Consensus Statements on Deployment-Related Respiratory Disease, Inclusive of Constrictive Bronchiolitis

BackgroundThe diagnosis of constrictive bronchiolitis (CB) in previously deployed individuals, and evaluation of respiratory symptoms more broadly, presents considerable challenges, including using consistent histopathologic criteria and clinical assessments.Research QuestionWhat are the recommended diagnostic workup and associated terminology of respiratory symptoms in previously deployed individuals?Study Design and MethodsNineteen experts participated in a three-round modified Delphi study, ranking their level of agreement for each statement with an a priori definition of consensus. Additionally, rank-order voting on the recommended diagnostic approach and terminology was performed.ResultsTwenty-five of 28 statements reached consensus, including the definition of CB as a histologic pattern of lung injury that occurs in some previously deployed individuals while recognizing the importance of considering alternative diagnoses. Consensus statements also identified a diagnostic approach for the previously deployed individual with respiratory symptoms, distinguishing assessments best performed at a local or specialty referral center. Also, deployment-related respiratory disease (DRRD) was proposed as a broad term to subsume a wide range of potential syndromes and conditions identified through noninvasive evaluation or when surgical lung biopsy reveals evidence of multicompartmental lung injury that may include CB.InterpretationUsing a modified Delphi technique, consensus statements provide a clinical approach to possible CB in previously deployed individuals. Use of DRRD provides a broad descriptor encompassing a range of postdeployment respiratory findings. Additional follow-up of individuals with DRRD is needed to assess disease progression and to define other features of its natural history, which could inform physicians better and lead to evolution in this nosology. The diagnosis of constrictive bronchiolitis (CB) in previously deployed individuals, and evaluation of respiratory symptoms more broadly, presents considerable challenges, including using consistent histopathologic criteria and clinical assessments. What are the recommended diagnostic workup and associated terminology of respiratory symptoms in previously deployed individuals? Nineteen experts participated in a three-round modified Delphi study, ranking their level of agreement for each statement with an a priori definition of consensus. Additionally, rank-order voting on the recommended diagnostic approach and terminology was performed. Twenty-five of 28 statements reached consensus, including the definition of CB as a histologic pattern of lung injury that occurs in some previously deployed individuals while recognizing the importance of considering alternative diagnoses. Consensus statements also identified a diagnostic approach for the previously deployed individual with respiratory symptoms, distinguishing assessments best performed at a local or specialty referral center. Also, deployment-related respiratory disease (DRRD) was proposed as a broad term to subsume a wide range of potential syndromes and conditions identified through noninvasive evaluation or when surgical lung biopsy reveals evidence of multicompartmental lung injury that may include CB. Using a modified Delphi technique, consensus statements provide a clinical approach to possible CB in previously deployed individuals. Use of DRRD provides a broad descriptor encompassing a range of postdeployment respiratory findings. Additional follow-up of individuals with DRRD is needed to assess disease progression and to define other features of its natural history, which could inform physicians better and lead to evolution in this nosology. Take-home PointsStudy Question: What is the recommended diagnostic workup and associated terminology of possible constrictive bronchiolitis (CB) or potentially related symptoms in previously deployed individuals?Results: Using a modified Delphi technique, an expert multidisciplinary panel achieved consensus on statements pertaining to the clinical presentation and evaluation of unexplained respiratory symptoms in previously deployed individuals. This included a definition of CB and recommendation of using deployment-related respiratory disease when referring to the broad set of respiratory symptoms or conditions observed after deployment, as well as for those who remain undiagnosed after a minimally invasive workup.Interpretation: Evaluating unexplained respiratory symptoms requires a systematic approach and consistent terminology to advance the health and care of previously deployed individuals. Study Question: What is the recommended diagnostic workup and associated terminology of possible constrictive bronchiolitis (CB) or potentially related symptoms in previously deployed individuals? Results: Using a modified Delphi technique, an expert multidisciplinary panel achieved consensus on statements pertaining to the clinical presentation and evaluation of unexplained respiratory symptoms in previously deployed individuals. This included a definition of CB and recommendation of using deployment-related respiratory disease when referring to the broad set of respiratory symptoms or conditions observed after deployment, as well as for those who remain undiagnosed after a minimally invasive workup. Interpretation: Evaluating unexplained respiratory symptoms requires a systematic approach and consistent terminology to advance the health and care of previously deployed individuals. More than 10 years ago, a case series was published describing symptomatic military personnel previously deployed to Southwest Asia referred for evaluation of unexplained dyspnea.1King M.S. Eisenberg R. Newman J.H. et al.Constrictive bronchiolitis in soldiers returning from Iraq and Afghanistan.N Engl J Med. 2011; 365: 222-230Crossref PubMed Scopus (194) Google Scholar Lung biopsies in 38 of 49 patients in this series were interpreted as manifesting features consistent with constrictive bronchiolitis (CB). Despite military personnel experiencing respiratory symptoms (eg, dyspnea, exercise intolerance, cough, inability to pass a military fitness test), other objective findings were limited to subtle impairments and abnormalities or normal cardiopulmonary function and chest imaging findings. This report garnered considerable attention within the medical and scientific community as well as the US Congress and news media. The active debate and discussion that has continued is summarized in both an American Thoracic Society workshop report2Garshick E. Abraham J.H. Baird C.P. et al.Respiratory health after military service in Southwest Asia and Afghanistan. An official American Thoracic Society workshop report.Ann Am Thorac Soc. 2019; 16: e1-e16Crossref PubMed Scopus (48) Google Scholar and a review by the National Academies of Sciences, Engineering and Medicine.3National Academies of SciencesEngineering and Medicine. Respiratory Health Effects of Airborne Hazards Exposures in the Southwest Asia Theater of Military Operations. National Academies Press, 2020Google Scholar Follow-up data specific to the questions of CB have been limited. In the largest study to date of postdeployment pulmonary pathologic diagnoses, an increase in the frequency of CB in previously deployed individuals seems to be present.4Lewin-Smith M.R. Martinez A. Brooks D.I. Franks T.J. Pulmonary pathology diagnoses in the US military during the global war on terrorism.Lung. 2021; 199: 345-355Crossref PubMed Scopus (5) Google Scholar Nonetheless, consensus around terminology as well as the optimal diagnostic approach to postdeployment respiratory symptoms remains unresolved. Several fundamental challenges exist that have impeded progress in the study of CB. First, bronchiolar disorders in general, and CB in particular, resulting from occupational and environmental exposures are uncommon and may occur after a variety of acute or indolent exposures.5Cummings K.J. Kreiss K. Occupational and environmental bronchiolar disorders.Semin Respir Crit Care Med. 2015; 36: 366-378Crossref PubMed Scopus (20) Google Scholar Second, CB has been associated with multiple clinical scenarios, including autoimmune or inflammatory bowel disease, lung or stem cell transplantation, and as a complication of certain medications, infections, or lymphoproliferative disorders. Third, CB is a pathologic diagnosis necessitating lung tissue obtained through invasive procedures (eg, surgical lung biopsy) likely to preclude study using a standard case referent design. Further, histopathologic evaluations of lung biopsy samples among previously deployed individuals demonstrate a wide spectrum of findings beyond CB.1King M.S. Eisenberg R. Newman J.H. et al.Constrictive bronchiolitis in soldiers returning from Iraq and Afghanistan.N Engl J Med. 2011; 365: 222-230Crossref PubMed Scopus (194) Google Scholar,4Lewin-Smith M.R. Martinez A. Brooks D.I. Franks T.J. Pulmonary pathology diagnoses in the US military during the global war on terrorism.Lung. 2021; 199: 345-355Crossref PubMed Scopus (5) Google Scholar,6Gutor S.S. Richmond B.W. Du R.H. et al.Postdeployment respiratory syndrome in soldiers with chronic exertional dyspnea.Am J Surg Pathol. 2021; 45: 1587-1596Crossref PubMed Scopus (12) Google Scholar,7Krefft S.D. Wolff J. Zell-Baran L. et al.Respiratory diseases in post-9/11 military personnel following Southwest Asia deployment.J Occup Environ Med. 2020; 62: 337-343Crossref PubMed Scopus (22) Google Scholar Even in the face of these challenges, because some previously deployed individuals with otherwise unexplained dyspnea do exhibit the histopathologic findings of CB, addressing the question of that entity in the wider context of small airways disease remains important. Expert consensus regarding the diagnostic approach to dyspnea and the cause and definition of CB could benefit patients, clinicians, researchers, and policy makers by advancing the field of small airways disease and other respiratory conditions in previously deployed individuals. The present modified Delphi study was motivated by recommendations from the National Academies of Sciences, Engineering and Medicine consensus study report,3National Academies of SciencesEngineering and Medicine. Respiratory Health Effects of Airborne Hazards Exposures in the Southwest Asia Theater of Military Operations. National Academies Press, 2020Google Scholar which broadly reviewed the scientific evidence on respiratory health outcomes in those previously deployed to the Southwest Asia region and Afghanistan. To this end, we convened a panel of clinical and research experts from academic centers, the US Department of Veterans Affairs, and the Department of Defense to arrive at a consensus on a variety of statements centered around the diagnostic approach to evaluating unexplained respiratory symptoms in previously deployed individuals. We used a modified Delphi technique to achieve consensus on a clinical approach to the diagnosis and management of respiratory conditions previously reported in case series among previously deployed individuals (Fig 1). The study was designed in accordance with reporting standards for Delphi studies8Junger S. Payne S.A. Brine J. Radbruch L. Brearley S.G. Guidance on Conducting and REporting DElphi Studies (CREDES) in palliative care: recommendations based on a methodological systematic review.Palliat Med. 2017; 31: 684-706Crossref PubMed Scopus (540) Google Scholar,9Diamond I.R. Grant R.C. Feldman B.M. et al.Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies.J Clin Epidemiol. 2014; 67: 401-409Abstract Full Text Full Text PDF PubMed Scopus (1310) Google Scholar by a steering committee and conducted using a web-based video platform. Initial activities through the final round of voting took place from November 2021 through February 2022. The initial design and the recruitment of experts were developed by the steering committee in consultation with collaborators. Pathologists, radiologists, pulmonologists, thoracic surgeons, and environmental and occupational medicine physicians were recruited specifically from both academia and federal agencies within the United States and were invited to participate via e-mail. An a priori goal of 12 final panelists was targeted across specialties, taking into account potential attrition. Potential panel members were excluded if they were unable to commit to participating in all components of this project. The study panel chair (J. M. D.) was recruited from an academic medical center based on clinical expertise in rare lung disease and previous experience leading multidisciplinary panels. To support development of statements, all panelists first attended a large group (videoconference) meeting followed by parallel small group (videoconference) meetings within 3 weeks of the initial meeting. The former was attended by the full panel and the latter were attended by six or fewer panelists per group. All meetings were facilitated by the study chair and steering committee, and a full transcription of the meetings was distributed to the panelists for review. In addition, supporting materials solicited from panelists (eg, publications) also were shared and distributed among the panel. The steering committee aggregated subject area content across pre-Delphi survey development activities (e-Appendix 1) to generate an initial list of 28 separate statements for initial consideration (round 1). Panelists completed surveys online using an electronic survey platform in which participants rated their agreement with statements using an 11-point Likert scale from 0 (strongly disagree) to 10 (strongly agree). An a priori definition of consensus was defined as ≥ 70% of panelists agreeing with a statement, a common approach used in Delphi studies,9Diamond I.R. Grant R.C. Feldman B.M. et al.Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies.J Clin Epidemiol. 2014; 67: 401-409Abstract Full Text Full Text PDF PubMed Scopus (1310) Google Scholar which we operationally achieved by selecting 8, 9, or 10 on the Likert scale. Panelists were also able to provide written comments on statements that were used to refine statements in subsequent rounds and to improve clarity. A deployed individual was defined operationally as any active or former military personnel, including contractors, previously deployed to the combat theater. In addition, undiagnosed respiratory condition was used to refer to a symptomatic deployed individual who, after a noninvasive evaluation, did not have a condition that met currently accepted diagnostic criteria for a respiratory disease. During pre-Delphi activities as well as during rounds of voting, the panel developed and refined a list of diagnostic assessments to be used in the evaluation of previously deployed individuals with persistent respiratory symptoms. This discussion included consideration of two components: when and where diagnostic assessments should best be performed. To address the first component, panelists were asked to rank order assessments across three levels. Levels 1 through 3 reflected basic, intermediate, or advanced evaluation, respectively, and were modeled after previously published algorithms in this population.10Krefft S.D. Meehan R. Rose C.S. Emerging spectrum of deployment-related respiratory diseases.Curr Opin Pulm Med. 2015; 21: 185-192Crossref PubMed Scopus (18) Google Scholar, 11Morris M.J. Grbach V.X. Deal L.E. Boyd S.Y. Morgan J.A. Johnson J.E. Evaluation of exertional dyspnea in the active duty patient: the diagnostic approach and the utility of clinical testing.Mil Med. 2002; 167: 281-288Crossref PubMed Scopus (56) Google Scholar, 12Morris M.J. Lucero P.F. Zanders T.B. Zacher L.L. Diagnosis and management of chronic lung disease in deployed military personnel.Therapeutic Advances in Respiratory Disease. 2013; 7: 235-245Crossref PubMed Scopus (17) Google Scholar, 13Rose C. Abraham J. Harkins D. et al.Overview and recommendations for medical screening and diagnostic evaluation for postdeployment lung disease in returning US warfighters.J Occup Environ Med. 2012; 54: 746-751Crossref PubMed Scopus (38) Google Scholar Separately, panelists also were asked to indicate whether corresponding assessments should be conducted at a standard facility or specialty referral center. The operational definition of a standard facility included a federal government hospital or community teaching hospital14Liu J.B. Kelz R.R. Types of hospitals in the United States.JAMA. 2018; 320: 1074Crossref PubMed Scopus (12) Google Scholar and specialty referral center and institute or similar entity housed within a federal or nonfederal hospital system with the resources (personnel expertise and equipment) and infrastructure to offer multimethod assessments specific to individuals with respiratory symptoms. For each assessment, panelists also could indicate whether the assessment was unnecessary or fell outside of the scope of their expertise. Assignment of diagnostic assessments was based on a simple majority (≥ 50%) of rank ordering. During pre-Delphi activities, discussions addressed terminology and case definitions that included a general exchange of views regarding importance to the field and the potential impact on clinical care and research activities of having agreed on nosology and terminologies. This included a preferred term for both: (1) a broad set of respiratory conditions that might be observed in previously deployed individuals, that is, a general name,7Krefft S.D. Wolff J. Zell-Baran L. et al.Respiratory diseases in post-9/11 military personnel following Southwest Asia deployment.J Occup Environ Med. 2020; 62: 337-343Crossref PubMed Scopus (22) Google Scholar and (2) respiratory conditions that remained undiagnosed after a comprehensive noninvasive clinical evaluation, that is, a specific condition name. During the Delphi rounds of voting, the panel had the opportunity to rank order terms or names (randomly ordered) that were proposed by the panelists (e-Appendix 1, e-Tables 1, 2). The preferred name (general or specific) was defined as that which was ranked highest by most panelists. Survey responses from each round were summarized using descriptive statistics (GraphPad Prism version 9.3.1 software [GraphPad Software]). For each of the final consensus statements, we report the median and 25th and 75th percentiles of the Likert scale responses. A total of 21 experts, including the study chair (Table 1), were invited to participate in this Delphi process, with 19 experts (seven more than the initial target number) completing the assessment (Fig 1). Desired specialties were represented with the exception of thoracic surgery. Dates of participation, average time for completion of each survey, and rates of participation are reported in Tables 1 and 2. Twenty-eight final statements across rounds 2 and 3 resulted, of which 25 statements ultimately reached consensus as defined (89.3%) (Table 3). Panelists also separately rank ordered diagnostic testing across three levels of complexity as well as the location of where testing should be conducted (Fig 2, e-Tables 3, 4).Table 1Characteristics of the Delphi Panel (n = 19)Expertise and ExperienceNo. of PanelistsaMissing response from one panelist.SpecialtybSome panelists had more than one subspecialty area. Pulmonology10 Environmental and occupational medicine6 Critical care4 Pathology3 Radiology3Experience, y evaluating deployed individuals < 53 5-107 11-194 20+4Evaluations, no. of deployed individuals in entire career 0-998 100-4997 500-1,0001 1,000+2Cases of constrictive bronchiolitis (suspected or probable) 04 1-397 40-993 100+4a Missing response from one panelist.b Some panelists had more than one subspecialty area. Open table in a new tab Table 2Characteristics of the Delphi Survey RoundsCharacteristicRound 1Round 2Round 3Survey dates12/19/21-12/27/211/15/22-1/25/222/2/22-2/8/22ParticipationaParticipation percentage is calculated as a percent of total panelists (n = 19).17 (89)18 (95)18 (95)Completion94100100Average time to complete, min61.2 ± 60.887.5 ± 105.241.4 ± 58.6No. of statements283414Data are presented as No. (%), percentage, or mean ± SD, unless otherwise indicated.a Participation percentage is calculated as a percent of total panelists (n = 19). Open table in a new tab Table 3Consensus Statement Results by Round and Question, Organized Into DomainsRoundQuestionMedian (Interquartile Range)StatementsClinical presentation of undiagnosed respiratory symptoms229 (8-9.75)There are a spectrum of known respiratory conditions that are associated with deployment to the Southwest Asia region, such as rhinitis, rhinosinusitis, and asthma. There are also deployment-related undiagnosed respiratory conditions that may include CB.239 (9-10)These undiagnosed respiratory conditions are found in, but not limited to, those previously deployed to Southwest Asia and Afghanistan.249 (7-10)These undiagnosed respiratory conditions present with signs and symptoms that are not exclusive to the small airways.259 (7.75-9.25)These undiagnosed respiratory conditions present with abnormalities attributable to pathology in multiple compartments within the lung including the small airways.269 (8-10)There are multiple exposures during deployment (eg, such as burn pits and other sources of VGDF [vapors, gases, dust, and fumes]) that are associated with these respiratory conditions.279 (6.5-9.75)Individuals with these undiagnosed respiratory conditions present with persistent respiratory symptoms, eg, unexplained shortness of breath, decreased exercise tolerance, and/or cough.299.5 (9-10)Respiratory symptoms may start during or after deployment and often continue to persist.Clinical evaluation of undiagnosed respiratory symptoms2810 (9-10)Individuals presenting with these undiagnosed respiratory symptoms should undergo a comprehensive occupational and environmental exposure history. Special consideration should be given to inhalational exposures during deployment, including vapors, gases, dust, and fumes.21210 (9-10)It is important to rule out other contributing factors and/or comorbid conditions including, but not limited to, asthma, sinusitis/rhinitis/rhinosinusitis, GERD, cardiac factors, laryngeal disorders, and anemia.21310 (8.25-10)It is highly recommended that there should be a standardized protocol to be followed in both standard facilities and specialty referral facilities.3139.5 (8.75-10)With the proper equipment and training, standard facilities and specialty referral facilities may be able to assess and treat some well-described postdeployment respiratory conditions, including chronic rhinitis, sinusitis, and asthma. An inadequate response to treatment and/or the persistence of additional symptoms or abnormal test results should consider prompt referral to a specialty facility knowledgeable in postdeployment respiratory health.328 (8-9)Surgical lung biopsy may be considered, but not mandatory, when noninvasive or minimally invasive diagnostic procedures do not yield a diagnosis and when there is a high clinical suspicion based on patient history, imaging, and/or PFT findings of diagnosis but not certainty.339 (8-10)Surgical lung biopsies (for undiagnosed respiratory conditions) should be obtained only at specialty centers and reviewed at a specialty center by an experienced pulmonary pathologist. Biopsy specimens must be properly prepared samples per recognized pathology guidelines.348 (5-9)To advance our understanding of undiagnosed respiratory conditions, surgical lung biopsy samples obtained from deployed individuals should undergo quantitative analysis.369 (7.25-10)In order to advance our understanding of undiagnosed respiratory conditions, surgical lung biopsies within the setting of a clinical trial would be helpful.CB359 (8-10)Pathological review and analysis of surgical lung biopsy samples in deployed individuals must move beyond diagnosis of CB in order to advance the field.378 (4.25-8)Surgical lung biopsy is necessary for the diagnosis of CB, otherwise you may have a high degree of suspicion, but not a definitive diagnosis.389 (8-10)CB is a histological pattern observed in small airways that is observed in some deployed individuals.399 (8-10)CB is a histological pattern of lung injury characterized by subepithelial fibrosis of the small airways that narrows and sometimes obliterates bronchiolar lumens.3109 (8-10)The longitudinal behavior of CB and other undiagnosed respiratory conditions in deployed individuals is not well characterized.3118 (6.5-9)At this time, the longitudinal progression of CB in deployed individuals appears to be slower than those among lung transplant patients.3149 (8-9)There is no proven treatment for individuals with CB; however, respiratory symptoms may be improved by managing comorbidities and/or pulmonary rehabilitation.2199 (7.25-10)There are many different respiratory conditions that are found in deployed individuals, including CB, but more data are required to determine what the prevalence of CB is.2239 (8-10)Histopathological evidence of CB on surgical lung biopsy may account for respiratory symptoms in deployed individuals.Recommended nosology and terminology22010 (8-10)Respiratory conditions that are found in deployed and postdeployed individuals should not be referred to as CB until proper diagnosis and/or testing is done in the appropriate clinical setting.22910 (9-10)Use of common terms and descriptors for these undiagnosed respiratory conditions as well as consistent evaluation approaches requires clear communication to patients and amongst their providers. This should be achieved through training and the provision of educational resources.2309 (7.25-10)It would be helpful to have a term initially to name the broad set of respiratory conditions, known and unknown, of a deployed individual, which could include deployment-related rhinitis, rhinosinusitis, asthma, CB, small airways disease or pleural disease, or various forms of ILD.2339 (7.25-10)It would be helpful to have a term to name the respiratory conditions that remain undiagnosed after a comprehensive noninvasive/minimally invasive workup.Boldface statements (n = 3) did not reach consensus. CB = constrictive bronchiolitis; GERD = gastroesophageal reflux disease; ILD = interstitial lung disease; PFT = pulmonary function test. Open table in a new tab Data are presented as No. (%), percentage, or mean ± SD, unless otherwise indicated. Boldface statements (n = 3) did not reach consensus. CB = constrictive bronchiolitis; GERD = gastroesophageal reflux disease; ILD = interstitial lung disease; PFT = pulmonary function test. Final statements with their associated statistics are presented in Table 3 and are grouped into the following categories: (1) clinical presentation of undiagnosed respiratory symptoms; (2) clinical evaluation of undiagnosed respiratory symptoms; (3) definition, diagnostic approaches, and treatment of CB; and (4) recommended nosology and terminology. Panelists reached consensus on all seven statements regarding clinical presentation of symptomatic previously deployed individuals (Table 3). This included agreement that a range of respiratory symptoms exists that could be associated with a variety of exposures and recognition that, despite other known conditions, respiratory conditions remain undiagnosed that could include CB. These undiagnosed respiratory conditions could be attributable to multiple compartments of the lung, including the small airways. Importantly, the symptomatic presentation of these conditions is not restricted to dyspnea. The results for the statements and associated rank ordering of diagnostic procedures and assessments for the previously deployed individual with unexplained respiratory symptoms were assimilated into Figure 2, which stratifies the diagnostic workup by the characteristics of the health-care center performing the requested test (standard hospital or specialty referral center). Within this category of statements, consensus was reached on seven of eight statements regarding the clinical evaluation of symptomatic individuals. Panelists agreed that a comprehensive assessment of exposure should be undertaken and that comorbidities should be assessed. These evaluations may take place at any medical facility; however, previously deployed individuals with persistent symptoms without a clear diagnosis should be referred to a specialty facility with expertise in postdeployment health. The panel agreed that under certain circumstances, noninvasive testing may not yield a diagnosis, and yet the clinical suspicion of underlying lung pathologic characteristics remains high. Under such circumstances, the panel agreed that surgical lung biopsy (eg, via video-assisted thoracoscopic surgery or other techniques) obtained at specialty centers and reviewed by experienced pulmonary pathologists should be considered. Consensus was not reached on whether to use quantitative histopathologic analysis (eg, molecular pathology) as opposed to standard qualitative histopathologic analysis. However, the panel agreed that surgical lung biopsy could occur as part of a clinical trial, such as when evaluating new diagnostic methods or to establish the diagnosis of CB before enrollment in a longitudinal observational or treatment trial. A total of 19 diagnostic assessments were proposed as part of a comprehensive evaluation of a previously deployed individual with persistent respiratory symptoms (e-Table 3), with 17 assessments reaching majority consensus opinion across three levels of increasing com