Seminar: Increased Computed Tomography trachea surface roughness is associated with worse symptom burden in Chronic Obstructive Pulmonary Disease

Student: Jason Bartlett

Supervisor: Dr. Miranda Kirby

Abstract

Introduction:

Chronic Obstructive Pulmonary Disease (COPD) is an under-diagnosed condition which accounts for 5% of all deaths worldwide (1). In patients with chronic obstructive pulmonary disease (COPD) chronic cough is a hallmark symptom, with more than a third of sufferers indicating that plays a significant role in impacting their quality of life (2,3). Chronic cough in COPD is due to several different factors, one is a heighten cough sensitivity brought on by acute inflammation of epithelial layers within the trachea and upper airways (4). It is known from airflow simulations of the upper airways that the surface composition of the trachea has a significant impact on how air flows into and out of the lungs (5–7).Tracheal abnormalities as measured by the trachea index (TI) on CT have been previously reported in COPD (8–10) and shown to be associated with airflow limitation. Further, investigations focused on trachea narrowing/topology and its impact on symptom burden in COPD patients has been limited (11).  However, the CT TI measurement uses only a single slice representing the narrowest region of the trachea to quantify morphology, which provides limited topological information about the overall shape of the trachea surface. Therefore, new approaches for quantifying trachea topology are required.

Objective: Our objective is to develop a fractal measurement that can quantify changes in surface roughness and topology in COPD using CT imaging. More specifically, we hypothesize that CT fractal measurements generated from trachea segmentations will be able to distinguish between those with COPD and those without and will be significantly associated with lung function and COPD symptoms.

Methods: A total of 1238 subjects were selected from The Canadian Cohort Obstructive Lung Disease (CanCOLD) (12) to be used for airway surface Fractal Analysis. Subjects were grouped based on disease severity according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines. For each subject, full-inspiration CT images were acquired, then segmented, labelled, and analyzed, using VIDA software (VIDA Diagnostics Inc.). Trachea surface roughness was generated by extracting the trachea segmentation and surface height profiles, projecting them onto a 2D grayscale image, and applying fractal dimension analysis using a Differential Box Counting method. The change in trachea surface roughness (TR) was calculated as the percent difference between the fractal dimension (FD) of the trachea surface profile and the FD of an equivalently sized smooth surface. Subjects’ spirometry lung function measurements used were forced expiratory volume in one second (FEV₁) and the ratio of FEV₁ over forced vital capacity (FEV₁/FVC).

Results: The 1238 CanCOLD participants were grouped by COPD disease severity into, n=260 never-smokers without COPD, n=364 at risk (ever-smokers without COPD), n=351 mild COPD and n=263 moderate+ COPD. A significant increase in surface roughness (TR) were observed between the at risk group and both the mild (at risk 3.84±1.10% vs. mild 4.02±0.96%, p=0.021) and the moderate+ group (at risk 3.84±1.10% vs. moderate+ 4.08±0.82%, p<0.001) after adjusting for covariates and factors. In a model with TI, TR was significantly and independently associated with FEV₁ (p<0.001) and FEV₁/FVC (p<0.001). In another model with TI, TR was also significantly associated with SGRQ (p=0.006) and Chronic cough (p=0.026). In a model with LAA₉₅₀ and Pi10, TR was significantly and independently associated with FEV₁ (p=0.002) and FEV₁/FVC (p<0.001). Finally in another model with LAA₉₅₀ and Pi10, TR was significantly associated with SGRQ (p=0.011) and Chronic cough (p=0.049).

Conclusion: We demonstrated FD measurements reflect abnormal surface topology/surface roughness in COPD and are related to airflow limitations. Taken together, these results indicate that CT airway topology is a novel imaging biomarker. We believe that such a biomarker could be potentially useful in identifying pre- and post-treatment differences related to inflammation of the trachea airway surfaces.

References

1.        World Health Organization. Projections of mortality and causes of death, 2016 to 2060 [Internet]. World Health Organization. World Health Organization; 2018 [cited 2020 Mar 18]. Available from: https://www.who.int/healthinfo/global_burden_disease/projections/en/

2.        O’Donnell DE, Banzett RB, Carrieri-Kohlman V, Casaburi R, Davenport PW, Gandevia SC, et al. Pathophysiology of dyspnea in chronic obstructive pulmonary disease: A roundtable. Proc Am Thorac Soc. 2007;4(2):145–68.

3.        Smith J, Woodcock A. Cough and its importance in COPD. Int J COPD [Internet]. 2006 Aug;1(3):305–14. Available from: http://www.atypon-link.com/DMP/doi/abs/10.2147/copd.2006.1.3.305

4.        Choudry NB, Fuller RW, Pride NB. Sensitivity of the Human Cough Reflex: Effect of Inflammatory Mediators Prostaglandin E 2 , Bradykinin, and Histamine. Am Rev Respir Dis [Internet]. 1989 Jul;140(1):137–41. Available from: http://www.atsjournals.org/doi/abs/10.1164/ajrccm/140.1.137

5.        Heidarinejad G, Roozbahani MH, Heidarinejad M. Studying airflow structures in periodic cylindrical hills of human tracheal cartilaginous rings. Respir Physiol Neurobiol [Internet]. 2019 Aug;266(March):103–14. Available from: https://doi.org/10.1016/j.resp.2019.04.012

6.        Broday DM, Agnon Y. Asymmetric human lung morphology induce particle deposition variation. J Aerosol Sci [Internet]. 2007 Jul;38(7):701–18. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0021850207000894

7.        Zhang B, Qi S, Yue Y, Shen J, Li C, Qian W, et al. Particle Disposition in the Realistic Airway Tree Models of Subjects with Tracheal Bronchus and COPD. Biomed Res Int. 2018;2018.

8.        Lee HJ, Seo JB, Chae EJ, Kim N, Lee CW, Oh Y-M, et al. Tracheal morphology and collapse in COPD: Correlation with CT indices and pulmonary function test. Eur J Radiol [Internet]. 2011 Dec;80(3):e531–5. Available from: http://dx.doi.org/10.1016/j.ejrad.2010.12.062

9.        Eom JS, Lee G, Lee HY, Oh JY, Woo SY, Jeon K, et al. The relationships between tracheal index and lung volumeparameters in mild-to-moderate COPD. Eur J Radiol. 2013;82(12).

10.      Greene R. “Saber-sheath” trachea: relation to chronic obstructive pulmonary disease. Am J Roentgenol [Internet]. 1978 Mar 1;130(3):441–5. Available from: https://www.ajronline.org/doi/10.2214/ajr.130.3.441

11.      Gupta PP, Yadav R, Verma M, Agarwal D, Kumar M. Correlation between high-resolution computed tomography features and patients’ characteristics in chronic obstructive pulmonary disease. Ann Thorac Med. 2008;3(3):87.

12.      Bourbeau J, Tan WC, Benedetti A, Aaron SD, Chapman KR, Coxson HO, et al. Canadian Cohort Obstructive Lung Disease (CanCOLD): Fulfilling the Need for Longitudinal Observational Studies in COPD. COPD J Chronic Obstr Pulm Dis [Internet]. 2014 Apr 20;11(2):125–32. Available from: http://www.tandfonline.com/doi/full/10.3109/15412555.2012.665520