If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Respiratory Medicine, Hannover Medical School and Biomedical Research in End-stage and Obstructive Lung disease (BREATH), German Center for Lung Research (DZL), Hannover, GermanyEuropean Reference Network on Rare and Complex Respiratory Diseases (ERN-LUNG), Frankfurt, Germany
Numbers of mycobacterial infections stagnate in German cystic fibrosis patients.
•
Mycobacterium abscessus causes roughly half of the cases, and Mycobacterium avium causes about one-third.
•
The risk of M. avium infection rises with age.
•
Aspergillus fumigatus is associated with mycobacterial infection.
Abstract
Objectives
People with cystic fibrosis (pwCF) are at risk for infection with nontuberculous mycobacteria (NTM). The epidemiology and screening practice of NTM among pwCF in Germany are largely unknown and require investigation.
Methods
We analyzed the data of the German Cystic Fibrosis Registry from 2016 to 2020 for NTM. The annual prevalence and incidence of any NTM, Mycobacterium abscessus complex (MABC), Mycobacterium avium complex (MAC), Mycobacterium gordonae, and other mycobacteria were determined and correlated to patient characteristics. Patients with incident MABC and MAC infection were compared.
Results
The annual NTM prevalence and incidence remained stable between 7.53% and 8.76%, as well as 3.31% and 4.95%, respectively, among the approximately 6000 registry participants. MABC was the most common NTM, whereas only the prevalence of MAC increased slightly. In each year, only about one-third of all patients were screened for NTM. An association between NTM infections and Aspergillus fumigatus infection and/or allergic bronchopulmonary aspergillosis was observed. On average, patients with incident MAC infection were older than patients with MABC infection.
Conclusion
The NTM burden in pwCF in Germany remained unchanged between 2016 and 2020. MABC was the dominant species detected, whereas only MAC infections increased with time and patient age. The previously observed association of Aspergillus fumigatus and NTM was reaffirmed. Awareness of NTM needs to be improved.
People with cystic fibrosis (CF), a fatal, autosomal inherited, multisystem disease are particularly susceptible to pulmonary infection with nontuberculous mycobacteria (NTM) due to impaired airway clearance and resulting alterations in lung microenvironment [
]. The associated morbidity ranges from airway colonization to fatal disease, depending on host factors, as well as NTM species, with Mycobacterium abscessus complex (MABC) usually considered the most virulent [
]. The reported prevalence of NTM infection among people with CF (pwCF) varies vastly worldwide between 3% and 20%, depending on geographic region and the type of data collection [
Changing epidemiology of the respiratory bacteriology of patients with cystic fibrosis-data from the European cystic fibrosis society patient registry.
]. According to official international recommendations, spontaneously expectorating patients with CF should be screened for NTM at least annually, but the variable implementation of this recommendation in practice is likely to influence the reported prevalence [
US Cystic Fibrosis Foundation and European Cystic Fibrosis Society consensus recommendations for the management of non-tuberculous mycobacteria in individuals with cystic fibrosis: executive summary.
]. The detection rates of NTM in CF airways increased to a variable extent in several European countries and the United States (US) during the first 15 years of this millennium [
]. MABC is the most prevalent NTM species in Europe, followed by Mycobacterium avium complex (MAC), whereas the reverse is true in the United States. The spread of dominant MABC clones in the global CF population by (indirect) patient-to-patient transmission is well documented, whereas MAC and other NTM infections are usually environmental [
], this important information is largely unknown for pwCF in Germany. To close this data gap, we investigated the data from the German CF Registry. The primary objectives of the current study were to determine the prevalence and the incidence of NTM detection and the NTM screening practice in pwCF participating in the German CF Registry over a 5-year period from 2016 to 2020. The secondary objective was to identify host factors associated with NTM detection.
Materials and methods
Data source and patient population
The patient datasets investigated were obtained from the German CF Registry. The inclusion criteria for subsequent analysis were a confirmed diagnosis of CF and a valid written informed consent to CF Registry participation. The exclusion criteria were a diagnosis of CF transmembrane receptor (CFTR)-associated disease or a diagnosis of CF screening positive, inconclusive diagnosis, and previous lung transplantation. In the German CF Registry (https://www.muko.info/englisch-version/registry), NTM detection in respiratory specimen is documented in the following categories: M. abscessus subsp. abscessus, M. abscessus subsp. bolletii, M. abscessus subsp. massiliense, M. abscessus complex, M. complex, Mycobacterium chelonae, Mycobacterium fortuitum, Mycobacterium kansasii, Mycobacterium marinum, Mycobacterium terrae, other mycobacteria, and unknown. The detection of Mycobacterium gordonae in the group of other mycobacteria was documented by breakdown of the “other mycobacteria” group after contacting each participating CF center. Neither the NTM detection method nor the type of culture/broth used is specified in the registry. For the current study, the following categories of NTM were defined: any NTM, MABC (all M. abscessus complex members, regardless of subspecies), MAC (all M. avium complex members, regardless of subspecies), M. gordonae, and other (NTM species other than MABC, MAC, or M. gordonae). In addition, the following age groups were defined in years: 0-5, 6-13, 14-17, 18-23, 24-29, 30-39, 40-49, and ≥50. Of note, the patients’ ability to provide spontaneously expectorated sputum samples, and thus the ability to spontaneously provide high-quality microbiology airway samples, has only been recorded in the German CF Registry since 2020.
Statistical analysis
Screening practice, as well as prevalence and incidence, were analyzed by descriptive statistics. To identify the risk factors for any NTM, MABC, and MAC infection among pwCF, we analyzed the predictors of the prevalent (i.e., established presence of NTM) and incident (i.e., new detection of NTM) NTM detection using multivariate logistic regression models with one dependent variable. We used the data from 2016 as the baseline for comparison to predict NTM detections from 2017 to 2020. The analyses were adjusted for confounders based on clinical relevance and previous literature and following recent journal editor recommendations for causal inference [
Control of confounding and reporting of results in causal inference studies. Guidance for authors from editors of respiratory, sleep, and critical care journals.
The following variables were investigated for their association with the prevalence and the incidence of any NTM, MABC, and MAC from 2017 to 2020, using 2016 as the baseline: age (in 10-year intervals); sex; allergic bronchopulmonary aspergillosis (ABPA); concomitant detections of Achromobacter xylosoxidans, Aspergillus fumigatus, Burkholderia spp., Candida albicans, and/or Stenotrophomonas maltophilia (at least once a year each); chronic Pseudomonas aeruginosa infection; recent intravenous antibiotic therapy; inhaled rhDNase, CFTR genotype (delF508 homozygous [F/F] or delF508 heterozygous [F/other] or other genotype); CF-related diabetes with or without insulin therapy; long-term azithromycin treatment (>4 weeks); hospitalizations; inhaled antibiotic therapy; body mass index; pulmonary exacerbations; CFTR modulator use; forced expiratory volume per second in percent predicted (ppFEV1; stratified into three groups: group I ppFEV1 >70%, group II ppFEV1 40%-70%, group III ppFEV1 <40%).
To further identify and characterize pwCF at risk for MABC or MAC infection, we performed the following grouped comparison: all patients with first MABC or MAC detection (incident detection), respectively, from 2017 to 2020 were grouped and the previously mentioned variables were compared between MABC and MAC groups. The Satterthwaite t-test was used for continuous variables when the skewness in the groups was between -1 and +1, or the chi-square test for categorical variables.
Results
Overall, the number of eligible participants in the German CF Registry steadily increased from 5498 in 2016 to 6295 in 2020 (Figure 1, Figure 2 and 2, Table 1). Notably, the mean age of registry participants increased over the years (mean ± SD: 20.9 ± 13.6, 21.3 ± 13.9, 21.5 ± 14.1, 21.6 ± 14.4, and 21.9 ± 14.5 years from 2016 to 2020, respectively) (Figure 2, Table 1). Similarly, we observed an increase in lower airway samples tested for mycobacteria, except in 2020 (Table 1).
Figure 1Prevalence and incidence rates of NTM detection among participants of the German Cystic Fibrosis Registry over the analyzed years 2016 to 2020. Panel a shows absolute numbers of prevalent NTM cases (continuous line) and incident NTM cases (dashed line) as well as numbers of all registry participants and those tested for NTM. Panel b shows prevalence (continuous line) and incidence (dashed line) rates of all registry participants.
Table 1Prevalence and incidence of nontuberculous mycobacteria in German Cystic Fibrosis Registry participants over the years 2016-2020. Incidence was not determined in 2016 (1st year of analysis).
2016
2017
2018
2019
2020
n
% all (n = 5498)
% tested (n = 1767)
n
% all (N = 5856)
% tested (n = 2022)
n
% all (n = 6023)
% tested (n = 2176)
n
% all (n = 6109)
% tested (n = 2317)
n
% all (n = 6295)
% tested (n = 2112)
Prevalence
All
133
2,42
7,53
175
2,99
8,65
168
2,79
7,72
180
2,95
7,77
185
2,94
8,76
MABC
79
1,44
4,47
90
1,54
4,45
82
1,36
3,77
84
1,38
3,63
85
1,35
4,02
MAC
38
0,69
2,15
48
0,82
2,37
55
0,91
2,53
51
0,83
2,20
65
1,03
3,08
Mycobacterium gordonae
4
0,07
0,23
15
0,26
0,74
15
0,25
0,69
17
0,28
0,73
13
0,21
0,62
Other
12
0,22
0,68
22
0,38
1,09
16
0,27
0,74
28
0,46
1,21
22
0,35
1,04
Incidence
All
100
1,71
4,95
72
1,20
3,31
78
1,28
3,37
83
1,32
3,93
MABC
39
0,67
1,93
29
0,48
1,33
29
0,47
1,25
27
0,43
1,28
MAC
30
0,51
1,48
26
0,43
1,19
21
0,34
0,91
34
0,54
1,61
M. gordonae
15
0,26
0,74
11
0,18
0,51
12
0,20
0,52
8
0,13
0,38
Other
16
0,27
0,79
6
0,10
0,28
16
0,26
0,69
14
0,22
0,66
All = all registry participants; tested = registry participants with at least one lower airway sample analyzed for mycobacteria in the respective year. MABC, Mycobacterium abscessus complex; MAC, Mycobacterium avium complex.
Prevalence and incidence of NTM detection, 2016-2020
Overall, neither the prevalence nor the incidence rates of NTM detection changed considerably from 2016 to 2020 or 2017 to 2020 (Figure 1, Table 1). MABC was the most common NTM species in all the years analyzed, with relatively stable prevalence and incidences rates during the study period, ranging from 3.63% and 4.47% and 1.25% and 1.93%, respectively (Table 1, Figure 1). Overall, the prevalence of MAC increased slightly over time (2.15% in 2016 to 3.08% in 2020), whereas there was a fluctuating trend in incidence (Table 1, Figure 1). Of note, the incidence of MAC exceeded that of MABC in 2020 (Table 1, Figure 1). M. gordonae, although rarely of clinical importance, is reported separately in our analyses as a quality indicator for laboratory detection of mycobacteria. Its numbers were low and stable (Table 1, Figure 1). In addition, there were only sporadic cases of other NTM species (Table 1, Figure 1).
Mycobacterial screening practice
Mycobacterial culture was performed at least once a year in approximately one-third of all pwCF in all years (32.1%, 34.5%, 36.1%, 37.9%, and 33.6% for 2016-2020), resulting in substantial differences in prevalence and incidence, depending on the reference population (all registry participants vs participants tested for mycobacteria; Table 1, Figure 3). Of note, sputum or other lower airway specimens were not available for 2525 registry participants (40.1%) in 2020.
Figure 3Panel a shows the proportions of German Cystic Fibrosis Registry participants according to their NTM screening status. Panel b shows prevalence (continuous lines) and incidence (dashed lines) of NTM as related to all registry participants and those tested for NTM, respectively, illustrating the diagnostic gap attributable to unscreened patients.
Predictors of prevalent and incident NTM detection
In the any NTM group, the probability of a prevalent detection significantly increased with increasing age (in 10-year steps: odds ratio [OR] 1.24, 95% confidence interval [CI] 1.07-1.43), low ppFEV1 (<40% vs 40-70%: OR 1.46, 95% CI 1.04.-2.06), concomitant ABPA (OR 1.70, 95% CI 1.02-2.83), and the detection of Aspergillus fumigatus (OR 1.63, 95% CI 1.21-2.20), whereas it decreased with chronic P. aeruginosa infection (OR 0.58, 95% CI 0.4-0.86), F/F, or F/other genotype (OR 0.56, 95% CI 0.37-0.84 and OR 0.50, 95% CI 0.33-0.76; respectively, Figure 4, Table S1). The ORs for any incident NTM detection significantly increased with age (in 10-year steps: OR 1.18, 95% CI 1.00-1.38) and low ppFEV1 (<40% vs 40-70%: OR 1.77, 95% CI 1.20.-2.61). The ORs for any incident NTM detection significantly decreased with the concomitant detection of Stenotrophomonas maltophilia (OR 0.21, 95% CI 0.05-0.88), F/F, or F/other genotype (OR 0.55, 95% CI 0.35-0.86 and OR 0.51, 95% CI 0.32-0.81; respectively; Figure 4, Table S2).
Figure 4Odds ratios (and 95% confidence intervals) of CF patient characteristics for prevalent and incident detections of any NTM, MABC and MAC. Only statistically significant associations (p<0.05) are shown. The displayed patient characteristics are defined as follows: Age: age in 10 year intervals; ABPA: allergic bronchopulmonary aspergillosis; Aspergillus fumigatus/Stenotrophomonas maltophilia: concomitant detection at least once a year, each; Pseudomonas aeruginosa: chronic Pseudomonas aeruginosa infection: Genotype F/F: delF508 homozygous; Genotype F/Other: delF508 heterozygous; ppFEV1: forced expiratory volume per second in percent predicted (group I ppFEV1 >70%, group II ppFEV1 40-70%, group III ppFEV1 <40%).
For MABC, the ORs for prevalent detection increased significantly with concomitant ABPA (OR 3.26, 95% CI 1.77-6.01), detection of Aspergillus fumigatus (OR 1.74, 95% CI 1.15-2.64), and low ppFEV1 (<40% vs 40-70%: OR 1.73, 95% CI 1.05.-2.84). They decreased only with a F/other genotype (OR 0.52, 95% CI 0.29-0.94; Figure 4, Table S3). The ORs for incident MABC increased with low ppFEV1 (<40% vs 40-70%: OR 2.07, 95% CI 1.12.-3.82; Figure 4, Table S4).
For MAC, increasing age (in 10-year steps; OR 1.38, 95% CI 1.12-1.71) and detection of Aspergillus fumigatus (OR 1.63, 95% CI 1.03-2.6) predicted prevalent detection, whereas chronic P. aeruginosa infection decreased this probability (OR 0.53, 95% CI 0.29-0.97; Figure 4, Table S5). Increasing age (in 10-year steps; OR 1.33, 95% CI 1.06.-1.68) and low ppFEV1 (<40% vs 40-70%: OR 2.20, 95% CI 1.18.-4.09) were predictors of incident MAC detection; (Figure 4; Table S6).
Group comparison of pwCF with incident MABC vs MAC detection
The group comparison of pwCF with incident MABC vs MAC detection in 2017-2020 showed a higher mean age at the time of first detection in patients with MAC (mean [SD] 30 [
] years, P <0.0001), whereas patients with MABC had a lower mean body mass index (BMI; 19.9 [2.8] kg/m² vs 21.1 [3.8] kg/m², P = 0.01). In addition, patients with MAC had a lower mean (SD) ppFEV1 at first detection (67. 7 [23.0] % vs 74.2 [25.3] %, P = 0.0417; Table 2). All the other variables analyzed did not differ significantly between groups.
Table 2Grouped comparison of patients with incident MABC or MAC infection over the years 2017-2020. Data are show as mean (standard deviation) or numbers (%), as appropriate.
FEV1, forced expiratory volume in the first second; GLI, Global Lung Initiative; MABC, Mycobacterium abscessus complex; MAC, Mycobacterium avium complex.
Our study revealed moderate NTM prevalence and incidence in pwCF in Germany, which remained stable during the study period. The mean prevalence of NTM was around 8% among German CF Registry participants screened for NTM and approximately 4% among all registry participants. These figures are consistent with recent reports from the United Kingdom (UK) (3.8% of all UK CF Registry participants in 2015) and the US (10% of all US CF Registry participants in 2020) CF Registry data [
]. Of note, we found a trend toward increasing numbers of MAC detections, with the incidence of MAC exceeding that of MABC for the first time in 2020. The stable rate of M. gordonae detections over the study period argues against fluctuating laboratory performance confounding these findings. The probability of MAC detection was found to increase with age, and at the same time, the mean age of the population covered by the German CF Registry increased every year. Thus, the increasing life expectancy of pwCF might increase the lifetime risk of MAC infection. The differences between pwCF with incident MABC and MAC detection were previously described in 2013 for the French CF population [
]. Catherinot and colleagues also found that age was the main discriminator between groups, with patients with MAC being significantly older than those with MABC. We confirmed these findings in our analysis; although, both patients with MAC and MABC were markedly older than in the French study, especially given that we were comparing incident infections, whereas Catherinot and colleagues looked only at the prevalent cases. We also found that ppFEV1 and BMI were significantly different between the two groups. However, because ppFEV1 decreases with age, whereas BMI increases with age in patients with CF within the reported range, we considered that these differences are more age-related and not related to MAC or MABC infection [
Moreover, the association between concomitant Aspergillus fumigatus detection, as well as ABPA with NTM infection, has long been established and was confirmed in our study [
]. Our study indicates a decreased risk of NTM infection (particularly MAC) in pwCF with chronic P. aeruginosa infection. This finding supports previous US data [
]. These controversial results on the association between chronic P. aeruginosa and NTM infection probably highlight a complex and multifaceted interaction, as well as methodological difficulties. On the one hand, the preference of both pathogens for conditions given in advanced CF lung disease promotes co-existence but does not prove pathogen interaction [
]. Nevertheless, direct interaction of the pathogens in CF airway biofilms is likely, but mutual interference is not well understood and could be both beneficial or detrimental [
Non-tuberculous Mycobacteria multispecies biofilms in cystic fibrosis: development of an in vitro Mycobacterium abscessus and Pseudomonas aeruginosa dual species biofilm model.
]. We suspect that the sample sizes of existing epidemiologic studies are too small and the populations and methods too heterogeneous to account for the complex relationship of both pathogens. Ultimately, only larger epidemiological studies complementing basic microbiological research could adequately address this problem.
We found a higher likelihood of NTM (including MABC and MAC) infection in patients with a ppFEV1 <40%, indicating a higher risk of infection in patients with advanced structural lung disease.
Only about one-third of patients in each of the years from 2016 to 2020 were tested for NTM, which is a comparatively low proportion. The lack of lower airway samples is a major contributor to this diagnostic gap because 40% of patients did not provide sputum samples in 2020, the first year in which these data were available. The difficulty in obtaining spontaneously expectorated sputum samples is particularly high in younger patients [
]. Nevertheless, a considerable number of patients were not tested for NTM despite the availability of sputum samples, demonstrating nonadherence of German centers to international recommendations on NTM screening practice in pwCF, according to which at least annual screening of these patients is indicated [
US Cystic Fibrosis Foundation and European Cystic Fibrosis Society consensus recommendations for the management of non-tuberculous mycobacteria in individuals with cystic fibrosis: executive summary.
]. In this context, it is to be expected that collection of qualified sputum samples will probably decrease due to the widespread use of highly potent CFTR modulators and this will definitely influence future NTM diagnostics and screening in pwCF.
The impact of CFTR modulator therapy on the detection of NTM remains unclear. Recently, a slightly reduced risk of acquiring NTM under CFTR modulator therapy was described in the US CF population from 2011 to 2018 [
]. Our analyses do not support this because the use of CFTR modulators did not significantly influence ORs in our analyses. Interestingly, however, delF508 genotypes—currently, apart from rarer gating mutations, a prerequisite for CFTR modulator therapy—were associated with a lower risk of NTM infection in our analyses. Clearly, the role of CFTR modulators on the epidemiology of NTM in pwCF requires further investigation, especially when more potent CFTR modulators become available for patients at younger ages.
Our study has strengths and limitations. Although it represents the most comprehensive analysis of the epidemiology of NTM in pwCF in Germany to date, the granularity of the German CF Registry does not allow us to differentiate between NTM detection or NTM infection and NTM pulmonary disease, respectively, or even to investigate the applied treatment modalities [
]. In this regard, our study revealed an obvious limitation of the German CF Registry that will be addressed in the near future and thus may contribute to further improvement in the care of pwCF and NTM infection.
Conclusion
In conclusion, the prevalence and incidence of NTM detection in pwCF remained stable during our study period in Germany. The increasing age of the German CF population might favor infections by MAC. So far, CFTR modulators have not yet reduced the burden of NTM infections in pwCF; so, CF physicians need to be alert to possible NTM infections. Improvement in the screening practices and the challenging task of obtaining lower airway specimens urgently needs to be addressed.
Declaration of competing interest
MS reports grants from Mukoviszidose e.V.; personal honoraria for lectures and sponsored seminars from Vertex; fees for clinical trial participation for his institution from Vertex, Gilead, and Corbus. FS reports fees for clinical trial participation from Vertex, Gilead, and Corbus; personal honoraria for lectures from Vertex, Biogen, and Novartis; personal payments for participation on advisory boards from Vertex and Gilead. LN reports grants or contracts from German Center for Lung Research, Vertex Pharmaceuticals, and Boehringer Ingelheim for clinical trial participating; participation on a trial steering committee for CF STORM, medical lead of the German CF Registry, pharmacovigilance study manager of the ECFSPR; receipt of medical writing from Articulate Science. FCR reports grants from the German Center for Lung Research (DZL), the German Center for Infection Research (DZIF), the Innovative Medicines Initiative (IMI; EU/EFPIA), and the iABC Consortium (including Alaxia, Basilea, Novartis, and Polyphor), Mukoviszidose Institute, Novartis, Insmed Germany, Grifols, Bayer Healthcare, and InfectoPharm paid to his institution; fees for clinical trial participation from Abbvie, AstraZeneca, Boehringer Ingelheim, Celtaxsys, Corbus, Insmed, Novartis, Parion, University of Dundee, Vertex, and Zambon paid to his institution; personal consulting fees from Parion Sciences, Grifols, Zambon, Insmed, and Helmholtz Center for Infection Research; personal honoraria for lectures from!DE Werbeagentur GmbH, Interkongress GmbH, AstraZeneca, Insmed, Grifols, University Hospital Frankfurt; payment for expert testimony from the Social Court Cologne; financial support for attending meetings from the German Kartagener Syndrome and Primary Ciliary Dyskinesia Patient Advisory Group, as well as the German Cystic Fibrosis Patient Advisory Group (Mukoviszidose e.V.); personal payments for participation on an advisory board from Insmed, Grifols, and Shionogi; as well as honorary (unpaid) commitment as coordinator of the ERNLUNG Bronchiectasis Core Network, co-chair of the German Bronchiectasis Registry PROGNOSIS, member of the steering committee of the European Bronchiectasis Registry EMBARC, member of the steering committee of the European NTM Registry EMBARC-NTM, co-speaker of the Medical Advisory Board of the German Kartagener Syndrome and Primary Ciliary Dyskinesia Patient Advisory Group, speaker of the Respiratory Infections and TB group of the German Respiratory Society (DGP), speaker of the Cystic Fibrosis group of German Respiratory Society (DGP), principal investigator of the German Center for Lung Research (DZL), member of the Protocol Review Committee of the PCD Clinical Trial Network, and member of Physician Association of the German Cystic Fibrosis Patient Advisory Group (Mukoviszidose e.V.). The other authors have no competing interests to declare.
Funding
The German Cystic Fibrosis Registry and its data curation is funded by the German Mukoviszidose Institut gGmbH. No specific funding was obtained for this study.
Ethical approval
All German CF Registry participants gave written informed consent for registry participation. Ethical approval for this study was covered by the ethics committee of the Justus-Liebig University Gießen FB Medizin (AZ 24/19).
Authors’ Contributions Statement
FCR and MS conceived the study and designed the analysis plan with intellectual input from SH, CR and LN. SH coordinated and supervised the study process. CR extracted and analyzed the data from the registry with scientific input from FCR, MS and LN. FCR, MS and FS put the study data in the scientific context. MS and FCR wrote the paper with significant intellectual input from all authors. All authors read and approved the final manuscript.
Changing epidemiology of the respiratory bacteriology of patients with cystic fibrosis-data from the European cystic fibrosis society patient registry.
US Cystic Fibrosis Foundation and European Cystic Fibrosis Society consensus recommendations for the management of non-tuberculous mycobacteria in individuals with cystic fibrosis: executive summary.
Control of confounding and reporting of results in causal inference studies. Guidance for authors from editors of respiratory, sleep, and critical care journals.
Non-tuberculous Mycobacteria multispecies biofilms in cystic fibrosis: development of an in vitro Mycobacterium abscessus and Pseudomonas aeruginosa dual species biofilm model.
00036-X&id=gr1.jpg){kind=link}
00036-X&id=gr2.jpg){kind=link}
00036-X&id=gr3.jpg){kind=link}
00036-X&id=gr4.jpg){kind=link}