Lung and Nodal Involvement in Nontuberculous Mycobacterial Disease: PET/CT Role

Introduction. Systematic use of 18F-FDG PET/CT has the potential to simultaneously assess both pulmonary and lymph node involvement in nontuberculous mycobacterial (NTM) lung infection. Objective. The aim of the study was to evaluate the role of 18F-FDG PET/CT in the assessment of both mediastinal lymph nodes and lung involvement in NTM patients compared with active tuberculosis (TB) patients. Methods. 26 patients with pulmonary NTM disease were selected; six consecutive patients had undergone 18F-FDG PET/CT and data was compared with 6 active TB patients. Results. NTM exhibited different radiological lung patterns with an average SUV max value at PET/CT scan of 3,59 ± 2,32 (range 1,14 to 9,01) on pulmonary lesions and a mean value of SUV max 1,21 ± 0,29 (range 0,90 to 1,70) on mediastinal lymph nodes. Pulmonary lesions in TB showed an average SUV max value of 10,07 ± 6,45 (range 1,20 to 22,75) whilst involved mediastinal lymph nodes exhibited a mean SUV max value of 7,23 ± 3,03 (range 1,78 to 15,72). Conclusions. The differences in PET uptake in a broad range of lung lesions and lymph nodes between NTM and M. tuberculosis patients suggest a potential role for PET/CT scan in the diagnosis and management of pulmonary mycobacterial disease.


Introduction
Epidemiological observations indicate an increase in the prevalence of NTM disease worldwide. NTM infection is caused by a group of versatile opportunistic bacterial pathogens and remains underdiagnosed due to difficulties in pathogen isolation and nonspecificity of clinical signs. Isolation of NTM from respiratory specimens does not always indicate disease; clinical, radiographic, and microbiologic criteria must all be met to make a diagnosis of NTM lung disease according to ATS guidelines [1,2].
Although ATS criteria represent the cornerstone of NTM diagnosis, some limitations need to be addressed: the uncertain significance of isolation of NTM; the experience and training required in using specific microbiological diagnostic kits; and use of X-ray investigations for NTM identification.
Radiological patterns of lung NTM disease include nodular lesions, cavitary pattern, bronchiectasis, lobar consolidation, and consolidation associated with fibrothorax, as well as possible mix patterns (e.g., nodular-bronchiectasis) [3]. Lymphadenitis is the main clinical manifestation of NTM in children aged 1-5 years [4,5], while skin and soft tissue involvement is not uncommon.
Lymph node involvement in NTM is not easily interpreted at thoracic CT scan unlike M. tuberculosis lung infections; NTM patients do not have a primary complex, colliquation of nodal stations is not detectable, and lymph node repair in calcification or fibrocalcification is absent. 18 F-FDG positron emission tomography-computed tomography (PET/CT) is a functional imaging technique which detects metabolically active areas through the accumulation of the radiotracer 18-fluorodeoxyglucose. PET/CT is widely used in cancer management [6,7] and is already employed in the assessment of inflammatory/infectious respiratory diseases with lymph node involvement such as sarcoidosis, TB, and intracellular infections [8][9][10][11].
The purpose of the study was to assess and compare, through the use of 18 F-FDG PET/CT, lung and mediastinal lymph node involvement in patients with NTM and patients with active TB. The data were collected at "Vincenzo Monaldi" Hospital, Naples, center for Phthisiology and clinical laboratory investigation in mycobacteriology.

Materials and Methods
Between 2008 and 2011 79 biological culture samples positive for NTM were evaluated in our mycobacteriology center.
In order to make an accurate disease diagnosis, according to ATS 2007 guidelines for NTM lung pathology, clinical, microbiological, and radiological criteria were all considered. 26 NTM patients were enrolled.
During the same period (2008-2011) 486 samples positive for Mycobacterium tuberculosis were evaluated. Only 6 patients affected by active TB according to medical history, clinical, radiology, and acute-phase reactants underwent 18 F-FDG PET/CT.
The clinical records and radiological examinations of patients with positive cultures for NTM species and TB by sputum or bronchial wash were reviewed retrospectively, including information about age, sex, respiratory symptoms, preexistent pulmonary and nonpulmonary illnesses, results of anti-HIV antibody, number of positive isolates, and available lung biopsy results (Table 1).
Six patients with NTM infection and 6 patients with diagnosis of TB who underwent an 18 F-FDG PET/CT were considered eligible for the study.

NTM Patients.
All were male, Caucasian, with mean age of 61 (between 38 and 77 years), HIV negative presenting with nonspecific symptoms such as cough with sputum, evening fever, and dyspnea. Comorbidities such as COPD, diabetes mellitus, and bronchial and allergic asthma were present, and one patient had received an aortic valve replacement. Four NTM patients had a positive microbiological diagnosis on sputum samples (M. kansasii = 2, MAC = 1, and M. xenopi = 1) and two on bronchial aspirate (M. kansasii = 1, MAC = 1). Expected average dose is 370-555 MBq. The administration of the radiotracer was followed by oral hydration (to promote an appropriate distribution of the tracer in the tissues and its urinary excretion) and by a rest period in which the patient was informed not to walk, speak, or make any kind of effort to avoid the physiological uptake by muscle overactivity especially at the level of the eye muscles and vocal cords. 50 minutes after the injection of 18 F-FDG the patient was positioned supine in the PET scanner. The PET/CT was composed of a dedicated PET scanner with a detector LSO with crystal dimensions of 4.0 × 4.0 × 20 mm, transaxial field of view of 585 mm, an axial field of view of 162 mm, intrinsic axial and transaxial resolution that was between 4.6 and 5.8 mm, and a multislice CT. The parameters used for the acquisition included CT 120 kV, 80 mA, 16 slices helical, and 0.5 s for rotation. Initial TC without contrast was carried out followed by a PET scan performed on the body of the patient in the same position from the head to knee level. In order to better assess the involvement of the lung and lymph node. Initial CT without contrast was carried out followed by a PET scan performed on the body of the patient in the same position from the head to knee level.

Active Tuberculosis
After reconstruction of the coronal, sagittal, and transverse planes, the images were interpreted qualitatively and semiquantitatively with the standardized uptake value (SUV).
An experienced radiologist in nuclear medicine evaluated radiotracer uptake in mediastinal lymph nodes.

Statistical Analysis.
The values were expressed as mean ± SD and were compared using the Mann-Whitney test for two unpaired samples. Differences were considered statistically significant when < 0, 05. The results have been obtained through the use of MATLAB program.
At CT scan more than one radiological pattern was simultaneously found in active tuberculosis patients. CT findings included parenchymal consolidation or ground glass opacity (GGO) ( = 6) with SUV max between 1,20 and 22,75, excavation ( = 3) SUV max 4,19-15,10, and miliary ( = 1) SUV max 8,25, with lymph node involvement; in one case a combination of nodular consolidation PET-positive and bilateral pleural parietobasal effusion PET-negative was identified. PET/CT showed the presence (for all of 6 patients) of extensive areas of high metabolic activity (SUV max range 1,20-22,75 ( Figure 4) and average values of SUV max of 10,07 ± 6,45).
The data obtained with PET/CT (Table 4) shows that patients with active TB have high metabolic activity in lung lesions and in mediastinal lymph nodes compared to NTM. Furthermore NTM patients, while presenting with lung lesions exhibiting high uptake, showed little or no lymph node involvement.

Discussion
Our study suggests that 18 F-FDG PET/CT is a useful diagnostic technique which may be helpful in the management of pulmonary NTM and M. tuberculosis infections. Indeed we have demonstrated that both NTM and M. tuberculosis patients exhibit an increase of SUV in lung lesions although average values were less in NTM than in M.  tuberculosis; lymph nodes SUV value was low or zero in NMT whilst results were high in active tuberculosis patients. Despite well-established diagnostic criteria, NTM lung disease remains challenging for clinicians and new approaches are required for improving disease management.
The literature provides only a few studies regarding the potential usefulness of PET/CT as a diagnostic tool in nontuberculous mycobacteriosis.
Currently the 18 F-FDG PET/CT is widely used in cancer management for diagnosis, staging, and response to therapy [7,12]. There is now a growing interest in the potential role of 18 F-FDG PET in the functional assessment of inflammatory and infectious pulmonary diseases including mycobacteria.
A study conducted by Goo et al. showed that the pulmonary tuberculoma commonly causes an increase in uptake of 18 F-FDG suggesting caution in differentiating benign from malignant pulmonary abnormalities [13].
Similarly, the literature reports various cases of pulmonary nodular lesions with PET/CT high uptake caused by infection with NTM (especially MAC) whose differential diagnosis with malignant nodule was performed through histological typing [14,15].
Another study by Demura et al. evaluated the use of 18 F-FDG PET/CT in the diagnosis and monitoring of therapy for mycobacteriosis. The average uptake was higher in patients with MAC than those with TB. The study also confirms that, following one or two years of treatment for MAC, despite the persistence of nodules on HRCT, the 18 F-FDG PET uptake was more likely to correlate with disease state [16].
Our data differ from that reported by Demura, which may be due to differing patient selection criteria including the lung lesion characteristics and subtype of NTM species in patients enrolled.
Davis et al. have shown the utility of PET/CT to detect tuberculous lesions and to monitor response to chemotherapy in animal models hypothesizing the possible application for humans [17].
In our study we have demonstrated the usefulness of 18 F-FDG PET/CT in reflecting the activity and extent of disease by monitoring metabolic activity not only in nodular lesions but also within a broad range of radiologically visible lung lesions in NTM and M. tuberculosis. This is the first study in which PET/CT characteristics of lesions different from tuberculoma or single nodular lesion have been evaluated.
The broad range of radiological patterns seen in NTM and M. tuberculosis includes nodular or pseudonodular lesions, parenchymal consolidation, cavitary lesions, pleural thickenings, and pleural effusions and may reflect differences in cellular immune response at the level of the diseased lung [18][19][20].
One of the most important findings from this study is the role of 18 F-FDG PET/CT in the comparative evaluation of both pulmonary lesions and mediastinal lymph nodes in patients with NTM and those with active TB.
Our results have shown that, in both NTM and M. tuberculosis, increased uptake of the radioisotope at PET/CT scan is representative of active inflammatory areas within lung lesions which appears to reflect the real extent of disease.
PET/CT scan allows simultaneous assessment of both parenchymal and lymph node lesions which is of great value in tuberculous disease where lung involvement is almost always accompanied by lymphadenopathy.
Although performed on a small number of patients, our observations suggest a potential role for PET/CT in assessing extent of disease, evolution, and follow-up in both NTM and M. tuberculosis patients. Further investigations are required to verify whether there is the possibility to identify specific SUV cutoffs which in turn may define the