Paratracheal Paraseptal Emphysema and Expiratory Central Airway Collapse in Smokers

Abstract

Rationale: Expiratory central airway collapse is associated with respiratory morbidity independent of underlying lung disease. However, not all smokers develop expiratory central airway collapse, and the etiology of expiratory central airway collapse in adult smokers is unclear. Paraseptal emphysema in the paratracheal location, by untethering airway walls, may predispose smokers to developing expiratory central airway collapse.

Objectives: To evaluate whether paratracheal paraseptal emphysema is associated with expiratory central airway collapse.

Methods: We analyzed paired inspiratory and expiratory computed tomography scans from participants enrolled in a multicenter study (Genetic Epidemiology of Chronic Obstructive Pulmonary Disease) of smokers aged 45 to 80 years. Expiratory central airway collapse was defined as greater than or equal to 50% reduction in cross-sectional area of the trachea during expiration. In a nested case-control design, participants with and without expiratory central airway collapse were included in a 1:2 fashion, and inspiratory scans were further analyzed using the Fleischner Society criteria for presence of centrilobular emphysema, paraseptal emphysema, airway wall thickening, and paratracheal paraseptal emphysema (maximal diameter ≥ 0.5 cm).

Results: A total of 1,320 patients were included, 440 with and 880 without expiratory central airway collapse. Those with expiratory central airway collapse were older, had higher body mass index, and were less likely to be men or current smokers. Paratracheal paraseptal emphysema was more frequent in those with expiratory central airway collapse than control subjects (16.6 vs. 11.8%; P = 0.016), and after adjustment for age, race, sex, body mass index, smoking pack-years, and forced expiratory volume in 1 second, paratracheal paraseptal emphysema was independently associated with expiratory central airway collapse (adjusted odds ratio, 1.53; 95% confidence interval, 1.18-1.98; P = 0.001). Furthermore, increasing size of paratracheal paraseptal emphysema (maximal diameter of at least 1 cm and 1.5 cm) was associated with greater odds of expiratory central airway collapse (adjusted odds ratio, 1.63; 95% confidence interval, 1.18-2.25; P = 0.003 and 1.77; 95% confidence interval, 1.19-2.64; P = 0.005, respectively).

Conclusions: Paraseptal emphysema adjacent to the trachea is associated with expiratory central airway collapse. The identification of this risk factor on inspiratory scans should prompt further evaluation for expiratory central airway collapse. Clinical trial registered with ClinicalTrials.gov (NCT 00608764).

Figure 1.

Axial paired computed tomography images at (A) end-inspiration and (B) end-expiration showing expiratory central airway collapse at the level of the aortic arch in a representative subject. The cross-sectional area decreased 70% from end-inspiration to end-expiration.

Figure 2.

Axial inspiratory computed tomographic image at the level of the T4 vertebra of a representative study subject with substantial paratracheal paraseptal emphysema.

Figure 3.

(A) Adjusted odds ratios for the association between paratracheal paraseptal emphysema (paratracheal PSE) when defined as greater than or equal to 0.5 cm, greater than or equal to 1 cm, and greater than or equal to 1.5 cm in maximal diameter. All associations are adjusted for age, sex, race, body mass index, pack-years smoking history, current smoking status, and forced expiratory volume in 1 second, with Global Initiative for Obstructive Lung Disease stage as strata variable. (B) Relationship between paratracheal PSE as a continuous measure and the probability of expiratory central airway collapse (ECAC). The graph illustrates the positive linear slope (b = 0.018, P = 0.002) between paratracheal PSE and the predicted probability of ECAC obtained from the multivariable logistic model to predict the presence of ECAC by paratracheal PSE with adjustment for age, sex, race, body mass index, pack-years of smoking, current smoking status, and forced expiratory volume in 1 second. Each scattered dot represents the predicted probability of ECAC for a subject. The linear slope for the relationship between paratracheal PSE and ECAC is presented with 95% confidence interval (CI). Multivariable logistic model showed that a participant with 1-cm greater paratracheal PSE is 29% more likely to have ECAC.