Questions were presented in multiple-choice format with exception of question 16 and question 19, which were free-text responses.
The responses to question 19, which asked respondents to characterize their practice setting, were divided into 1 of 7 discrete categories that best fit the individual’s response. Those categories included Interventional Pulmonology, Combined Pulmonology/Critical Care, Fellowship/Training, Combined Interventional Pulmonology/Pulmonology/Critical Care, Pulmonology, Thoracic Surgery, and Undetermined.
Bronchoscopic images were illustrated in question 7 (Fig. 1) as such: Image (a) concentric narrowing of TBM, image (b) posterior membrane bulging of the trachea representative of EDAC, and image (c) crescent-type narrowing representative of TBM.
Paired dynamic expiratory-inspiratory computed tomographic (CT) scans were illustrated in question 9 (Fig. 1) as such: Image (a) reveals expiratory posterior membrane bulging of the trachea consistent with EDAC, whereas image (b) reveals expiratory anterolateral wall collapse of the trachea consistent with TBM.
The survey was created and displayed on surveymonkey.com. The link to the survey was sent to the American College of Chest Physicians who disseminated it by email to members of their Interventional Chest Diagnostic Procedures Network. These individuals were targeted as they represent those clinicians with interests in TBM and thereby are considered to have the strongest knowledge based on the topic. The link to the survey was sent out in January of 2015 and was maintained active through July of 2015.
Upon survey completion deadline, results were extracted from the web into a Microsoft Excel spreadsheet. Cross tabulations were performed using the Fisher t tests and χ2 analyses to investigate whether certain groups of respondents were more likely to demonstrate better awareness or competencies with TBM, or exhibit different practice patterns. Among the trends we wanted to analyze were whether interventional pulmonology (IP) gave different answers than the other respondents and whether clinicians who said they felt comfortable making the diagnosis of TBM, displayed better understanding of the disorder and competency with its diagnosis.
One hundred sixty-five clinicians participated in the survey from January to July 2015. Every question had at least 159 responses except questions 5a to d (105 to 125 respondents), question 16 (26), and question 19 (131). Ninety-three percent of respondents considered TBM a treatable entity. Key findings are summarized in Table 3.
The vast majority (87%) of respondents recognized that TBM is a distinct entity from EDAC, and most (77%) were able to define it correctly.
Awareness and Practice Patterns
Eighty-two percent of respondents will consider TBM in COPD patients that are uncontrolled despite adequate maintenance bronchodilator therapy, although only 73% said they proceed with a diagnostic workup. Most clinicians who suspect the condition will utilize pulmonary function tests (PFTs) (88%), CT (83%), and bronchoscopy (89%) to work it up. Eight respondents said they would perform diagnostic fluoroscopy. A total of 90% of respondents agreed that patients who have unexplained extubation failures should be considered for TBM.
Twenty-two percent of respondents said they never consider TBM in their patients with cough, sputum production, dyspnea, and recurrent infections. Seventy-three percent of respondents used PFTs to influence their diagnostic plan. Thirty-eight percent of respondents said that if PFTs were normal, they would not proceed with further evaluation of TBM. Only 18% of respondents use a classification system to describe the severity of TBM and only 6 respondents (4%) reported using the FEMOS classification. Of clinicians who would refer their patients for management of TBM, 26% would refer to either a thoracic surgeon or ENT. Eighty percent of respondents do not diagnose TBM solely based on paired dynamic CT imaging and 90% make the diagnosis of TBM based on bronchoscopic findings.
Ability to Differentiate TBM From EDAC on Imaging and Bronchoscopy
In question 7, only 29% of respondents correctly identified the bronchoscopic findings of TBM represented by answer D (images “a” and “c”). Thirty-three percent selected answer C and 28% selected E (A, B, and C).
In question 9, 39% correctly identified B as the image displaying TBM, whereas 47% selected C (both).
Seventy-one percent of respondents perform dynamic functional bronchoscopy. Of the 131 respondents who answered question 19, the breakdown is listed in Table 1. The breakdown of years in practice is listed in Table 2. Over 50% of respondents had been in practice for 0 to 20 years.
Breakdown by Type of Practice (IP vs. Non-IP)
Twenty-one percent of clinicians in IP pursued a diagnosis of TBM in 50% or more of their patients with cough, sputum production, dyspnea, and recurrent infections. Meanwhile, only 6% of non-IP clinicians worked up TBM in at least 50% of these patients (P=0.0129).
Respondents that practice IP demonstrated a better knowledge base of TBM and were also able to more frequently make a diagnosis of TBM when compared with clinicians who did not have an IP practice. This was demonstrated by their ability to more often correctly define TBM (95% vs. 72%, P=0.0075) and recognize TBM as a separate entity from EDAC (100% vs. 84%, P=0.0111) as well as correctly identify TBM on CT (71% vs. 28%, P<0.0001).
When compared with other practice categories, respondents in IP were less likely to use PFTs to influence their decision to proceed with diagnostic evaluation of TBM (60% vs. 21%, P=0.0204). They were more likely to use a classification system to describe the degree of severity of TBM (40% vs. 14%, P=0.0004).
Breakdown Based on Years in Practice
Those with >30 years of experience were more likely to use fluoroscopy to workup TBM (19% vs. 4%, P=0.0214).
Although the result did not reach statistical significance, there was a trend showing that those with over 30 years of practice were less likely to select the appropriate CT images corresponding to a diagnosis of TBM (16% vs. 32%, P=0.0633).
The respondents who felt comfortable with making the diagnosis were more likely to pursue the diagnostic evaluation of TBM in patients with poorly controlled COPD (77% vs. 58%, P=0.0349).
Those that felt comfortable making the diagnosis of TBM were not significantly better in their ability to identify the condition on CT (27% vs. 35%, P=0.3706) or bronchoscopy (43% vs. 25%, P=0.0586) than the respondents who did not feel comfortable making the diagnosis.
This survey supports our hypothesis that pulmonologists have an insufficient knowledge base regarding TBM and exposes that deficits exist in their ability to confidently and correctly diagnose and manage TBM.
The terminology for TBM has been notoriously ambiguous. Twenty-three percent of respondents were unable to correctly define TBM. Clinicians may struggle to properly define this disorder due to the inconsistent use of terminology for describing disorders of expiratory airway collapse such as those historically grouped under TBM, trachea-bronchial collapse, tracheobronchial dyskinesia, dynamic airway collapse, or other terminologies.2 More recently, these conditions have been further categorized into 2 separate processes, namely TBM and EDAC,3 which recognizes and respects their different underlying pathophysiology. TBM classically results from acquired (ie, prolonged intubation, cancer, radiation, and relapsing polychondritis) or congenital disorders. Meanwhile, EDAC is often a consequence of obesity or obstructive lung disease.4
It is essential for clinicians to recognize the classic presentations of TBM to avoid missed diagnostic opportunities. In general, recurrent cough, dyspnea, or infections that have no other identifiable etiology should prompt evaluation. In addition, patients with COPD who experience frequent exacerbations or recurrent pneumonia despite maximal bronchodilator therapy may suffer from ECAC. Patients with known thoracic or airway trauma or connective tissue disease (ie, relapsing polychondritis) should be considered for TBM if clinically suspected.5
Not all providers execute a diagnostic evaluation for TBM in warranted situations. Many respondents consider TBM for the difficult-to-control COPD patients, yet not all proceed to diagnostic evaluation. Over 20% of respondents never consider TBM in patients with suggestive symptoms. Clinicians may be reluctant to initiate workup for TBM because of their unfamiliarity with its diagnostic evaluation.
The majority of respondents utilize PFTs to evaluate TBM and 38% do not pursue TBM further if PFTs were normal. PFT and flow-volume loop (FVL) have poor correlation with the severity of disease.6 A trend exists for patients with TBM to manifest obstructive defects on PFTs and low maximum forced expiratory flow on FVL; however, in 1 study evaluating 90 subjects with moderate to severe TBM, 21% had normal PFTs and 17% had normal FVLs,7 suggesting that a normal PFT and FVL cannot exclude TBM. Our respondent’s use of PFTs to definitively diagnose TBM suggests an unfamiliarity with this data.
Eighty percent of respondents do not diagnose TBM on dynamic inspiratory/expiratory CT. This is a reasonable response as any patient found to have signs of TBM on dynamic CT should ultimately undergo bronchoscopy to confirm the diagnosis. However, paired dynamic inspiratory/expiratory CT is useful to evaluate TBM and strongly correlates with bronchoscopic findings.8,9 One study found that dynamic inspiratory/expiratory CT detected the condition in 97% of patients with bronchoscopy-proven TBM.10 Clinicians’ awareness of dynamic CT’s ability to detect TBM, may promote the use of this noninvasive method to assess these patients.
Ninety percent of respondents diagnose TBM on dynamic bronchoscopy, the gold standard for diagnosis. This assessment allows for more accurate diagnosis and classification by the FEMOS scale, which combines qualitative assessment of Functional Class (F), Extent (E), Morphology (M), Origin (O), and Severity (S).2 Because of its multidimensional and detailed assessment, FEMOS should be considered as the most ideal classification system for TBM. However, only 6 respondents (4%), use FEMOS to grade severity of TBM. Respondents had difficulty diagnosing TBM and differentiating it from EDAC both on bronchoscopy and CT. The accurate diagnosis of TBM requires understanding of the morphologic differences between the 2 disorders. TBM tends to produce asymmetric collapse of the trachea and often is categorized as concentric, crescent like, or saber-sheath. EDAC only involves the posterior membrane of the trachea.11 Although most respondents differentiated the definitions for EDAC and TBM, a large majority had difficulty differentiating the 2 conditions on both bronchoscopic and CT images. This suggests a need to better educate providers on the proper interpretation of bronchoscopic and paired dynamic inspiratory/expiratory CT scan findings of these conditions.
An expected finding was that interventional pulmonologists had greater familiarity with the definitions, diagnostic workup, and morphologic characteristics of TBM. This reflects the increasing role interventional pulmonologists have played in evaluating and treating this condition, as well as other disorders of the central airways. Interestingly, of the respondents that would refer patients suspected of having TBM, 26% would refer to an ENT or thoracic surgeon. Although many specialties offer different areas of expertise within the airway, it is worth noting that interventional pulmonologists are often the best suited to assist in the holistic evaluation of TBM given their experience with both diagnosis and treatment, as well as their role in performing bronchoscopy.12
It is encouraging to find that physicians who consider themselves competent in diagnosing TBM are more likely to pursue a diagnosis in patients who warrant further investigation. However, physicians’ perception of competence in the diagnostic evaluation of TBM did not translate into their actual competence as was reflected by their ability to correctly interpret diagnostic studies. This exposes the opportunity to not only further raise awareness about this condition to promote its justified diagnostic evaluation, but to also improve upon educating those who already consider this diagnostic entity within their differential diagnosis.
We recognize that a self-reported survey may be limited by recall bias. Another limitation includes our relatively small study population that may have limited the power of the study, specifically where trends were noted. Still-image CT and bronchoscopy were used in the questionnaire and whether or not complete CT images/video bronchoscopy would yield different results is questionable and therefore a limitation of our study.
There is a paucity of literature regarding TBM; however, it is not rare with a prevalence of 4% to 12% cited in the adult population with different respiratory diseases.13 We believe that TBM is heavily underrecognized and therefore likely underdiagnosed by both clinicians and radiologists. This concept of underrecognition is highlighted by our ongoing retrospective radiologic study that has identified a 92% underrecognition of ECAC in combined inspiratory and end-expiratory CT by interpreting radiologists (W. Richter and I. Matus, unpublished data).
The overall deficits in knowledge revealed by our survey are not surprising given the fact that our understanding of this clinical entity is evolving and universally accepted diagnostic and therapeutic algorithms are lacking. It is evident that promotion of awareness and further education of pulmonologists about TBM is warranted in addition to collaborative efforts via research and expert consensus committees to progress our knowledge and management of this disease.
The authors would like to acknowledge the work of Eshetu Tefera who provided assistance with performing statistical analysis of our data. The authors acknowledge Septimiu Murgu for allowing us to use his bronchoscopic and CT images in our survey.
1. Murgu S, Colt H. Tracheobronchomalacia
and excessive dynamic airway collapse
. Respirology. 2006;11:388–406.
2. Murgu S, Colt H. Tracheobronchomalacia
and excessive dynamic airway collapse
. Clin Chest Med. 2013;34:527–555.
3. Choo E, Seaman J, Musani A. Tracheomalacia/tracheobronchomalacia
and hyperdynamic airway collapse. Immunol Allergy Clin N Am. 2013;33:23–34.
4. Boiselle P, Michaud G, Roberts D, et al.. Dynamic expiratory tracheal collapse in COPD. Chest. 2012;142:1539–1544.
5. Lawrence D, Branson B, Oliva I, et al.. The wonderful world of the windpipe: a review of central airway anatomy and pathology. Can Assoc Radiol J. 2015;66:30–43.
6. Majid A, Fernandez L, Fernandez-Bussy S, et al.. Tracheobronchomalacia
. Arch Bronconeumol. 2010;46:196–202.
7. Majid A, Sosa A, Ernst A, et al.. Pulmonary function and flow-volume loop patterns in patients with tracheobronchomalacia
. Respir Care. 2013;58:1521–1526.
8. Kugler C, Stanzel F. Tracheomalacia. Thorac Surg Clin. 2014;24:51–58.
9. Gilkeson R, Ciancibello L, Hejal R, et al.. Tracheobronchomalacia
: dynamic airway evaluation with multidetector CT. AJR. 2001;176:205–210.
10. Lee K, Sun M, Ernst A, et al.. Comparison of dynamic expiratory CT with bronchoscopy for diagnosing airway malacia. Chest. 2007;131:758–764.
11. Dal Negro R, Tognella S, Guerriero M, et al.. Prevalence of tracheobronchomalacia
and excessive dynamic airway collapse
in bronchial asthma of different severity. Multidiscip Respir Med. 2013;8:32.
12. Barros Casas D, Fernandez S, Folch E, et al.. Non-malignant central airway obstruction [Non-malignant central airway obstruction]. Arch Bronconeumol. 2014;50:345–354.
13. Ridge CA, O’donnell CR, Lee EY, et al.. Tracheobroncholmalacia: current concepts and controversies. J Thorac Imaging. 2011;26:278–289.
Keywords:Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
tracheobronchomalacia; excessive dynamic airway collapse; expiratory central airway collapse; interventional pulmonology