Background
Nontuberculous mycobacterial pulmonary disease (NTM-PD) cases are steadily increasing worldwide, with an annual prevalence of 3.2–9.8 per 100,000 people in North America [
1,
2] and 14.7 per 100,000 person-years in Japan in 2014, approximately 2.6 of the incidence rate reported in 2007 [
3]. This disease causes chronic cough, sputum, hemoptysis, fatigue, malaise, and weight loss [
4] and reduces the quality of life in advanced stages.
Bronchial asthma (BA) is a common chronic airway inflammatory disease affecting 1%-29% of the population [
5] and a common cause of chronic cough [
6]. It is defined by a history of characteristic respiratory symptoms, such as wheezing, shortness of breath, chest tightness, cough, and evidence of variable expiratory airflow limitation [GINA 2023 [
5]. However, the diagnosis of asthmatic cough is sometimes difficult because there are no validated tests or approved diagnostic criteria for asthma [
7].
Many asthmatics have inflammation driven by type 2 cytokines [
8], and increased fractional exhaled nitric oxide (FeNO) release from bronchial epithelium [
9]. FeNO is modestly correlated with airway eosinophilic inflammation [
10] and airway hyperresponsive [
11], making it a useful marker for diagnosing asthma [
12];-therefore, several diagnostic guidelines including FeNO have been proposed [
13,
14]. These guidelines have established FeNO cutoff values and advocate a process whereby if asthma symptoms are present and the FeNO value is above the cutoff value, the patient should be diagnosed with asthma and begin treatment, including an inhaled corticosteroid (ICS). FeNO has also proven to be a useful marker for diagnosing asthmatic cough, showing a relatively high specificity of 0.85 in predicting adult asthma with chronic cough [
15].
Although it is easy to speculate that the asthmatic component may be hidden in the coughing of patients with mycobacterial infections, asthmatic cough may be overlooked because NTM-PD itself causes a chronic cough. In addition, while ICS is an essential treatment for BA, there is some concern that this approach may lead to exacerbation of NTM-PD [
16]. Conversely, ICS might have a positive impact on asthma symptoms in NTM and may lead to improved quality of life in patients.
The aim of the present study was to examine the diagnostic impact of measuring FeNO in predicting asthmatic cough (BA and cough variant asthma) in patients with NTM-PD, and the effects of ICS on NTM-PD complicated by asthma.
Methods
Study design and patients
This retrospective study was performed at the National Hospital Organization, Osaka Toneyama Medical Center (Osaka, Japan), a referral center for respiratory diseases. We enrolled patients with NTM-PD with FeNO measured because of a cough suspected of having asthma between April 1, 2014, and March 30, 2019, and for whom follow-up data were available for at least six months after the first FeNO measurement. (for details regarding the inclusion criteria and exclusion criteria, see the online
supplemental methods).
This study was conducted in accordance with the principles of the Declaration of Helsinki and the experimental protocol for data involving human followed the Ethical Guidelines of the Japan Ministries of Health and Labour for Medical and Health Research Involving Human Subjects. The protocol was approved by the Clinical Research Review Board of the National Hospital Organization Osaka Toneyama Medical Center (approval number: TNH-A-2021026), and the need to obtain written informed consent was waived owing to the retrospective nature of the study.
Data collection
Baseline clinical parameters, including age, sex, body mass index (BMI), smoking status, allergic disease history (allergic rhinitis, pollinosis, infantile asthma, and atopic dermatitis), and other comorbidities, were collected from each patient’s medical records at the first FeNO measurement. Data on the total eosinophil count, serum immunoglobulin E (IgE), anti-glycopeptidolipid-core IgA antibody, FeNO, spirometry, chest radiography, and high-resolution computed tomography (HRCT) findings were also collected. FeNO was measured using a NO analyzer (NIOX MINO or NIOX VERO; Aerocrine, Solna, Sweden).
Definition of the NTM+BA and NTM groups
Asthma was defined by a history of variable respiratory symptoms, such as wheezing, shortness of breath, cough, and reversible expiratory airflow limitation, according to the Global Initiative for Asthma (GINA) [
17] . However, airway reversibility can be difficult to prove in some asthmatics [
18]. Therefore, we applied the following criteria to define NTM with asthmatic components (the NTM + BA group).
1) NTM patients with symptoms consistent with asthma, and
2) NTM patients with symptomatic improvement after diagnostic therapy with ICS ± a long-acting beta2-agonist (LABA).
It was preferred to the above conditions as the diagnosis of asthma that reversible FEV1, defined as an increase in the FEV1 of ≥12% and ≥200 mL after 2 inhalations of salbutamol, demonstrated an increase in the FEV1 of ≥200 mL after diagnostic therapy with ICS±LABA. Diagnostic ICS treatment for NTM-PD with a history of variable cough was performed, considering any findings suggestive of asthma, such as increased eosinophilia, an FeNO value ≥22 [
19], atopic disposition, and a positive bronchial challenge test with methacholine. The remaining patients were assigned to the non-tuberculous mycobacterial (NTM) group. The grouping was performed by two chest physicians with more than 10 years of clinical experience.
Radiological evaluations
HRCT was performed using a 64-row multidetector row CT scanner (SOMATOM Definition AS+; Siemens, Munich, Germany). Images at the first FeNO measurement ± 3 months were independently analyzed by two chest physicians, in the direction of a radiologist. We used the CT scoring system proposed by Lee et al. to assess HRCT findings of NTM-PD (Supplemental Table
1) [
20].
Table 1
Baseline characteristics of NTM-PD patients
Age (years) | 70.0 (60.5-76.0) 67.03333 | 69.0 (54.0-79.0) | 70.0 (61.0-75.0) | 0.9897 |
Sex, female, n (%) | 83 (93.3%) | 28 (90.3%) | 55 (94.8%) | 0.4220 |
Body mass index (kg/m2) | 19.6 (17.8-21.8) | 19.6 (18.4-21.2) | 19.6 (17.4-22.0) | 0.9083 |
onset age of NTM (y) | 61.0 (52.5-69.0) | 61.0 (52.0-74.0) | 61.5 (52.8-68.0) | 0.9656 |
Duration (y) of NTM | 5.7 (1.0-11.5) | 4.8 (0.7-11.7) | 5.9 (1.6-11.4) | 0.4641 |
Duration (y) of asthma | - | 0.67 (0.0-5.9) | - | - |
Smoking status (Never/Ex) n | 79/10 | 28/3 | 51/7 | >0.9999 |
Underlying diseases, n (%) |
Diabetes mellitus | 6 | 1 | 5 | 0.6605 |
Collagen disease | 4 | 2 | 2 | 0.6082 |
immunosuppressive agent use | 2 | 1 | 1 | >0.9999 |
Allergic disease, n (%) | 29 (32.6%) | 16 (51.6%) | 13 (22.4%) | 0.0085 |
FeNO (ppb) | 18.0 (12.0-28.0) | 23.0 (15.0-43.0) | 17.0 (11.8-23.0) | 0.0151 |
Blood eosinophil count (cells/μl) | 140.0 (90.0-227.5) | 160.0 (90.0-270.0) | 130.0 (80.0-205.0) | 0.1361 |
IgE (IU/ml) | 60.0 (16.0-160.5) | 58.5 (16.0-228.0) | 60.0 (16.0-130.5) | 0.4046 |
NB form | 87 (97.8%) | 29 (93.5%) | 58 (100%) | 0.1187 |
Cavity | 23 (25.8%) | 5 (16.1%) | 18 (31.3%) | 0.2031 |
AFB smear, positive | 23 (25.8%) | 10 (32.3%) | 13 (22.4%) | 0.3224 |
Anti-GPL core IgA antibody levels (U/ml) | 2.50 (0.84-10.22) | 1.35 (0.76-5.43) | 4.20 (0.84-12.04) | 0.1613 |
Mycobacterium species |
M. avium | 54 (60.7%) | 19 (61.3%) | 35 (60.3%) | >0.9999 |
M. intracellulare | 22 (24.7%) | 8 (25.8%) | 14 (24.1%) | >0.9999 |
M. avium & intracellulare | 3 (3.3%) | 1 (3.2%) | 2 (3.5%) | >0.9999 |
M. abscessus complex | 6 (6.7%) | 2 (3.3%) | 4 (8.4%) | >0.9999 |
M. intracellulare+chelonae | 2 (2.2%) | 1 (3.2) | 1 (1.7%) | >0.9999 |
M. gordonae | 2 (2.2%) | 0 (0.0%) | 2 (3.4%) | 0.5409 |
Treatment of mycobacterium |
No treatment | 37 (41.6%) | 15 (48.4%) | 22 (37.9%) | 0.3733 |
On treatment | 52 (58.4%) | 16 (51.6%) | 36 (62.1%) | |
EM or CAM monotherapy monotherapy | 21(23.6%) | 6 (19.4%) | 15 (25.9%) | 0.6042 |
CAM -included regimen ≧2 | 5 (5.6%) | 1 (3.2%) | 4 (6.9%) | 0.6543 |
CAM -included regimen ≧3 | 25 (28.1%) | 9 (29.0%) | 16 (27.6%) | >0.9999 |
Other regimen | 1 (1.1%) | 0 (0.0%) | 1 (1.7%) | >0.9999 |
Duration from start any treatment of mycobacterium (y) | 3.80 (1.40- 7.55) | 2.05 (1.08-3.60) | 3.80 (1.40-7.55) | 0.0988 |
Pulmonary function test* |
FEV1, L | 1.98 (1.71-2.25) | 2.00 (1.67-2.35) | 1.85 (1.76-2.25) | 0.3812 |
FEV1, % predicted | 85.2 (66.1-100.0) | 87.3 (66.5-103.2) | 82.7 (66.0-98.7) | 0.7156 |
FEV1/FVC, % | 79.2 (76.9-80.5) | 79.2 (75.7-80.8) | 78.8 (77.2-80.4) | 0.8932 |
FEF50, L/s | 1.56 (1.05-2.26) | 1.71 (1.04-2.47) | 1.49 (1.02-1.92) | 0.4743 |
FEF25, L/s | 0.40 (0.20-0.64) | 0.47 (0.26-0.98) | 0.36 (0.19-0.63) | 0.3947 |
FEF50 % predicted | 54.8 (35.6-80.3) | 60.9 (35.5-84.6) | 52.1 (35.2-76.3) | 0.5205 |
FEF25 % predicted | 34.1 (21.4-53.6) | 35.6 (23.9-70.6) | 33.7(19.3-53.6) | 0.3947 |
Asthma treatment |
ICS | 11 (12,4%) | 11 (35.5%) | 0 (0.0%) | <0.0001 |
Dose of ICS (μg/day)† (FP Conversion amountdose) | 0 (0-0) | 0 (0-320) | 0 (0-0) | <0.0001 |
OCS | 1 (1.1%) | 1 (3.2%) | 0 (0%) | 0.3483 |
LAMA | 2 (2.3 %) | 0 (0.0%) | 2 (3.5%) | >0.9999 |
LABA | 17 (19.1 %) | 13 (41.9%) | 4 (6.9%) | 0.0001 |
leukotriene modifier, n (%) | 9 (10.1 %) | 9 (29.0%) | 0 (0.0%) | <0.0001 |
theophylline, n (%) | 2 (2.3%) | 2 (6.5%) | 0 (0.0%) | 0.1187 |
Statistical analyses
All statistical analyses were performed using the GraphPad Prism ver. 9 (GraphPad Software, San Diego, CA, USA) and JMP version 11 (SAS Institute, Cary, NC, USA). Continuous variables are expressed as median and interquartile range (IQR), and categorical data are expressed as numbers and percentages. Patient groups were compared using the Mann–Whitney U test for continuous variables and the χ
2 test or Fisher’s exact probability test for categorical variables. The results among more than two groups were compared using the Kruskal-Wallis test followed by the Wilcoxon test. Statistical significance was set at
P <0.05. Several patients had missing pulmonary function test and CT at the first FeNO measurement, in which case the number of patients was indicated in the tables and figures, and the missing values were omitted from the analysis. The sample size was estimated according to the previous report related to the FeNO cutoff values for the diagnosis of asthma [
19], and the minimum number of patients was estimated to be 14 for each group using EZR (version 1.64, Saitama Medical Center, Jichi Medical University, Saitama, Japan).
Discussion
The NTM+BA group, which included NTM cases with asthmatic components, more frequently had an allergic history and had higher FeNO values than the NTM group, and the FeNO cutoff value for discriminating these groups was 21.5 ppb according to the ROC curve. The NTM-preceding subgroup treated with ICS+LABA after a tentative asthma diagnosis showed prompt improvement in coughing and no exacerbation of AFB culture within six months in the majority of cases. In addition, several patients showed culture conversion after initiation of ICS therapy.
Some asthma cases do not show clear airway reversibility [
5], and several guidelines have proposed the use of FeNO to guide diagnosis [
13,
14]. The FeNO cut-off value for differentiating between NTM and asthma-associated NTM was calculated as 21.5 ppb in this study, which is close to 22 ppb, the value used to differentiate between asthmatic and normal Japanese patients [
19]. Furthermore, some NTM+BA patients were worsened their coughing after a common cold, pneumonia with fever, or hay fever attack. These findings agree with the frequently reported issue of asthmatic cough exacerbation triggered by viral infections [
22] or hay fever [
23], and indicate that the process of diagnosing and treating asthma complicated by NTM does not differ from that of usual asthma cases. Furthermore, the present study showed no marked difference in pulmonary function test findings between the NTM+BA and NTM groups, including in peripheral airway-related indices (FEF50, FEF25), indicating the usefulness of FeNO measurement in addition to the usual pulmonary function tests for asthma associated with NTM-PD as well as the usual diagnosis of asthma, as described above. However, there have been few detailed reports on coughing in NTM-PD, so further investigations into the mechanism of coughing are warranted.
The safety of ICSs in patients with NTM remains unclear. Bak et al. reported using a mouse model that comorbid allergic asthma exacerbated
M. avium pulmonary infection (Mav-PI) by reducing the mycobacterium-specific Th17 response, which plays an important role in defense against intracellular pathogens [
24]. They further showed that the aggravation of Mav-PI with a suppressed Th17 immune response was more prominent when allergic asthma was induced after
M. avium infection than when asthma preceded it, and the comorbid allergic asthma in Mav-PI showed reduced disease progression over time, accompanied by a diminished degree of goblet cell hyperplasia with reduced IL-13 production. Crosstalk between Th17 and Th2 has also been shown to be mutually exclusive in patients with chronic rhinosinusitis with nasal polyps [
25] .
In the present study, we report cases in which asthma symptoms improved without exacerbation of NTM-PD for six months after ICS therapy was started or intensified. Given the present findings and previous reports of Th2-Th17 crosstalk over several months, suppression of Th2 inflammation by ICS may lead to the improvement of clinical symptoms without exacerbation of NTM-PD. Based on the findings of this study, the NTM-PD patients with atopic disposition and variable symptoms or triggering symptoms suggestive of asthma should be aggressively examined for airway reversibility and FeNO to determine whether or not they have asthma complications. If possible, bronchial challenge testing with methacholine should be performed before starting ICS treatment to strengthen the asthma diagnosis. In this study, only four patients in the NTM+BA group underwent bronchial challenge testing with methacholine and showed a positive reaction. In this small number of cases, we found, surprisingly, that ICS-based asthma treatment improved in several cases of refractory NTM-PD itself. There is a major question of how the ICS dose should be determined, but a short-term response to treatment with an ICS dose similar to that of usual asthma treatment can be considered for patients with NTM-PD who have positive findings suggestive of asthma. However, prolonged long-term ICS administration might exacerbate NTM-PD, as previously reported, and it may be necessary during ICS treatment to assess Th2 inflammation and reduce the ICS dose appropriately over time, with discontinuation of ICSs considered when asthma symptoms improve sufficiently. However, in some cases, continued ICS administration can be necessary due to increased Th2 inflammation. Further study is needed to determine the appropriate duration and dose of ICS therapy for asthma-complicated NTM-PD.
In this study, we also showed that the FeNO value was not correlated with the severity of total CT images or the degree of bronchiectasis or cellular bronchiolitis in NTM-PD patients. Bronchiectasis and bronchiolitis progress with the development of NTM-PD [
20]. In the present study, both bronchiectasis and cellular bronchiolitis were more severe in the NTM group than in the NTM+BA group, but FeNO values were lower in the NTM group. Regarding bronchiectasis, it was reported that bronchiectasis patients with
Pseudomonas aeruginosa infection had significantly lower FeNO values than those without
Pseudomonas infection, and that there was no correlation between FeNO and the number of bronchiectatic lung lobes [
26]. Our results were consistent with this report, suggesting that high FeNO values in NTM patients were indicative of asthma, regardless of the degree of imaging findings.
Limitations
The present study has several limitations. First, this study had a retrospective design and was a single-center study with a relatively small sample size. Admittedly, the NTM and NTM+BA groups had slightly different backgrounds, including smoking history. However, the major statistical backgrounds (e.g. age, sex, and severity of bacteriology of NTM) were consistent in both groups, Second, the diagnosis of the NTM+BA group was a clinical diagnosis, and the possibility cannot be ruled out that other bronchiolitis responsive to ICS/LABA may be included in the group. Bronchial challenge testing with methacholine should be performed for the more accurate diagnosis of asthma. Third, we were unable to identify a clear indicator that could distinguish pure NTM from NTM with asthmatic complications. Fourth, the majority of NTM-PD patients with asthma complications in this study had been asthmatic for only a few months at the measurement of FeNO. Larger and longer studies of multiple centers are needed to confirm the results of this study, and further investigations are required to identify patients indicated for ICS therapy and optimal dosage of ICS for NTM complicated by asthma in order to prevent the development of NTM.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.