to be verified scientifically. This suggests melioidosis is endemic to Bangladesh and a problem of underdiagnosis or under-reporting exists there. most likely due to a lack of adequate laboratory facilities in affected rural areas.

Northeast Thailand has the highest incidence of melioidosis recorded in the world (21.3 cases of melioidosis per 100,000 people per year). In Northeast Thailand, 80% of children are positive for antibodies against B. pseudomallei by the age of 4; the figures are lower in other parts of the world.

Melioidosis is a recognised disease in animals, including cats, goats, sheep, and horses. Cattle, water buffalo, and crocodiles are considered to be relatively resistant to melioidosis despite their constant exposure to mud. An outbreak at the Paris Zoo in the 1970s ("L’affaire du jardin des plantes") was thought to have originated from an imported panda.

B. pseudomallei is normally found in soil and surface water; a history of contact with soil or surface water is, therefore, almost invariable in patients with melioidosis; that said, the majority of patients who do have contact with infected soil suffer no ill effects. Even within an area, the distribution of B. pseudomallei within the soil can be extremely patchy, and competition with other Burkholderia species has been suggested as a possible reason. Contaminated ground water was implicated in one outbreak in northern Australia. Also implicated are severe weather events such as flooding tsunamis and typhoons.

Based on whole genome sequencing, humans may play a role in moving B. pseudomallei from place to place.

The single most important risk factor for developing severe melioidosis is diabetes mellitus. Other risk factors include thalassaemia, kidney disease, occupation (rice paddy farmers), and cystic fibrosis. The mode of infection is believed to be either through a break in the skin, or through the inhalation of aerosolized B. pseudomallei cells. Person-to-person spread has been described, but is extremely unusual. HIV infection does not predispose to melioidosis.

The disease is clearly associated with increased rainfall, with the number (and severity) of cases increasing following increased precipitation.

Synonyms

• Pseudoglanders
• Whitmore's disease (after Captain Alfred Whitmore, who first described the disease)
• Nightcliff gardener's disease (Nightcliff is a suburb of Darwin, Australia where melioidosis is endemic)
• Paddy-field disease
• Morphia injector's septicaemia