© 2003 American Thoracic Society
Emergence of Nontuberculous Mycobacteria as Pathogens in Cystic FibrosisUniversity of Texas Health Center Tyler, Texas Establishing nontuberculous mycobacteria (NTM) as pathogens in any clinical situation has been difficult for as long as these organisms have been known (1). A recent example, which shares an important similarity to the two reports considered here (2, 3), was the recognition of NTM as significant pathogens in AIDS (4). In that instance, NTM infection occurred late in the course of a severe and progressive underlying disease and was further complicated by the presence of other more virulent infectious agents. Initially, many clinicians felt that NTM were a reflection of underlying immunosuppression and had little impact on the patients' symptoms or survival (5, 6). Olivier and coworkers (2, 3) were faced with similar obstacles when evaluating the impact of NTM for patients with another serious pre-existing disease, cystic fibrosis (CF). Two studies in this issue AJRCCM (pp. 828–834 and pp. 835–840) by Olivier and coworkers (2, 3) begin the process of defining the impact of NTM. In the first report (2), 13% of 986 CF patients had NTM isolated from sputum; Mycobacterium avium complex accounted for 72% of the isolates, and Mycobacterium abscessus accounted for 16%. Seventy percent of culture-positive patients had only one out of three sputum cultures positive for NTM. Compared with CF patients who had negative cultures, the patients with positive cultures were older and had better pulmonary function. In the second report (3), the authors evaluated clinical parameters from 60 patients with CF who had positive sputum cultures for NTM. Most (57%) had only one positive sputum culture for NTM. Specimens smear-positive for acid-fast bacilli were more common in patients with multiple positive cultures. The authors compared clinical variables, such as symptoms, pulmonary function tests, chest radiographs, and chest computed tomography scans in three groups of patients. The groups consisted of 99 control CF patients who did not have positive NTM cultures, 38 CF patients who had positive NTM cultures but who did not meet American Thoracic Society criteria for NTM lung disease, and 22 CF patients who had positive NTM cultures and who also met American Thoracic Society diagnostic criteria for NTM lung disease. There were no significant clinical differences between the three groups with the exception of findings on chest computed tomography. Two or more chest computed tomography abnormalities compatible with NTM disease as well as progressive abnormalities were more common in the patients with positive NTM cultures; these findings, however, sometimes also occurred in control patients. Computed tomography was not performed in all study participants; specifically, it was performed in only 8% of control patients at the exit from the study. The results from these studies are surprising, especially the apparent relatively benign effect of NTM on this population. Multiple questions stay unanswered. Important caveats need to be considered when interpreting the results of these studies. The authors considered all NTM isolated from CF patients as potential pathogens, but not all NTM are equally virulent. For instance, the Mycobacterium gordonae isolates were likely due to contamination of specimens (7). Inclusion of those patients may have skewed the results in a way favoring a lack of clinical difference between patients who had positive cultures and patients who had negative cultures. Conversely, focusing on more virulent pathogens, such as M. abscessus, could accentuate the clinical differences between culture-positive and culture-negative patients. Additionally, the CF patients in these studies almost certainly have a wide variation in severity of NTM disease and in the rate of progression. Some patients with mild or slowly progressive disease would be extremely difficult to detect during the short duration of these studies, especially against the backdrop of progressive CF. The studies do not provide clear guidance for diagnosing NTM lung disease in CF patients. The difficulty of diagnosing NTM disease in this population is illustrated by the observation that 11 patients who were originally designated control subjects developed positive NTM cultures during the course of the studies (3). Furthermore, other control patients with multiple negative NTM cultures but with radiographic abnormalities compatible with NTM disease might have had mild and slowly progressive NTM disease. Given a sufficiently long period of time, those patients would also likely become NTM culture positive. The American Thoracic Society diagnostic criteria, based on symptoms and radiologic and microbiologic findings, are probably not adequate in this setting (7). Symptom evaluation is difficult because all CF patients are symptomatic and NTM culture-positive patients were not more symptomatic than NTM culture-negative patients (3). Microbiologically, as in non-CF patients, single culture-positive, smear-negative sputum specimens are of uncertain clinical importance (3, 7). Of note, smear-positive specimens were frequently associated with multiple positive cultures, and thus, predicted patients who would meet American Thoracic Society microbiologic diagnostic criteria. The principal American Thoracic Society radiographic diagnostic criterion is the presence of an infiltrate not explained by other processes. The data from Olivier and coworkers suggest that the plain chest radiograph is of no diagnostic value (3). Computed tomography abnormalities were more helpful, although abnormalities compatible with NTM infection displayed overlap between patients with and patients without positive NTM cultures. It would have been helpful if there had been more extensive follow-up computed tomography data for NTM culture-negative patients. The comparison scans could help define the evolution and variability of abnormalities in NTM culture-negative patients and perhaps better characterize and accentuate the differences in abnormalities that develop over time in NTM culture-positive patients. Additional diagnostic criteria specifically applied to CF patients will probably be required. Consideration of symptoms, such as a fever that is not responsive to other treatment modalities, might be helpful. Assessing the virulence of the NTM species recovered is a reasonable initial step. For instance, M. abscessus isolates are probably less likely to be caused by contamination and more often associated with severe and/or progressive disease than M. avium complex (3). Documenting the progression of radiographic abnormalities may be as important as the presence of the abnormalities themselves. Positive acid-fast bacilli smears reflect a higher burden of organisms and serve as an indicator of significant disease (3). It is possible that quantitative cultures might be similarly useful. Finally, consideration of pulmonary function changes over time is probably helpful for individual patients even though the trend in pulmonary function in this study population did not differ between culture-positive and culture-negative patients (3). The role of bronchoscopy in the evaluation of these patients remains undefined. In this inexorably progressive disease, it is tempting to treat any pathogen that might worsen the prognosis of the CF patient. The data from these studies, however, do not offer clear guidance to identify which CF patients would benefit most from therapy. Although not explicitly described in these reports, there are subpopulations of CF patients who have positive sputum cultures for NTM, progressive infiltrates, and deteriorating pulmonary function, with no findings other than NTM infection to explain the clinical decline (8–10). These patients would probably benefit more from therapy directed against NTM than would clinically stable patients with single smear-negative, culture-positive specimens. The decision about which patients to treat for NTM infection is not made easier by the long and potentially toxic multidrug therapy for NTM (11). Therapy for M. abscessus disease is especially difficult with no therapeutic strategy of proven value (12). The recognition of NTM lung infection in CF patients is likely tied to improved life expectancy in CF patients and could accurately be described as an emerging disease. As life expectancy of CF patients continues to improve, the problem of NTM lung infection will almost certainly grow. Establishing which CF patients would benefit most from therapy of NTM disease is a high priority and should become easier with further separation of clinical findings caused by NTM infection and CF. The studies by Olivier and coworkers are an important start (2, 3). REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
