Opportunist Mycobacteria
Total Page:16
File Type:pdf, Size:1020Kb
Thorax: first published as 10.1136/thx.44.6.449 on 1 June 1989. Downloaded from Thorax 1989;44:449454 Editorial Treatment of pulmonary disease caused by opportunist mycobacteria During the early 1950s it was recognised that recommended that surgical treatment for opportunist mycobacteria other than Mycobacterium tuberculosis mycobacterial infection should be given to those could cause pulmonary disease in man.' Over 30 years patients who are suitable surgical candidates.'7 The later there is no general agreement about the treatment failure of chemotherapy was often attributed to drug of patients with these mycobacterial infections. The resistance,'2 18 but the importance of prolonging the greatest controversy surrounds the treatment of in- duration of chemotherapy in opportunist mycobac- fection caused by the M aviwn-intracellulare- terial disease beyond that which would normally be scrofulaceum complex (MAIS), for which various required in tuberculosis was not appreciated. In treatments have been advocated, including chemo- several surgically treated series preoperative chemo- therapy with three23 or more' drugs or, alternatively, therapy was given on average for only four to seven surgical resection of the affected lung.78 Although the months,'3 19-22 some patients receiving as little as eight treatment of infection caused by M kansasii is less weeks of treatment before surgery.2324 In contrast, controversial there is no uniform approach to treat- chemotherapy alone, with isoniazid, p-aminosalicyclic ment. Disease caused by M xenopi has been described acid, and streptomycin for 24 months, produced by some as easy to treat with chemotherapy,9 whereas successful results in 80-100% of patients with M others have found the response to drug treatment to be kansasii infection despite reports of in vitro resistance unpredictable.'' " to these agents.2125 This diversity ofopinion and approach to treatment has arisen for two reasons. Firstly, there have been no In vitro drug sensitivity testing in opportunist mycobac- http://thorax.bmj.com/ large prospective studies designed to compare treat- terial disease ment regimens. Secondly, the failure to consider infection caused by M kansasii, the MAIS complex, Many studies show that the results of in vitro drug and other opportunist mycobacteria as entities distinct sensitivity tests performed against opportunist myco- from infection caused by M tuberculosis has led to bacteria bear little relation to the clinical response ob- inappropriate comparisons with tuberculosis. As a served when these drugs are used in treatment.2310 25-30 result principles that govern the treatment of tuber- The results of treatment with agents such as cyclo- culosis, in particular the role ofin vitro drug sensitivity and which show in serine ethionamide, often good on September 27, 2021 by guest. Protected copyright. tests, and criteria by which response to treatment is vitro action, has been disappointing2 '° and may reflect assessed have been empirically extended to disease drug toxicity and poor compliance by patients.23' caused by other mycobacterial species. It was realised Paradoxically, successful treatment has been achieved in early reports that patients with infection caused by with combinations of standard antimycobacterial M kansasii often remained sputum culture positive agents, despite reported in vitro resistance, though the and showed persistent cavities on their chest duration oftreatment has had to be prolonged in these radiographs after a few months' treatment with com- circumstances.2313132 Good in vitro sensitivity to first binations of isoniazid, para-aminosalicyclic acid line agents has allowed the duration ofchemotherapy (PAS) and streptomycin.'2"16 The results were poor by to be shortened-for example, short course chemo- comparison with the prompt bacteriological and therapy for infection caused by M kansasii is possible radiographic responses in patients with tuberculosis with regimens containing rifampicin.333 given the same treatment and raised doubts about the Critical levels ofin vitro resistance that have clinical use of chemotherapy in opportunist mycobacterial importance have been defined for M tuberculosis?' but disease.'2-"6 In 1963 the American Thoracic Society not for other mycobacterial species. Nevertheless, isolates of opportunist mycobacteria are reported as "resistant" if the minimum inhibitory concentrations Address for reprint requests: Dr J Banks, Department of Chest ofdrugs that suppress their growth in vitro exceed the Medicine, Liandough Hospital, Penarth, S Glam CF6 lXX. minimum inhibitory concentrations for drug sensitive 449 Thorax: first published as 10.1136/thx.44.6.449 on 1 June 1989. Downloaded from 450 strains of M tuberculosis.3" The comparison with MAIS COMPLEX Mtuberculosis makes the assumption that critical There is no consensus about the best treatment for levels of in vitro resistance that predict treatment pulmonary infection caused by the MAIS complex. failure in M tuberculosis infection are the same for all Surgical resection of affected lung is advocated by mycobacteria, irrespective of species. Clearly this is some,78 and others have recommended chemotherapy not so. Isolates of mycobacteria belonging to the with standard antimycobacterial drugs.23 The results MAIS complex or of M xenopi or M malmoense following surgical treatment have undoubtedly been should not therefore be classified as resistant until in good, conversion to sputum culture negative being vitro criteria that correlate with the clinical results of achieved postoperatively in 93-100% ofcases in some treatment of these infections have been established. series, and relapses occurring in only 5% of patients Drug resistance cannot be defined simply by relating during prolonged follow up.78 These results are more in vitro minimum inhibitory concentrations to serum impressive than those obtained with chemotherapy drug concentrations as the latter do not necessarily alone,'6424 but comparisons of this nature do not reflect drug concentrations achieved within tissues and compare like with like. Patients included in surgically macrophages. For example, concentrations of etham- treated series are often selected in so far as they have butol in both normal and caseous lung tissue are less extensive disease and have sufficiently good car- several times higher than the concentrations in plas- diopulmonary reserve to withstand thoracotomy and ma,38 and even higher drug concentrations are pulmonary resection.78 Even in centres where surgery achieved within alveolar macrophages.39 4 In addition, is strongly advocated they have comprised no more concentrations of ethambutol needed to kill phago- than 30-40% of all patients treated.8 In addition, cytosed bacilli are lower than the bactericidal concen- many have received adjunctive chemotherapy, which trations required in culture medium.4' Such factors in some cases has been continued for two or three may account for the effectiveness of some drugs in years.78 In contrast, patients treated with chemo- treatment despite their poor in vitro action. therapy alone have often had other serious pulmonary conditions, which in some cases were considered to be more important prognostically than the mycobacterial Recommendations for treatment infection itself.45 Interestingly, 94% of patients in one series who had no coexistent lung disorder, and who Pulmonary disease most often results from infection were not breathless on admission, converted to with M kansasii, the MAIS complex, M xenopi, or M sputum culture negative and showed radiographic http://thorax.bmj.com/ malmoense. improvement in response to chemotherapy alone.3 The median follow up was 22 months, during which time M KANSASII 6% of patients relapsed. These initial results are as Infection with M kansasii should be treated with good as those obtained with surgery and, although regimens that include both rifampicin and etham- follow up was relatively short, are far better than the butol. The combination ofrifampicin and ethambutol results reported in many other medically treated series. alone produced successful results in one series.3 The Surgical treatment, even in selected patients, has not preliminary results of a prospective multicentre study always been successful. In one series 33% of patients conducted in the United Kingdom by the British ultimately relapsed after having surgery45; in another on September 27, 2021 by guest. Protected copyright. Thoracic Society suggests that nine months' treatment series complications, including bronchopleural fistula, with regimens that include both rifampicin and etham- developed in 24% of patients and there was a 7% butol will be adequate.35 Of 75 patients who had postoperative mortality.7 completed treatment by March 1986, 14 had received Medical treatment is the only therapeutic option rifampicin combined with ethambutol and 61 had available for many patients. Pessimistic reports on the been given rifampicin and ethambutol combined with results of chemotherapy have often failed to indicate one or more additional drugs, usually isoniazid. These the duration of treatment given before it was con- additional drugs were given for less than three months cluded to be unsuccessful.'628424 Treatment with at the start of treatment in most patients. One failure rifampicin and isoniazid combined with either etham- of treatment resulted from non-compliance with drug butol or streptomycin was successful in 84% of cases treatment and two deaths occurred from unrelated