Background: Disease accumulates in the small airways without being detected by conventional measurements. Objectives: To quantify small airway disease using a novel computed tomography (CT) inspiratory-to-expiratory approach called the disease probability measure (DPM) and to investigate the association with pulmonary function measurements. Methods: Participants from the population-based CanCOLD study were evaluated using full-inspiration/full-expiration CT and pulmonary function measurements. Full-inspiration and full-expiration CT images were registered, and each voxel was classified as emphysema, gas trapping (GasTrap) related to functional small airway disease, or normal using two classification approaches: parametric response map (PRM) and DPM (VIDA Diagnostics, Inc., Coralville, IA, USA). Results: The participants included never-smokers (n = 135), at risk (n = 97), Global Initiative for Chronic Obstructive Lung Disease I (GOLD I) (n = 140), and GOLD II chronic obstructive pulmonary disease (n = 96). PRMGasTrap and DPMGasTrap measurements were significantly elevated in GOLD II compared to never-smokers (p < 0.01) and at risk (p < 0.01), and for GOLD I compared to at risk (p < 0.05). Gas trapping measurements were significantly elevated in GOLD II compared to GOLD I (p < 0.0001) using the DPM classification only. Overall, DPM classified significantly more voxels as gas trapping than PRM (p < 0.0001); a spatial comparison revealed that the expiratory CT Hounsfield units (HU) for voxels classified as DPMGasTrap but PRMNormal (PRMNormal- DPMGasTrap = -785 ± 72 HU) were significantly reduced compared to voxels classified normal by both approaches (PRMNormal-DPMNormal = -722 ± 89 HU; p < 0.0001). DPM and PRMGasTrap measurements showed similar, significantly associations with forced expiratory volume in 1 s (FEV1) (p < 0.01), FEV1/forced vital capacity (p < 0.0001), residual volume/total lung capacity (p < 0.0001), bronchodilator response (p < 0.0001), and dyspnea (p < 0.05). Conclusion: CT inspiratory-to-expiratory gas trapping measurements are significantly associated with pulmonary function and symptoms. There are quantitative and spatial differences between PRM and DPM classification that need pathological investigation.

The concept that the major site of increased resistance to airflow in normal lungs is the small conducting airways was first proposed by Rohrer in 1915 [1]. This opinion remained well established until the appearance of Weibel's famous monograph in 1963 depicting a model of the tracheobronchial system [2]. Weibel's data showed that the total cross-sectional area of airway lumina increased exponentially at each generation of airway branching, and Green's [3] calculations later indicated that the rapid expansion in the total cross-sectional area of the smaller airway lumina results in a sharp reduction in the resistance to flow. This was later confirmed by Macklem and Mead [4] using a novel technique that allowed the total airways resistance to be partitioned into two components located central and peripheral to a catheter tip wedged into airways 2 mm in diameter. Based on this finding and using the same technique, Hogg et al. [5] demonstrated that the small airways offer little resistance in the normal human lungs, but become the major site of increased resistance to flow in chronic obstructive pulmonary disease (COPD). Taken together, Mead postulated that the smaller conducting airways <2 mm in diameter represent a “quiet zone” within the normal human lung where disease can accumulate over many years without being noticed [6]. Collectively, these developments led to the decades-long search for noninvasive methods to assess the small airways.

The introduction and rapid development of computed tomography (CT), coupled with the development of quantitative imaging methods, have led to their increased use in the diagnosis and management of many chronic lung diseases [7]. However, even with the advances in CT technology, the spatial resolution is still not great enough to visualize the small airways directly. Recently, a novel analysis method called the parametric response map (PRM) was introduced to indirectly assess the functional effects of small airway disease [8]. By registering CT images acquired at full inspiration to images acquired at full expiration and applying established Hounsfield unit (HU) thresholds, regions of the lung that trap gas that are not related to emphysema, and hence may be attributed to small airway disease, can be quantified. Although measurements of gas trapping related to “functional” small airway disease generated using this technique have been shown to be reproducible over short periods of time [9], to be correlated with pulmonary function [8], to show spatial agreement with other imaging modalities [10], and to be associated with longitudinal changes in forced expiratory volume in 1 s (FEV1) [11], these measurements have yet to be pathologically validated. Furthermore, measurements generated using a single HU threshold have obvious limitations, and therefore there is motivation to investigate other CT inspiratory-to-expiratory classification approaches.

The purpose of this report was to investigate a novel approach for quantifying CT inspiratory-to-expiratory gas trapping related to functional small airway disease called the disease probability measure (DPM) (VIDA Diagnostics, Inc., Coralville, IA, USA). Furthermore, we aimed to determine the association between CT inspiratory-to-expiratory gas trapping related to small airway disease measurements with pulmonary function and symptoms.

Participants

Participants between 45 and 90 years from the multisite, population-based CanCOLD study were evaluated [12,13]; the sites include Vancouver, Calgary, Ottawa, Quebec City, Montreal, Toronto, Halifax, Kingston, and Saskatoon. Ethics approval was obtained from the institutional review board at each CanCOLD study site. COPD status and severity grade were defined using the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria whereby participants with a FEV1/forced vital capacity (FVC) ratio <0.7 were considered to be part of the COPD group [14]. Current or ex-smokers with normal lung function (FEV1/FVC >0.7) were defined as “at risk”, and never-smokers had normal lung function with no smoking history [14]. From this entire cohort, a subset of 504 CanCOLD participants (never-smokers, at risk, GOLD I, and GOLD II) were randomly selected for DPM analysis. The 504 participants were enrolled between February 2010 and December 2013. Of the 504 participants selected for analysis, 36 were excluded due to missing plethysmography lung volumes, and therefore a total of 468 participants were investigated. Given the paucity of GOLD III/IV participants in studies that recruit from the population compared to convenience samples, no GOLD III/IV participants were included in this analysis.

Pulmonary Function Tests

Spirometry and plethysmography were performed according to the American Thoracic Society guidelines [15,16,17]. For spirometry, FEV1, FVC, and FEV1/FVC were reported. Bronchodilator response was defined as the percent change in FEV1 (L) after bronchodilator inhalation: [(FEV1 post-BD - FEV1 pre-BD)/FEV1 pre-BD] × 100. Whole-body plethysmography was also performed for measurement of functional residual capacity, residual volume (RV), total lung capacity (TLC), and diffusing capacity of the lung for carbon monoxide.

Symptom Scores

Dyspnea was evaluated using the modified Medical Research Council scale [18]. Participants were also asked whether COPD symptoms of chronic cough, chronic phlegm, and wheeze were present or absent (binary variable). As previously described [19,20], chronic cough and chronic phlegm were defined as cough or phlegm on most days for at least 3 months in two consecutive years; wheeze was defined as whistling or wheezing in the chest at any time in the last 12 months.

CT Image Acquisition

CT images were acquired at multiple sites using various CT system makes and models with the participant supine at suspended full inspiration and full expiration from the apex to the base of the lung [13]. The CT parameters for image acquisition in the participants investigated were as follows: 100 kVp, 50 mAs, 0.5 s gantry rotation, pitch of 1.375, 1.00-1.25 mm slice thickness, and an intermediate reconstruction kernel (GE: Standard; Siemens: b35; Philips: B) was used for quantitative analysis.

CT Image Analysis

All CT analyses were performed by VIDA Diagnostics, Inc. The full-inspiration and full-expiration CT images were registered using a nonlinear registration algorithm, as previously described [21]. The registration was validated using manually selected landmarks in participants with and without COPD; the target registration error was approximately 1-2 voxels (data not shown). Classification of the registered inspiratory-to-expiratory CT lung voxels was performed using both the established PRM [8] and the DPM approach. PRM classification of emphysema and gas trapping were generated using the -950-HU and -856-HU thresholds on registered full-inspiration and full-expiration CT images, respectively [8]. The following measurements were generated: PRMNormal: percentage of lung voxels without gas trapping and emphysema; PRMGasTrap: percentage of lung voxels with gas trapping but no emphysema; and PRMEmph: percentage of lung voxels with both gas trapping and emphysema. The DPM classification developed by VIDA Diagnostics, Inc. is described in Figure 1 for a GOLD II COPD participant (56-year-old male, FEV1 = 77%pred, FEV1/FVC = 56). In contrast to the PRM method which applied the same single threshold across the whole lung for each participant, the DPM method takes into account the HU value within each registered voxel at inspiration and expiration. First, an exponential decay function determined the continuous probability of gas trapping using the difference between CT inspiration and expiration intensity; the exponential decay function was optimized using an external cohort [22]. For example, a difference between CT inspiration and expiration of 0 HU, indicating no lung emptying on expiration, corresponded to a gas trapping probability of 100%, while a difference between CT inspiration and expiration of ≥0 HU corresponded to a lower probability of gas trapping (exponentially approaching 0%). The continuous probability that the voxel contained emphysema was also determined using an optimized exponential function computed from the average normalized CT inspiration and expiration intensity values. The normalized HU values for both inspiration and expiration were defined as 1 for ≥0 HU and 0 for ≤-1,000 HU, and intensity values between these boundaries were assigned linearly. For example, a CT inspiration and expiration voxel with -1,000 HU (pure air) had 100% probability of emphysema. After the continuous probability of gas trapping and emphysema had been determined, voxels with <50% probability of gas trapping and <50% probability of emphysema were classified as normal (DPMNormal), voxels with >50% probability of gas trapping and >50% probability of emphysema were classified as emphysema (DPMEmph), and voxels with >50% probability of gas trapping and <50% probability of emphysema were classified as gas trapping (DPMGasTrap).

Fig. 1

CT inspiratory-to-expiratory DPM classification methodology. CT inspiratory and expiratory images were first registered using a nonlinear, deformable registration algorithm. Next, the DPM method took into account the HU value within each voxel at inspiration and expiration and, using probability density functions, computed the probability (between 0 and 100%) that each voxel contains gas trapping and emphysema. CT, computed tomography; DPM, disease probability measure; HU, Hounsfield unit.

Fig. 1

CT inspiratory-to-expiratory DPM classification methodology. CT inspiratory and expiratory images were first registered using a nonlinear, deformable registration algorithm. Next, the DPM method took into account the HU value within each voxel at inspiration and expiration and, using probability density functions, computed the probability (between 0 and 100%) that each voxel contains gas trapping and emphysema. CT, computed tomography; DPM, disease probability measure; HU, Hounsfield unit.

Close modal

Statistical Analysis

A one-way analysis of variance was used for comparisons between never-smokers, at risk, GOLD I, and GOLD II participants for demographics and pulmonary function, and a two-way analysis of variance was used for comparison of PRM and DPM measurements between never-smokers, at risk, GOLD I, and GOLD II participants; multiple-comparisons test correction was performed using a Tukey test (GraphPad Prism 6, La Jolla, CA, USA). To compare the spatial overlap between PRM and DPM measurements, we calculated the Dice similarity coefficient (DSC) [23]; all voxels with measurements <0.5% were excluded from the analysis. A multivariable linear regression analysis (SAS 9.4 software, Cary, NC, USA) was used to determine the association between pulmonary function and clinical measurements (symptom score response variables were treated as ordinal) with PRMGasTrap and DPMGasTrap adjusted for age, sex, race, body mass index, and scanner model.

Participant Demographics and Pulmonary Function Measurements

Table 1 shows the demographics and pulmonary function measurements for all participants. The GOLD I participants were slightly but significantly older than the never-smokers (never-smokers 65 ± 10 years, GOLD I participants 68 ± 10 years, p > 0.05) and had significantly fewer female participants than the never-smokers (never-smokers 50%, GOLD I 36%) and at risk (51%) groups. There were no differences between the groups for body mass index. Pulmonary function measurements and smoking history worsened significantly according to GOLD grade.

Table 1

Subject demographics and pulmonary function measurements for all participants

Subject demographics and pulmonary function measurements for all participants
Subject demographics and pulmonary function measurements for all participants

PRM and DPM Measurements

Figure 2 shows a representative image of the expiratory CT, the registered inspiration CT, and the corresponding PRM- and DPM-derived map for a never-smoker, at risk, GOLD I, and GOLD II COPD participant. While the number of voxels classified as PRMGasTrap and DPMGasTrap (shown in yellow) increased with increasing GOLD grade, reflective of more severe gas trapping related to functional small airway disease, there were visually a greater number of voxels classified as DPMGasTrap than PRMGasTrap in each group. There were relatively few voxels classified as PRMEmph and DPMEmph (shown in red); however, these voxels also increased with increasing GOLD grade, indicating more emphysematous tissue destruction.

Fig. 2

Expiratory CT and registered inspiratory CT images with corresponding PRM and DPM map for a never-smoker, at risk, GOLD I, and GOLD II subject. Never-smoker: 47-year-old male, FEV1 = 107%pred, FEV1/FVC = 85%, PRMNormal = 100%, PRMEmph = 0%, PRMGasTrap = 0%, DPMNormal = 99%, DPMEmph = 0%, DPMGasTrap = 1%. At risk: 82-year-old female ex-smoker, pack-years = 21, FEV1 = 99%pred, FEV1/FVC = 73%, PRMNormal = 97%, PRMEmph = 0%, PRMGasTrap = 3%, DPMNormal = 89%, DPMEmph = 0%, DPMGasTrap = 11%. GOLD I: 59-year-old female ex-smoker, pack-years = 59, FEV1 = 87%pred, FEV1/FVC = 69%, PRMNormal = 92%, PRMEmph = 2%, PRMGasTrap = 6%, DPMNormal = 71%, DPMEmph = 2%, DPMGasTrap = 27%. GOLD II: 56-year-old male current smoker, pack-years = 86, FEV1 = 77%pred, FEV1/ FVC = 56%, PRMNormal = 92%, PRMEmph = 1%, PRMGasTrap = 7%, DPMNormal = 62%, DPMEmph = 1%, DPMGasTrap = 37%. CT, computed tomography; DPM, disease probability measure; FEV1, forced expiratory volume in 1 s; fSAD, functional small airway disease; FVC, forced vital capacity; GOLD, Global Initiative for Chronic Obstructive Lung Disease; PRM, parametric response map.

Fig. 2

Expiratory CT and registered inspiratory CT images with corresponding PRM and DPM map for a never-smoker, at risk, GOLD I, and GOLD II subject. Never-smoker: 47-year-old male, FEV1 = 107%pred, FEV1/FVC = 85%, PRMNormal = 100%, PRMEmph = 0%, PRMGasTrap = 0%, DPMNormal = 99%, DPMEmph = 0%, DPMGasTrap = 1%. At risk: 82-year-old female ex-smoker, pack-years = 21, FEV1 = 99%pred, FEV1/FVC = 73%, PRMNormal = 97%, PRMEmph = 0%, PRMGasTrap = 3%, DPMNormal = 89%, DPMEmph = 0%, DPMGasTrap = 11%. GOLD I: 59-year-old female ex-smoker, pack-years = 59, FEV1 = 87%pred, FEV1/FVC = 69%, PRMNormal = 92%, PRMEmph = 2%, PRMGasTrap = 6%, DPMNormal = 71%, DPMEmph = 2%, DPMGasTrap = 27%. GOLD II: 56-year-old male current smoker, pack-years = 86, FEV1 = 77%pred, FEV1/ FVC = 56%, PRMNormal = 92%, PRMEmph = 1%, PRMGasTrap = 7%, DPMNormal = 62%, DPMEmph = 1%, DPMGasTrap = 37%. CT, computed tomography; DPM, disease probability measure; FEV1, forced expiratory volume in 1 s; fSAD, functional small airway disease; FVC, forced vital capacity; GOLD, Global Initiative for Chronic Obstructive Lung Disease; PRM, parametric response map.

Close modal

A quantitative comparison between the PRM and DPM measurements for never-smokers, at risk, GOLD I, and GOLD II is shown in Figure 3. Overall, DPMGasTrap measurements were significantly elevated compared to PRMGasTrap measurements (p < 0.0001). For both PRMGasTrap and DPMGasTrap, measurements were significantly elevated in GOLD II compared to never-smokers (p = 0.007 and p < 0.0001, respectively) and at risk (p = 0.006 and p < 0.0001, respectively) participants, and for GOLD I compared to at risk (p = 0.04 and p = 0.002, respectively); DPMGasTrap measurements were significantly elevated in GOLD II compared to GOLD I (p < 0.0001), but not PRMGasTrap measurements.

Fig. 3

Comparison of PRMGasTrap and DPMGasTrap (a) and PRMEmph and DPMEmph (b) measurements for never-smokers, at risk, GOLD I, and GOLD II COPD. COPD, chronic obstructive pulmonary disease; DPM, disease probability measure; GOLD, Global Initiative for Chronic Obstructive Lung Disease; PRM, parametric response map.

Fig. 3

Comparison of PRMGasTrap and DPMGasTrap (a) and PRMEmph and DPMEmph (b) measurements for never-smokers, at risk, GOLD I, and GOLD II COPD. COPD, chronic obstructive pulmonary disease; DPM, disease probability measure; GOLD, Global Initiative for Chronic Obstructive Lung Disease; PRM, parametric response map.

Close modal

There were no overall differences between PRMEmph and DPMEmph measurements (p = 0.77). For both PRMEmph and DPMEmph, measurements were significantly elevated in GOLD II compared to never-smokers (p = 0.003 and p = 0.003, respectively), at risk (p = 0.007 and p = 0.008, respectively), and GOLD I (p = 0.04 and p = 0.03, respectively), although there were no other differences between the groups.

PRM and DPM Measurements: Spatial Comparison

Figure 4 shows the spatial comparison between the PRM and the DPM classification. The greatest spatial agreement for the PRM and DPM approach, as indicated by the DSC, was for voxels classified as normal (DSC = 0.65 ± 0.29), followed by emphysema (DSC = 0.50 ± 0.12), and the lowest spatial agreement was for gas trapping (DSC = 0.36 ± 0.18) (Fig. 4a).

Fig. 4

Box and whisker plots of DSCs for PRM and DPM voxel classification. a Box and whisker plot (whiskers represent the 5th to 95th percentiles) shows DSCs for PRMNormal and DPMNormal measurements (DSC = 0.65 ± 0.29), PRMEmph and DPMEmph (DSC = 0.50 ± 0.12), and PRMGasTrap and DPMGasTrap measurements (DSC = 0.36 ± 0.18). b The expiratory CT HU value for voxels classified as PRMNormal and DPMNormal (-722 ± 89 HU) was significantly greater than voxels classified as PRMGasTrap and DPMGasTrap (-884 ± 21 HU, p < 0.0001) and PRMNormal but DPMGasTrap (-785 ± 72 HU, p < 0.0001). The expiratory CT HU value for voxels classified as PRMNormal but DPMGasTrap was significantly greater than voxels classified as PRMGasTrap and DPMGasTrap (p < 0.0001). * p < 0.05. CT, computed tomography; DPM, disease probability measure; DSC, Dice similarity coefficient; HU, Hounsfield unit(s); PRM, parametric response map.

Fig. 4

Box and whisker plots of DSCs for PRM and DPM voxel classification. a Box and whisker plot (whiskers represent the 5th to 95th percentiles) shows DSCs for PRMNormal and DPMNormal measurements (DSC = 0.65 ± 0.29), PRMEmph and DPMEmph (DSC = 0.50 ± 0.12), and PRMGasTrap and DPMGasTrap measurements (DSC = 0.36 ± 0.18). b The expiratory CT HU value for voxels classified as PRMNormal and DPMNormal (-722 ± 89 HU) was significantly greater than voxels classified as PRMGasTrap and DPMGasTrap (-884 ± 21 HU, p < 0.0001) and PRMNormal but DPMGasTrap (-785 ± 72 HU, p < 0.0001). The expiratory CT HU value for voxels classified as PRMNormal but DPMGasTrap was significantly greater than voxels classified as PRMGasTrap and DPMGasTrap (p < 0.0001). * p < 0.05. CT, computed tomography; DPM, disease probability measure; DSC, Dice similarity coefficient; HU, Hounsfield unit(s); PRM, parametric response map.

Close modal

To determine whether voxels classified as PRMNormal but DPMGasTrap showed evidence of greater expiratory CT gas trapping, the expiratory CT HU values were investigated. As shown in Figure 4b, the mean expiration CT HU measurement for voxels classified as PRMNormal but DPMGasTrap (PRMNormal-DPMGasTrap = -785 ± 72 HU) was significantly greater compared to voxels classified as gas trapping by both approaches (PRMGasTrap- DPMGasTrap = -884 ± 21 HU, p < 0.0001), but significantly reduced compared to voxels classified as normal by both approaches (PRMNormal-DPMNormal = -722 ± 89 HU; p < 0.0001).

PRM and DPM Measurements: Associations with Pulmonary Function Measurements and Symptoms

Table 2 shows separate multivariable linear regression models for pulmonary function measurements and symptoms with PRMGasTrap and DPMGasTrap measurements. After adjusting for confounding variables, both PRMGasTrap and DPMGasTrap were significantly associated with FEV1 (p < 0.001), FEV1/FVC (p < 0.0001), RV/TLC (p < 0.0001), bronchodilator response (p < 0.0001), and dyspnea (p < 0.05).

Table 2

Multivariable regression models for pulmonary function and symptoms with PRMGasTrap and DPMGasTrap measurements

Multivariable regression models for pulmonary function and symptoms with PRMGasTrap and DPMGasTrap measurements
Multivariable regression models for pulmonary function and symptoms with PRMGasTrap and DPMGasTrap measurements

The search for reliable and meaningful biomarkers of early COPD-related changes has been ongoing for several decades. The use of these biomarkers is thought to be vital if therapeutic interventions at an early and potentially modifiable time point are to be developed. Our data from a population-based study of mild-to-moderate COPD participants indicate that (1) both PRM and DPM measurements of gas trapping related to functional small airway disease distinguished participants with mild and moderate COPD from those at risk, (2) DPM classified significantly more voxels as gas trapping than PRM and spatial analysis of the HU values indicated that DPMGasTrap but PRMNormal voxels had significantly lower HU values than voxels classified as normal by both approaches, and (3) both PRM and DPM showed similar, significant associations with pulmonary function measurements and dyspnea.

Both the PRMGasTrap and DPMGasTrap measurements were significantly elevated in GOLD I and GOLD II participants compared to those who had never smoked or were at risk of COPD but had normal lung function. However, only DPMGasTrap distinguished between GOLD I and GOLD II COPD, and overall, the DPM approach classified significantly more voxels as gas trapping related to small airway disease than PRM. While this suggests that DPMGasTrap measurements may be more sensitive than PRM, without pathological validation it is unclear whether DPM improved specificity as well as sensitivity. We do note that emphysema measurements were consistent between PRM and DPM. Both PRM and DPM emphysema measurements were low in all participant groups and were only significantly elevated in GOLD II participants.

To provide a deeper understanding of the differences in the voxels that PRM and DPM classified as gas trapping related to small airway disease, we evaluated the spatial relationship of the two techniques using DSCs. These data showed that the gas trapping measurements had the lowest spatial agreement in comparison to voxels classified as normal and emphysema. Investigation of this discordance revealed that voxels classified as normal using PRM but gas trapping using DPM classification had significantly more expiratory gas trapping than voxels which both approaches classified as normal, thus indicating that DPM may identify regions that trap gas but do not reach the -856-HU CT threshold, which may explain the apparent greater sensitivity. Longitudinal and/or pathological studies are required to confirm that the DPM approach identifies gas trapping regions before they reach the -856-HU threshold and are detected by the PRM approach.

Finally, we demonstrated that both PRM and DPM showed similar, significant associations with pulmonary function measurements and dyspnea. This finding that both PRM and DPM are associated with established physiological tests of small airways obstruction and COPD symptoms bodes well for CT inspiratory-to-expiratory small airway disease measurements, and indicates that future investigation of both methods is warranted.

There are some limitations of this study that deserve mention. An important consideration and source of variability [24] for CT inspiratory-to-expiratory measurements are insufficient inhalation and/or exhalation to the target lung volume during image acquisition as well as technical factors such as the mismatch of reconstruction kernels and slice thickness between inspiration and expiration CT images [24]. Unlike other multicenter COPD cohort studies such as COPDGene [22] and SPIROMICS [25], the CanCOLD study acquired plethysmography measurements, and therefore we were able to directly compare plethysmography-derived lung volumes to CT-derived lung volumes for quality control. Our data show that even without a standardized breath-hold protocol to carefully control for lung volume, there was good agreement between CT and pulmonary function testing-derived lung volumes, indicating that routine clinical instructions during image acquisition allow for adequate inspiratory and expiratory images to be acquired (online suppl. Table 1; for all online suppl. material, see www.karger.com/doi/10.1159/000478865). While we acknowledge that efforts should be made to acquire CT images at the target lung volumes, such as implementing a standardized protocol for coaching prior to and during image acquisition [26], this finding bodes well for this advanced analysis technique in future clinical research studies conducted at multiple sites, with different CT systems and technologists. It is also important to note that we reported significant associations for PRM and DPM measurements with pulmonary function measurements and dyspnea regardless of whether we investigated all participants, only those with CT images acquired using the same convolution kernel/slice thickness between full-inspiration and full-expiration images, or only those with CT-measured lung volumes that were in agreement with the plethysmography-measured lung volumes (see online suppl. material).

We would also like to acknowledge that while the gas trapping measurements generated by the two classification approaches were not in good agreement, it was beyond the scope of this study to determine which approach more closely represents ground truth. However, obtaining inspiration and expiration CT before surgical resection and performing morphometric analysis on the excised lung tissue samples may provide a way to validate these measurements. Furthermore, following these participants longitudinally will aid in determining whether these measurements provide prognostic information and reflect longitudinal changes in clinical outcomes. It is important to note, however, that both the PRM and DPM methods are equally accessible and can be obtained commercially (Imbio, Minneapolis, MN, USA; VIDA Diagnostics, Inc.).

In conclusion, CT registration-based measurements may provide a method to untangle the underlying small airway disease and emphysema disease contributions. The results of our study indicate that the functional effects of small airway obstruction quantified using inspiratory-to-expiratory CT PRM and DPM measurements are associated with pulmonary function measurements and COPD symptoms. Future research involving longitudinal evaluations will help determine the potential of these measurements as intermediate endpoints in clinical trials.

We acknowledge funding support from the Canadian Institute of Heath Research (CIHR/Rx&D Collaborative Research Program Operating Grant 93326); the Respiratory Health Network of the Fonds de Recherche Santé Québec (FRSQ); the Canadian Respiratory Research Network (CRRN); the Canadian Lung Association (CLA)/Canadian Thoracic Society (CTS); the British Columbia Lung Association; and the industry partners Astra Zeneca Canada, Inc., Boehringer Ingelheim Canada, Inc., GlaxoSmithKline Canada, Inc., Merck, Novartis Pharma Canada, Inc., Nycomed Canada, Inc., and Pfizer Canada, Ltd. Dr. Kirby gratefully acknowledges postdoctoral support from the Canadian Institutes of Health Research (CIHR) Banting Program. Dr. Coxson was a Roberta R. Miller Fellow in Thoracic Imaging from the British Columbia Lung Association. The investigators would also like to acknowledge VIDA Diagnostics, Inc., Clinical Image Analysis Services for support with the analysis of the CT images.

M. Kirby, W.C. Tan, J. Leipsic, C.J. Hague, J. Bourbeau, D.D. Sin, J.C. Hogg, and H.O. Coxson have no conflicts of interest to declare. Y. Yin and J. Tschirren are employed by VIDA Diagnostics, Inc., a software company that commercializes lung image analysis software.

1.
Rohrer F: Flow resistance in human air passages and the effect of irregular branching of the bronchial system on the respiratory process in various regions of the lungs. Arch Ges Physiol 1915;162:225-299.
2.
Weibel ER: Morphometry of the Human Lung. New York, Academic Press, 1963.
3.
Green M: How big are the bronchioles? St Thomas Hosp Gaz 1965;63:136-139.
4.
Macklem PT, Mead J: Resistance of central and peripheral airways measured by a retrograde catheter. J Appl Physiol 1967;22:395-401.
5.
Hogg JC, Macklem PT, Thurlbeck WM: Site and nature of airway obstruction in chronic obstructive lung disease. N Engl J Med 1968;278:1355-1360.
6.
Mead J: The lung's “quiet zone.” N Engl J Med 1970;282:1318-1319.
7.
Celli BR, Decramer M, Wedzicha JA, Wilson KC, Agustí A, Criner GJ, MacNee W, Make BJ, Rennard SI, Stockley RA, Vogelmeier C, Anzueto A, Au DH, Barnes PJ, Burgel PR, Calverley PM, Casanova C, Clini EM, Cooper CB, Coxson HO, Dusser DJ, Fabbri LM, Fahy B, Ferguson GT, Fisher A, Fletcher MJ, Hayot M, Hurst JR, Jones PW, Mahler DA, Maltais F, Mannino DM, Martinez FJ, Miravitlles M, Meek PM, Papi A, Rabe KF, Roche N, Sciurba FC, Sethi S, Siafakas N, Sin DD, Soriano JB, Stoller JK, Tashkin DP, Troosters T, Verleden GM, Verschakelen J, Vestbo J, Walsh JW, Washko GR, Wise RA, Wouters EF, ZuWallack RL; ATS/ERS Task Force for COPD Research: An Official American Thoracic Society/European Respiratory Society Statement: research questions in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2015;191:e4-e27.
8.
Galban CJ, Han MK, Boes JL, Chughtai KA, Meyer CR, Johnson TD, Galban S, Rehemtulla A, Kazerooni EA, Martinez FJ, Ross BD: Computed tomography-based biomarker provides unique signature for diagnosis of COPD phenotypes and disease progression. Nat Med 2012;18:1711-1715.
9.
Boes JL, Hoff BA, Bule M, Johnson TD, Rehemtulla A, Chamberlain R, Hoffman EA, Kazerooni EA, Martinez FJ, Han MK, Ross BD, Galban CJ: Parametric response mapping monitors temporal changes on lung CT scans in the Subpopulations and Intermediate Outcome Measures in COPD Study (SPIROMICS). Acad Radiol 2015;22:186-194.
10.
Capaldi DP, Zha N, Guo F, Pike D, McCormack DG, Kirby M, Parraga G: Pulmonary imaging biomarkers of gas trapping and emphysema in COPD: (3)He MR imaging and CT parametric response maps. Radiology 2016;279:597-608.
11.
Bhatt SP, Soler X, Wang X, Murray S, Anzueto AR, Beaty TH, Boriek AM, Casaburi R, Criner GJ, Diaz AA, Dransfield MT, Curran-Everett D, Galban CJ, Hoffman EA, Hogg JC, Kazerooni EA, Kim V, Kinney GL, Lagstein A, Lynch DA, Make BJ, Martinez FJ, Ramsdell JW, Reddy R, Ross BD, Rossiter HB, Steiner RM, Strand MJ, van Beek EJ, Wan ES, Washko GR, Wells JM, Wendt CH, Wise RA, Silverman EK, Crapo JD, Bowler RP, Han MK; COPDGene Investigators: Association between functional small airway disease and FEV1 decline in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2016;194:178-184.
12.
Tan WC, Sin DD, Bourbeau J, Hernandez P, Chapman KR, Cowie R, FitzGerald JM, Marciniuk DD, Maltais F, Buist AS, Road J, Hogg JC, Kirby M, Coxson H, Hague C, Leipsic J, O'Donnell DE, Aaron SD; CanCOLD Collaborative Research Group: Characteristics of COPD in never-smokers and ever-smokers in the general population: results from the CanCOLD study. Thorax 2015;70:822-829.
13.
Bourbeau J, Tan WC, Benedetti A, Aaron SD, Chapman KR, Coxson HO, Cowie R, Fitzgerald M, Goldstein R, Hernandez P, Leipsic J, Maltais F, Marciniuk D, O'Donnell D, Sin DD, CanCOLD Study Group: Canadian Cohort Obstructive Lung Disease (CanCOLD): fulfilling the need for longitudinal observational studies in COPD. COPD 2014;11:125-132.
14.
Vestbo J, Hurd SS, Rodriguez-Roisin R: The 2011 revision of the global strategy for the diagnosis, management and prevention of COPD (GOLD) - why and what? Clin Respir J 2012;6:208-214.
15.
Macintyre N, Crapo RO, Viegi G, Johnson DC, van der Grinten CP, Brusasco V, Burgos F, Casaburi R, Coates A, Enright P, Gustafsson P, Hankinson J, Jensen R, McKay R, Miller MR, Navajas D, Pedersen OF, Pellegrino R, Wanger J: Standardisation of the single-breath determination of carbon monoxide uptake in the lung. Eur Respir J 2005;26:720-735.
16.
Wanger J, Clausen JL, Coates A, Pedersen OF, Brusasco V, Burgos F, Casaburi R, Crapo R, Enright P, van der Grinten CP, Gustafsson P, Hankinson J, Jensen R, Johnson D, Macintyre N, McKay R, Miller MR, Navajas D, Pellegrino R, Viegi G: Standardisation of the measurement of lung volumes. Eur Respir J 2005;26:511-522.
17.
Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, Crapo R, Enright P, van der Grinten CP, Gustafsson P, Jensen R, Johnson DC, MacIntyre N, McKay R, Navajas D, Pedersen OF, Pellegrino R, Viegi G, Wanger J; ATS/ERS Task Force: Standardisation of spirometry. Eur Respir J 2005;26:319-338.
18.
Mahler DA, Wells CK: Evaluation of clinical methods for rating dyspnea. Chest 1988;93:580-586.
19.
Tan WC, Bourbeau J, Hernandez P, Chapman KR, Cowie R, FitzGerald JM, Marciniuk DD, Maltais F, Buist AS, O'Donnell DE, Sin DD, Aaron SD; CanCOLD Collaborative Research Group: Exacerbation-like respiratory symptoms in individuals without chronic obstructive pulmonary disease: results from a population-based study. Thorax 2014;69:709-717.
20.
Buist AS, McBurnie MA, Vollmer WM, Gillespie S, Burney P, Mannino DM, Menezes AM, Sullivan SD, Lee TA, Weiss KB, Jensen RL, Marks GB, Gulsvik A, Nizankowska-Mogilnicka E; BOLD Collaborative Research Group: International variation in the prevalence of COPD (the BOLD Study): a population-based prevalence study. Lancet 2007;370:741-750.
21.
Yin Y, Raffy P, Sieren JP, Newell JD, Hoffman EA: Quantitative CT in SPIROMICS: lobar evaluation via image warping of TLC masks onto RV images. Am J Respir Crit Care Med 2014;189:A4332.
22.
Regan EA, Hokanson JE, Murphy JR, Make B, Lynch DA, Beaty TH, Curran-Everett D, Silverman EK, Crapo JD: Genetic epidemiology of COPD (COPDGene) study design. COPD 2010;7:32-43.
23.
Dice LR: Measures of the amount of ecologic association between species. Ecology 1945;26:297-302.
24.
Boes JL, Bule M, Hoff BA, Chamberlain R, Lynch DA, Stojanovska J, Martinez FJ, Han MK, Kazerooni EA, Ross BD, Galban CJ: The impact of sources of variability on parametric response mapping of lung CT scans. Tomography 2015;1:69-77.
25.
Couper D, Lavange LM, Han M, Barr RG, Bleecker E, Hoffman EA, Kanner R, Kleerup E, Martinez FJ, Woodruff PG, Rennard S; SPIROMICS Research Group: Design of the Subpopulations and Intermediate Outcomes in COPD Study (SPIROMICS). Thorax 2014;69:491-494.
26.
Hayward WS, Green MH: Inhibition of Escherichia coli and bacteriophage lambda messenger RNA synthesis by T4. Proc Natl Acad Sci USA 1965;54:1675-1678.
Copyright / Drug Dosage / Disclaimer
Copyright: All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher.
Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.
Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.