Nontuberculous mycobacterial (NTM) pulmonary infections continue to be a diagnostic and therapeutic challenge. There is a general consensus that clinical, radiographical and microbiological findings should be considered to be equally important in determining the diagnosis of a NTM infection. The American Thoracic Society (ATS) has set forth guidelines for the diagnosis, treatment and prevention for these infections. These guidelines identify several abnormalities on chest computed tomography (CT) as diagnostic criteria that may be associated with an active infection, including pulmonary nodules, cavities and bronchiectasis. Nevertheless, these criteria have been mainly described for infections caused by Mycobacterium avium complex, M. kansasii and M. abscessus. CT abnormalities and how they relate to symptomatology and microbiological findings in other less common species of NTM are less well known.
M. xenopi is an emerging pulmonary pathogen with significant morbidity and mortality in both normal and immunocompromised hosts.2,3 It has been isolated worldwide, including several countries in Europe and South America, Canada and the United States, and reports of this infection continue to surface in parts of Asia. The association of the CT abnormalities with symptoms and microbiological data is critical in making therapeutic decisions and predicting prognosis. There have been only a few studies that describe this correlation in patients with M. xenopi, but these involve small series with limited conclusions.2,4 Moreover, the positivity of one single culture has always raised the dilemma: does it represent colonization (generally regarded as not pathogenic) or is it a manifestation of an active infection? This predicament has to be resolved before a patient is committed to months of therapy that is not devoid of toxicity.
In this issue of Respirology, Marras and colleagues report the findings in a cohort of 70 immunocompetent patients from the province of Ontario, Canada with at least one or more respiratory isolates for M. xenopi and a chest CT within 6 months of the positive culture. The imaging studies were screened for all abnormalities rather than only for the criteria described by the ATS guidelines. Clinical data, when available, was used to further characterize the patients. Using this information, the patients were classified into three groups: (i) definite disease: fulfil ATS criteria (pulmonary symptoms and nodules, cavitary lesions or multifocal bronchiectasis on CT, and definitive microbiological confirmation); (ii) possible disease: fulfil ATS microbiological criteria (two or more positive cultures from expectorated sputum, or a positive culture from bronchoscopic sampling or a positive culture from a biopsy) and an abnormal CT as described earlier and unknown symptoms; and (iii) no disease: only one positive expectorated sputum for M. xenopi, or no symptoms or no relevant abnormalities on CT. In the present study, patients with tree-in-bud opacities and >10 small (<5 mm) or >5 large (>10 mm) nodules were also considered to have imaging consistent with active disease. Because the data were gathered retrospectively, in a minority of patients, it was not possible to document pulmonary symptoms.
The present study represents one of the largest series of patients with positive isolates of M. xenopi where CT imaging was available. CT imaging is significantly more sensitive compared with a chest radiograph and has become the standard in the evaluation of patients with suspected NTM infection. Additionally, the imaging was evaluated for abnormalities beyond those described in the ATS guidelines, providing a better understanding of their association with symptoms and microbiological data. The study revealed that cavitary disease, bronchiectasis, and tree-in-bud and large nodules were associated more frequently with definitive disease. Cavitary disease and large nodules were associated with fulfilment of the microbiological ATS criteria and positive acid-fast bacilli staining. These findings are in line with previous studies, where cavitary lesions were strongly associated with M. xenopi active disease.2–4,7,8 In all of these studies (including the present one), abnormalities tended to be more significant in the upper lobes. Contrary to other studies were pre-existing pulmonary conditions (e.g. chronic obstructive pulmonary disease, tuberculosis) were a common feature of patients with an active infection, the present study showed that prevalence of pre-existing lung diseases was similar among patients with definite disease compared with patients with no disease.2,7,9 This finding was unexpected and can be partially explained by the inclusion criteria, where patients were included based on a single positive culture of expectorated sputum in contrast with other studies where ATS criteria were fully met. This criterion was chosen to discern the significance of a single isolate of M. xenopi in association with symptomatology and radiographical findings. Additionally, in contrast with the present study, prior studies included immunosuppressed patients, where previous lung disease may further facilitate infections by this pathogen.
Studies that have evaluated the CT findings in other NTM pulmonary infections have shown that the presence of upper lobe cavitary lesions was a salient feature.10,11 Interestingly, not all of the NTM species have cavitary lesions as an important presenting feature; cavities have been reported in as few as 13% of patients fulfilling ATS criteria for NTM infection. Why some NTM infections present with cavitary lesions and others do not is not clear. The presence of cavities may also vary significantly even when evaluating a single species. For example, the incidence of cavities in M. avium complex infections can vary from 14% to 65%12,13 and in M. kansasi infections from 32% to 83%.14,15 The reasons for the radiographical ‘variability’, even within a single species of NTM, remain uncertain, and larger studies are needed to better define phenotypes or subgroups.
As with other NTM infections, for M. xenopi, there may still be more questions than answers. Even though prospective studies that evaluate serial imaging with an emphasis on frequent clinical and microbiological surveillance are needed, the present study sheds light on the association of the CT abnormalities with microbiological and clinical findings. In immunocompetent patients with M. xenopi isolation and cavitary lesions, especially if they are in the upper lobes, the diagnosis of NTM infection should be considered.