To the Editor:
Chronic obstructive pulmonary disease (COPD) and asthma are major sources of morbidity and mortality. Inhaled bronchodilators are a cornerstone of disease management, but treatment is not equally effective in all patients (1, 2). Identifying factors associated with bronchodilator efficacy may expand opportunities for precision medicine and, critically, identify patient subgroups for which new therapies are urgently needed.
The airway tree is the target of inhaled therapies. Using computed tomography (CT), we have demonstrated that variation in native airway tree caliber relative to lung size (i.e., dysanapsis) is common in the general population (3), extends to the peripheral airways (4), and is associated with COPD risk (3). Although computational fluid dynamic studies of particle deposition (5) suggest that dysanapsis may modify inhaled pharmacotherapy efficacy, in vivo evidence is lacking.
This study evaluated whether dysanapsis quantified by CT is associated with the spirometric response to a standardized dose of inhaled bronchodilator.
Methods
Participants.
The COLD (Canadian Obstructive Lung Disease) prevalence study used census data to recruit a random sample of noninstitutionalized adults ⩾40 years old from nine communities (2005–2009). In 2010–2014, the ongoing CanCOLD (Canadian Cohort Obstructive Lung Disease) study enrolled COLD participants with COPD, and representative random subsets of COLD nonsmoking and smoking participants without COPD matched on age and sex (6). Data from the baseline CanCOLD visit were included in this analysis. Institutional review board approval was obtained, and all participants provided written informed consent.
CT assessment of dysanapsis.
Full-inspiration CT was performed on helical scanners according to a standardized protocol. Airway lumen diameters at 19 standard anatomic locations (trachea-to-subsegments) and total lung volume were measured from CT images using Apollo Software (VIDA Diagnostics) by trained readers unaware of other participant information. Trained readers achieved a minimum training set interrater intraclass correlation of 0.9 (3).
Dysanapsis was quantified as the mean of airway lumen diameters in centimeters divided by the cube root of total lung volume in cubic centimeters (airway-to-lung ratio). Lower values of airway-to-lung ratio indicate smaller airway tree-to-lung size, and higher values indicate larger airway tree-to-lung size. Secondary dysanapsis measures were percent-predicted airway tree caliber calculated from reference equations that account for age, sex, height, and lung volume (3), and the mean of airway lumen diameters in centimeters.
Lung function.
Spirometry was performed before and 15 minutes after inhalation of 200 μg of salbutamol (albuterol) that was administered from a metered-dose inhaler with spacer device (100 μg/actuation) (7). Participants were instructed to not use regular inhaled medications 6–24 hours before spirometry depending on medication class. Spirometric COPD was defined by post-bronchodilator FEV1/FVC <0.7 (1). Bronchodilator-associated change in FEV1 (ΔFEV1) and FEV1/FVC (ΔFEV1/FVC) were calculated as post-bronchodilator minus prebronchodilator values.
Other variables.
Age, sex, cigarette smoking status, and regular inhaled medication use were self-reported. Clinical diagnoses of asthma and COPD corresponded to an affirmative response to questions of whether a healthcare provider told you that you have, respectively, asthma, or COPD, chronic bronchitis, or emphysema. The percentage of emphysema-like lung was quantified as the lung volume below −950 Hounsfield units (3).
Statistical analysis.
A linear regression model of ΔFEV1 was fit with airway-to-lung ratio as the independent variable of interest. The main model adjusted for age, age2, sex, height, height2, cigarette smoking status, pack-years, asthma diagnosis, COPD diagnosis, percent emphysema, and regular inhaler medication use. Secondary analyses were restricted to nonsmokers; participants without diagnosis of asthma or COPD; participants with diagnosis of asthma, with diagnosis of COPD, and with spirometry-defined COPD; and replacing airway-to-lung ratio with percent-predicted airway tree caliber, with mean airway lumen diameter, and replacing ΔFEV1 with ΔFEV1/FVC. A two-sided P value <0.05 was considered statistically significant (SAS 9.4; SAS Institute).
Results
Table 1 summarizes participant characteristics by airway-to-lung ratio quartile. Among the 1,272 participants (mean ± SD age, 67 ± 10 years; 44.3% female; 93.8% non-Hispanic White), the airway-to-lung ratio was 0.032 ± 0.003 and a standardized dose of inhaled short-acting β2-agonist was associated with ΔFEV1 of 96 ± 157 ml. Participants with smaller airway-to-lung ratio were more likely to report a diagnosis of asthma or COPD.
Table 1.
Participant Characteristics by Airway-to-Lung Ratio Quartile
| Quartile of Airway-to-Lung Ratio |
||||
|---|---|---|---|---|
| Q1 (Smallest Airway-to-Lung Ratio) | Q2 | Q3 | Q4 (Largest Airway-to-Lung Ratio) | |
| No. of participants | 318 | 318 | 318 | 318 |
| Age, yr | 65 ± 10 | 65 ± 8 | 66 ± 10 | 70 ± 10 |
| Sex, n (%) | ||||
| F | 166 (52.1) | 153 (48.2) | 138 (43.2) | 123 (38.6) |
| M | 152 (47.9) | 165 (51.9) | 181 (56.8) | 195 (61.4) |
| Height, cm | 169 ± 10 | 168 ± 8 | 168 ± 10 | 167 ± 10 |
| Body mass index, kg/m2 | 28 ± 6 | 28 ± 5 | 27 ± 5 | 28 ± 5 |
| Race/ethnicity, n (%) | ||||
| Non-Hispanic White | 302 (94.9) | 306 (96.2) | 292 (92.1) | 296 (93.2) |
| Non-Hispanic Black | 5 (1.5) | 3 (0.8) | 3 (0.8) | 5 (1.7) |
| Non-Hispanic Chinese | 10 (3.2) | 5 (1.6) | 18 (5.61) | 10 (3.1) |
| Hispanic | 0 (0) | 0 (0) | 2 (0.6) | 2 (0.6) |
| Other | 1 (0.4) | 5 (1.4) | 4 (1.1) | 5 (1.5) |
| Cigarette smoking status, n (%) | ||||
| Never | 144 (45.4) | 168 (52.9) | 170 (53.5) | 163 (51.1) |
| Former | 123 (38.6) | 110 (34.5) | 111 (34.8) | 130 (41.0) |
| Current | 51 (16.0) | 40 (12.5) | 37 (13.7) | 25 (9.9) |
| Pack-years among smoking participants, median (IQR) | 22 (6–44) | 20 (10–33) | 15 (5–32) | 19 (9–37) |
| Asthma diagnosis ever, n (%) | 83 (26.1) | 68 (21.5) | 49 (15.4) | 42 (13.2) |
| COPD diagnosis ever, n (%) | 51 (15.9) | 32 (10.0) | 35 (11.1) | 24 (7.6) |
| Regular inhaled medication use, n (%) | 65 (20.6) | 46 (14.3) | 44 (13.7) | 39 (12.2) |
| Spirometry | ||||
| Prebronchodilator FEV1, ml | 2,333 ± 598 | 2,603 ± 725 | 2,720 ± 843 | 2,570 ± 785 |
| Prebronchodilator % predicted FEV1 | 77.1 ± 20.1 | 86.6 ± 19.2 | 92.0 ± 20.5 | 94.4 ± 21.0 |
| Prebronchodilator FVC, ml | 3,604 ± 824 | 3,730 ± 996 | 3,743 ± 1,108 | 3,462 ± 1,024 |
| Prebronchodilator % predicted FVC | 96.4 ± 18.1 | 99.2 ± 16.8 | 99.6 ± 18.3 | 97.4 ± 18.8 |
| Prebronchodilator FEV1/FVC | 0.65 ± 0.08 | 0.70 ± 0.07 | 0.73 ± 0.07 | 0.74 ± 0.08 |
| Bronchodilator-associated ΔFEV1, ml | 155 ± 115 | 105 ± 122 | 96 ± 180 | 61 ± 186 |
| Bronchodilator-associated ΔFEV1/FVC | 0.035 ± 0.029 | 0.031 ± 0.043 | 0.028 ± 0.035 | 0.026 ± 0.045 |
| CT measures of dysanapsis | ||||
| Airway-to-lung ratio* | 0.028 ± 0.001 | 0.030 ± 0.001 | 0.032 ± 0.001 | 0.036 ± 0.002 |
| Percent-predicted airway tree caliber† | 82.1 ± 5.8 | 90.4 ± 4.6 | 96.6 ± 4.7 | 105.0 ± 7.0 |
| Mean of airway lumen diameters, cm‡ | 0.49 ± 0.04 | 0.53 ± 0.04 | 0.57 ± 0.05 | 0.61 ± 0.06 |
| Percent-predicted total lung volume§ | 97 ± 14 | 95 ± 15 | 92 ± 15 | 87 ± 16 |
| Percent emphysema-like lung, median (IQR)ǁ | 1.0 (0.3–1.6) | 0.8 (0.1–1.5) | 0.8 (0.2–1.4) | 0.8 (0.05–1.33) |
Definition of abbreviations: COPD = chronic obstructive lung disease; CT = computed tomography; IQR = interquartile range; Q = quartile.
The mean of airway lumen diameters in centimeters measured at 19 standard anatomic locations divided by the cube root of lung volume in cubic centimeters. The 19 standard anatomic airways were trachea, right mainstem, left mainstem, bronchus intermedius, right upper lobe, right middle lobe, right lower lobe, left upper lobe, left lower lobe, RB1, RB4, RB10, LB1, and LB10 bronchi, as well as the average airway lumen diameters of the subsegments along each of these segmental paths (sRB1, sRB4, sRB10, sLB1, and sLB10).
The percent-predicted airway tree caliber for each participant was quantified as the mean of percent-predicted airway lumen diameters measured at 19 standard anatomic locations defined by sex-stratified airway-specific lumen diameter reference equations with terms for total lung volume, age, and height.
The mean of airway lumen diameters (mm) measured at 19 standard anatomic locations.
The percent-predicted total lung volume was calculated as the total lung volume measured at inspiratory CT divided by the predicted TLC (12) multiplied by 100%.
The percentage of lung volume below −950 Hounsfield units on inspiratory CT.
Participants in the smallest quartile of airway-to-lung ratio exhibited significantly greater ΔFEV1 than those in the largest airway-to-lung ratio quartile (adjusted mean difference, 86 ml; 95% confidence interval [CI], 60–113 ml; P < 0.001). Differences were consistent in unadjusted and adjusted models and various subgroups (Table 2), and when dysanapsis was quantified as percent-predicted airway tree caliber or mean airway lumen diameter (P < 0.001), and when ΔFEV1 was replaced by ΔFEV1/FVC (adjusted mean difference, 0.012; 95% CI, 0.006–0.018; P < 0.001).
Table 2.
Dysanapsis Quantified as the Airway-to-Lung Ratio by CT and the Spirometric Response to a Standardized Dose of Inhaled Short-Acting Bronchodilator
| Mean (95% CI) ΔFEV1, ml |
Mean (95% CI) Difference in ΔFEV1 between the Lowest and Highest Quartile of Airway-to-Lung Ratio (ml), P Value | |||
|---|---|---|---|---|
| Entire Stratum | Participants in the Lowest Quartile of Airway-to-Lung Ratio (Smallest Airway-to-Lung Ratio) | Participants in the Highest Quartile of Airway-to-Lung Ratio (Largest Airway-to-Lung Ratio) | ||
| All participants (n = 1,272) | ||||
| Unadjusted | 96 (88–105) | 155 (134–176) | 61 (46–76) | 94 (68–120), P < 0.0001 |
| Adjusted | 104 (95–113) | 158 (137–179) | 72 (56–88) | 86 (60–113), P < 0.0001 |
| Never-smoking participants (n = 652) | ||||
| Unadjusted | 123 (108–138) | 185 (154–216) | 66 (37–95) | 119 (77–162), P < 0.0001 |
| Adjusted | 123 (109–137) | 178 (148–209) | 79 (50–108) | 99 (56–142), P < 0.0001 |
| Participants without physician diagnosis of asthma or COPD (n = 912) | ||||
| Unadjusted | 105 (94–116) | 172 (148–196) | 66 (46–86) | 106 (75–137), P < 0.0001 |
| Adjusted | 105 (95–116) | 166 (142–190) | 71 (51–92) | 94 (63–127), P < 0.0001 |
| Participants with physician diagnosis of asthma (n = 227) | ||||
| Unadjusted | 164 (145–183) | 211 (181–242) | 69 (26–113) | 142 (89–195), P < 0.0001 |
| Adjusted | 164 (147–181) | 204 (175–232) | 90 (48–131) | 114 (63–165), P < 0.0001 |
| Participants with physician diagnosis of COPD (n = 133) | ||||
| Unadjusted | 141 (121–161) | 158 (123–194) | 73 (25–122) | 85 (25–145), P = 0.005 |
| Adjusted | 141 (123–160) | 164 (130–198) | 78 (30–126) | 86 (25–147), P = 0.006 |
| Participants with spirometry-defined COPD (n = 313)* | ||||
| Unadjusted | 155 (141–169) | 197 (176–219) | 84 (50–118) | 113 (73–153), P < 0.0001 |
| Adjusted | 155 (143–166) | 190 (170–211) | 105 (72–139) | 85 (45–125), P < 0.0001 |
Definition of abbreviations: CI = confidence interval; COPD = chronic obstructive pulmonary disease; CT = computed tomography; ΔFEV1 = post-bronchodilator FEV1 − prebronchodilator FEV1.
The mean and mean differences were estimated from linear regression models of ΔFEV1. The adjusted models included the following covariables: age, age2, sex, height, height2, cigarette smoking status, pack-years, percent emphysema, regular inhaler use, and physician diagnoses of asthma and COPD. The P value for the linear trend of airway-to-lung ratio with outcome was <0.001 for all models.
The subgroup analysis of spirometry-defined COPD was defined by a post-bronchodilator FEV1/FVC <0.7.
Discussion
CT-assessed dysanapsis was associated with spirometric response to a standardized dose of inhaled bronchodilator among older adults. This association was independent of age, sex, height, smoking history and clinical diagnoses of asthma or COPD. Moreover, it was consistent among nonsmokers and among participants with and without these diagnoses. These observations suggest that native airway tree structure variation that is common in the general population may modify inhaled bronchodilator efficacy with potential implications for precision medicine.
The mean ΔFEV1 associated with standardized inhalation of β2-agonist was 86 ml higher among participants in the smallest airway-to-lung ratio quartile than among participants in the largest, representing 83% of the mean ΔFEV1 of the entire sample (104 ml), 84% of the mean ΔFEV1 associated with long-acting β2-agonist in COPD (102 ml) (8), and 77% of the mean ΔFEV1 in asthma (112 ml) (9). Although short-term spirometric response to short-acting β2-agonist is not equivalent to longer-term efficacy of long-acting bronchodilators, these therapies share the same therapeutic target (airway tree), primary mechanism of action (smooth muscle relaxation), and physiological effect (improved airflow). Moreover, spirometric response to short-acting bronchodilators predicts response to long-acting bronchodilators (10, 11). Dysanapsis-associated differences in short-term bronchodilator response may also have implications for the spirometric diagnoses of asthma and COPD.
Mechanisms were not assessed. Dysanapsis may alter the distribution of inhaled bronchodilators within the airway tree. Alternatively, dysanapsis-associated differences in airway smooth muscle content, tone, wall compliance, or lung volume–associated differences in thoracic gas compression may influence bronchodilator responsiveness.
Limitations.
Airway-to-lung ratio may also reflect acquired lung pathologies, although the observed differences were independent of potential confounders and consistent in subgroups with and without obstructive lung disease. Most participants were non-Hispanic White, limiting generalizability. A single formulation of short-acting β2-agonist was evaluated, and the time interval between spirometry may have underestimated effect. Nevertheless, we believe this study provides the clinical equipoise and mechanistic plausibility needed to ethically justify controlled trials investigating heterogeneity of guideline-recommended pharmacotherapy efficacy.
Conclusions.
CT-assessed dysanapsis was associated with spirometric response to a standardized dose of inhaled bronchodilator independent of cigarette smoking and clinical obstructive lung diseases. These findings suggest that native airway tree structure may modify inhaled pharmacotherapy efficacy, representing a potential avenue for precision therapy.
Footnotes
A full list of participating CanCOLD investigators and institutions can be found at www.cancold.ca.
Supported by NIH/NHLBI grant R01-HL130506, Canadian Institute of Health Research (CIHR) grant PJT-162335, and a Canadian Lung Association/CIHR Catalyst Grant. M.V. was supported by Vanier Canada Graduate Scholarship. CanCOLD (Canadian Cohort Obstructive Lung Disease) was supported by the Canadian Respiratory Research Network; industry partners: AstraZeneca Canada Ltd.; Boehringer Ingelheim Canada Ltd.; GlaxoSmithKline Canada Ltd.; and Novartis. Researchers at Research Institute of the McGill University Health Centre Montreal and iCAPTURE Centre Vancouver led the project. Previous funding partners are the CIHR (CIHR/Rx&D Collaborative Research Program Operating Grant 93326); the Respiratory Health Network of the Fonds de la recherche en santé du Québec; industry partners: Almirall; Merck Nycomed; Pfizer Canada Ltd.; and Theratechnologies.
Author Contributions: Concept and design: W.C.T., J.B., M.V., and B.M.S. Drafting of the manuscript: M.V. and B.M.S. Statistical analysis: M.V. and B.M.S. Obtained funding: W.C.T., J.B., and B.M.S. Acquisition, analysis, or interpretation of data and critical revision of the manuscript for important intellectual content: all authors. B.M.S. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Originally Published in Press as DOI: 10.1164/rccm.202107-1574LE on July 15, 2021
Author disclosures are available with the text of this letter at www.atsjournals.org.
Contributor Information
Collaborators: on behalf of the CanCOLD Investigators, Yves Lacasse, Denis O’Donnell, Robert Cowie, Kenneth Chapman, Roger Goldstein, Darcy Marciniuk, Aaron Shawn, Andrea Benedetti, Paul Hernandez, Mark Fitzgerald, Teresa To, Hélène Perrault, Tanja Taivassalo, William Sheel, Peter Pare, and James C. Hogg
References
- 1. Vogelmeier CF, Criner GJ, Martinez FJ, Anzueto A, Barnes PJ, Bourbeau J, et al. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Lung Disease 2017 Report. GOLD Executive Summary. Am J Respir Crit Care Med. 2017;195:557–582. doi: 10.1164/rccm.201701-0218PP. [DOI] [PubMed] [Google Scholar]
- 2.Global Initiative for Asthma. Global Strategy for Asthma Management and Prevention. 2021https://ginasthma.org/.
- 3. Smith BM, Kirby M, Hoffman EA, Kronmal RA, Aaron SD, Allen NB, et al. MESA Lung, CanCOLD, and SPIROMICS Investigators. Association of dysanapsis with chronic obstructive pulmonary disease among older adults. JAMA. 2020;323:2268–2280. doi: 10.1001/jama.2020.6918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Vameghestahbanati M, Kirby M, Tanabe N, Vasilescu DM, Janssens W, Everaerts S, et al. Central airway tree dysanapsis extends to the peripheral airways. Am J Respir Crit Care Med. 2021;203:378–381. doi: 10.1164/rccm.202007-3025LE. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Hofmann W. Modelling inhaled particle deposition in the human lung-A review. J Aerosol Sci. 2011;42:693–724. [Google Scholar]
- 6. Bourbeau J, Tan WC, Benedetti A, Aaron SD, Chapman KR, Coxson HO, et al. Cancold Study Group. Canadian Cohort Obstructive Lung Disease (CanCOLD): fulfilling the need for longitudinal observational studies in COPD. COPD. 2014;11:125–132. doi: 10.3109/15412555.2012.665520. [DOI] [PubMed] [Google Scholar]
- 7. Buist AS, McBurnie MA, Vollmer WM, Gillespie S, Burney P, Mannino DM, et al. BOLD Collaborative Research Group. International variation in the prevalence of COPD (the BOLD Study): a population-based prevalence study. Lancet. 2007;370:741–750. doi: 10.1016/S0140-6736(07)61377-4. [DOI] [PubMed] [Google Scholar]
- 8. Kew KM, Dias S, Cates CJ. Long-acting inhaled therapy (beta-agonists, anticholinergics and steroids) for COPD: a network meta-analysis. Cochrane Database Syst Rev. 2014;(3):CD010844. doi: 10.1002/14651858.CD010844.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Ducharme FM, Ni Chroinin M, Greenstone I, Lasserson TJ. Addition of long-acting beta2-agonists to inhaled corticosteroids versus same dose inhaled corticosteroids for chronic asthma in adults and children. Cochrane Database Syst Rev. 2010;(5):CD005535. doi: 10.1002/14651858.CD005535.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Peters SP, Bleecker ER, Kunselman SJ, Icitovic N, Moore WC, Pascual R, et al. National Heart, Lung, and Blood Institute’s Asthma Clinical Research Network. Predictors of response to tiotropium versus salmeterol in asthmatic adults. J Allergy Clin Immunol. 2013;132:1068–1074.e1. doi: 10.1016/j.jaci.2013.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Kang J, Kim KT, Lee JH, Kim EK, Kim TH, Yoo KH, et al. Predicting treatable traits for long-acting bronchodilators in patients with stable COPD. Int J Chron Obstruct Pulmon Dis. 2017;12:3557–3565. doi: 10.2147/COPD.S151909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Hall G, Filipow N, Ruppel G, Okitika T, Thompson B, Kirkby J, et al. contributing GLI Network members Official ERS technical standard: Global Lung Function Initiative reference values for static lung volumes in individuals of European ancestry. Eur Resp J. 2021;57:2000289. doi: 10.1183/13993003.00289-2020. [DOI] [PubMed] [Google Scholar]