Environmental Mycobacteria


Consultant Respiratory Physician
Aintree and Cardiothoracic Centre (NHS) Hospital Trusts
Director of the Tuberculosis Research Unit,
Cardiothoracic Centre, Thomas Drive, Liverpool L14 3PE.
Received January 1999


Key Words: Environmental mycobacteria Atypical mycobacteria Pulmonary disease Soft tissue infection Treatment


Environmental mycobacteria can only be distinguished from their more agressive cousin M.tuberculosis by bacteriological culture. Clinically the presentation of pulmonary disease is similar to tuberculosis with chronic cough weight loss and malaise. Soft tissue disease particularly of the lymph glands may occur in children. On microscopic smear and by the histological changes they cause, they are indistinguishable from the tubercle bacillus. Treatment of pulmonary disease is less effective than treatment for tuberculosis but Rifampicin and Ethambutol should be tried. Recent evidence suggests that Ciprofloxacin and Clarithromycin can also be effective.


What's in a name?

The first problem is what to call these bacteria collectively. The term Atypical Mycobacteria seems inappropriate as of all the Mycobacteria, M.tuberculosis with its Asian and African variants, and M.leprae, are the only ones which are obligative parasites. All the others can live free in the environment. Hence Environmental mycobacteria may be a better term. The Americans may win the day with "Mycobacteria other than tuberculosis" (MOTT), which is cumbersome, or "Non-tuberculous mycobacteria," which is inaccurate, as these bacteria form tubercles in the infected host identical to those produced by M.tuberculosis.(l)

Table 1 Major Environmental mycobacteria causing human disease.(2)


Species Sites of infection
M.avium-intracellulare Lungs
Lymph Glands
Disseminated (in HIV)
M.kansasii Lungs
M.xenopi Lungs
M.malmoense Lungs
M.fortuitum Soft tissues
M.chelonei Soft tissues
M.ulcerans Skin
M.marinum Soft tissues


M.avium-intracellulare (MAC) used to be the least common of the four major species causing lung disease. The advent of HIV, which for reasons unknown, seems to attract MAC infection specifically, has now made MAC the commonest in developed countries with an appreciable HIV infected population Among HIV negative individuals, M.kansasii is the commonest species found
in Western Europe and parts of the USA. It is also the commonest species in to commoner in the UK, although M.xenopi be South East M.malmoense the North. These bacteria are free and M.molmoense living; occuring in soil, house dust and tap water. They may infect wild or domestic animals as well as humans.

Bacteriology and Pathology

Environmental mycobacteria do not cause the disease we call tuberculosis. Infection with these organisms is not transmissable, therefore they do not have public health implications and diseases caused by them, unlike those caused by M.tuberculosis are not notifiable.

Histologically tissues infected by these mycobacteria appear identical to those infected by M.tuberculosos with characteristic granulomata formation leading to caseous necrosis.

On direct smear all mycobacteria stain positive to acid and alcohol fast Zehl-Neelsen and are therefore indistinguishable from one another. Only on culture do the species specific charteristics permit identification (Fig 1). The result is that patients, with pulmonary infections caused by environmental mycobacteria, are usually first diagnosed as having tuberculosis with the public health implications which that brings. Diagnosis is only revised when cultures are available some six weeks later, which may result in confusion of patients and doctors alike. Molecular techniques such as the Polymerase Chain Reaction (PCR) can provide more rapid species identification than culture and may be used more frequently in future by those countries able to afford it.


culture.jpg (8188 bytes)

Figure1: Culture of M.malmoense growing on a slope of Loenstein Gensen medium.

Clinical Presentation

Disease commonly occurs in the latter half of life among those with pre-exisiting lung disease such as previous tuberculosis, chronic bronchitis and emphysema, and less commonly bronchiectasis. in my experience infection with mycobacteria is strongly associated with smoking. A recent survey of over forty patients fom Liverpool showed only one to be a life long non-smoker.(3)

Patients may present with acute or chronic illness that is and radiographically clinically indistinguishable from tuberculosis. Symptoms are usually chronic and include cough with sputum and sometimes haemoptysis, weight loss, malaise and night sweats. Symptoms may be insidious and occur over many months before the patient presents. A recent study looking at the
symptomatic presentation of M.kansasii  infection compared with tuberculosis, found that with exception of diabetes and alcohol intake, which favoured a diagnosis of tuberculosis, there were no features between the two, likely to be diagnostically helpful.(4)

Some patients have disease presenting insidiously over more than a year. Weight loss and cachexia are often accompanied by long standing respiratory symptoms.  These are often attributed to the pre-existing pulmonary disease. Mycobacterial infection may be missed if it is not considered and the appropriate investigations requested.

Chest Radiography and Diagnosis

The chest radiograph may be difficult to interpret, often showing chronic abnormalites which may be attributed to the preexisting lung disease. (Fig 2). Genuine infection likely if several specimens show the same strain of mycobacteria from symptomatic patients whose chest x-ray shows abnormalities consistant with mycobacterial disease. Chest radiography usually shows upper lobe shadowing, which is more often unilateral than bilateral. Cavitation tends to be more frequent than in
tuberculosis and fibrosis more marked, but these features cannot be used to distinguish infection by one organism from the other.(5)

Figure 2.
Chest radiograph of a 55 year old, HIV negative male showing extensive left sided fibrosis. Sputum was heavily smear positive on Ziehl Neelsen stainng. MAC was grown from the sputum.

envmfig2.jpg (5649 bytes)

Sometimes the cavities can be very thick walled making penetration by drugs difficult (Fig. 3) A single isolate from sputum which is not repeated is unlikely to be of significance as these mycobacteria from the environment may easily contaminate specimens or cultures.

Figure 3.
CT scan showing a very thick walled cavity in the right lung. M avium intracellulare was grown from the sputum. The patient remained relatively well but smear positive despite 2 years of drug therapy

envmfig4.jpg (6751 bytes)


Disseminated Disease

Before the HIV epidemic disseminated infection with environmental mycobacteria was rare. MAC and M.kansasii have been described as causing disseminated disease in some patients. In the presence of HIV infection co-infection with MAC usually occurs at very low CD4 counts and dissemination is usual.

Patients may have pronounced immunodeficiency and complain of fever with weight loss and general malaise. Gastrointestinal symptoms usually predominate with pulmonary disease less marked. Enlarged lymph glands, hepatosplenomegaly and skin lesions are common. Chest x-ray may show hilar gland lymphadenopathy, pulmonary nodules or patchy alveolar

cavitation in disseminated disease is infiltrates. Pulmonary uncommon. The diagnosis may be made from tissue biosy particularly bone marrow, and even blood cultures. Infected tissue typically shows poorly formed granulomata teeming with acid fast bacilli. Serially performed blood cultures can be used as a way of monitoring the efficacy of treatment.

More commonly than blood born spread of disease to distant sites. environmental mycobacteria may infect soft tissue by puncture wounds of the skin, incuding sites of surgical incision. Soft tissue infections from swimming pools or fish tanks as result of M.balnei infections, may cause soft tissue inflammation and possible ulceration.


Treatment of pulmonary disease caused by environmental mycobacteria is much more problematic than the treatment of M.tuberculosis. In vitro sensitivity testing is thought to be a poor guiue to success. Environmental mycobacteria are usually resistant to all fist line anti-tuberculosis drugs with the exception of ethambutol. Uncontrolled trials have shown that treatment with rifampicin (R) and ethambutol (E) is effective in most patients infected with M.kansasii and M.xenopi. The length of treatment has been a subject of debate but a recent study showed a 9% relapse rate after as little as nine months of RE in the treatment of M.kansasii.(6)

The same agents have been said to be effective against the other common environmental pulmonary infectors, M.malmoense and MAC but a treatment length of two years is recommended. Recently the newer drugs, particularly clarithromycin and ciprofloxacin have been said to be effective but trial data is scarce.

Treatment is not always successful and the patient may remain sputum smear or culture positive despite months or years of treatment. It may be necessary to treat concomitant lung disease such as Chronic Bronchitis with broad spectrum antibiotics and bronchodilators. A number of different factors must be taken into account when assessing the efficacy of treatment. These will include general well being and symptomatology of the patient, weight gain, sputum conversion to a smear negative and culture negative state and radiological improvement. The last is probably the poorest guide to overall improvement as disease is often long standing and fibrosis permanent. In my experience cigarette smoking is strongly associated with environmental pulmonary infections and it is very difficult to eliminate bacteria from patients who continue to smoke. If the patient is relatively fit and the area of infection well contained, in an abscess cavity for example, removal of the infected area by surgery is an option. Fig 3 shows the chest radiograph of a 73 year old man who grew M.malmoense from the sputum. Two years of therapy eliminated the bacteria for a time but he relapsed with a return of symptoms and smear and culture presence of M.malmoense. Surgical excision of the right apical cavity effected a cure.

Sometimes the patient may be relatively asymptomatic, despite being heavily sputum smear positive for one of these organisms. In these cases a period of observation, while witholding treatment may be advisable bearing in mind the possible adverse effects of the treatment. In this respect M.malmoense and MAC in HIV serum negative individuals may have a very low virulence and may not be causing any additional pathology. M.xenopi, M.kansasii and the much rarer M.szulgei tend to be more virulent and will probably result in spread of disease if not actively treated. Other organisms such as M.gordonai or M.fortuitum, which have been isolated in pulmonary disease, are unlikely to cause active pathology.

Sensitivity testing often shows bacterial susceptibility to the older first line drugs such as ethionamide and cycloserine. The efficacy of these in vivo remains very difficult to quantify and the common adverse events they cause probably precludes their use.

Treatment of soft tissue infection is even more problematical. Drug treatment is rarely effective and surgical exicion is often the only treatment. In children with lympatic glands infected with these organisms total exision is often used. Alternatively they may be left untreated as they usually cause no systemic disease, and regress in time.



1. Davies PDO, Infection with Mycobacterium kansasii. (Editorial)
Thorax 1994;49:435-6.
2. Banks J. Environmental mycobacteria. in: Clinical Tuberculosis;
Ed Davies PDO. Chapman and Hall. London 1994 pp265-276.
3. Davies L, Davies PDO Non-tuberculosis mycobacterial infections: no fire without a smoke. Am J Resp Crit Care Med 1995;151: A7I5.
4. Evans SA, Colville A, Evans Al, Crisp Al, Johnston IDA. Pulmonary mycobacterium kansasii infection: comparison of the clinical features, treatment and outcome with pulmonary tuberculosis Thorax 1996;51:1248-1252.
5. Evans SA, Colville A, Evans AJ, Crisp Al, Johnston IDA. Pulmonary mycobacterium kansasii
infection: compariso of radiological features with pulmonary tuberculosis.
Thorax 1996;51:1243-1247.
6. Research Committee of the British Thoracic Society. Mycobacterium kansasii pulmonary
infection: a prospective study of the results of nine months of treatment with rifampicin and ethambutol. Thorax 1994;49:442-5.



Home • Journals • Search • Rules for Authors • Submit a Paper • Sponsor us   

Chest Med
Rules for Authors
Submit a Paper
Sponsor Us



Default text | Increase text size