Management of COPD: Is There a Role for Quantitative Imaging?

Miranda Kirby PhD1,2, Edwin J.R. van Beek MD, PhD3, Joon Beom Seo MD, PhD,4 Juergen Biederer MD5,6,7, Yasutaka Nakano MD, PhD8, Harvey O. Coxson PhD1,2 and Grace Parraga PhD9,10

1Department of Radiology, University of British Columbia, Vancouver, Canada; 2UBC James Hogg Research Center & The Institute of Heart and Lung Health, St. Paul's Hospital, Vancouver, Canada; 3Clinical Research Imaging Centre, Queen’s Medical Research Institute, University of Edinburgh, Edinburgh, UK. 4Department of Radiology, University of Ulsan, South Korea; 5Department of Diagnostic and Interventional Radiology, University Hospital of Heidelberg, Germany; 6Translational Lung Research Center Heidelberg (TLRC), Member of the German Lung Research Center (DZL); 7Radiologie Darmstadt, Gross-Gerau County Hospital, Germany 8Division of Respiratory Medicine, Department of Internal Medicine, Shiga University of Medical Science, Shiga, Japan; 9Robarts Research Institute, The University of Western Ontario, London, Canada; 10Department of Medical Biophysics, The University of Western Ontario, London, Canada.

Correspondence to: G. Parraga PhDRobarts Research Institute1151 Richmond St NLondon, CanadaN6A 5B7Telephone: (519) 931-5265Email: gparraga@robarts.ca

Key words: COPD, Imaging, Computed Tomography, Magnetic Resonance Imaging, Optical Coherence Tomography

Word Count: 3913/3000 wordsFigures: 3/4 maxReferences: 104/50 max

Authors:Miranda KirbyEdwin van BeekJoon BeomJuergen BiedererYasutaka NakanoHarvey CoxsonGrace Parraga

Email Addresses:Miranda.Kirby@hli.ubc.caedwin-vanbeek@ed.ac.ukseojb@amc.seoul.krbiederer@ radiologie-darmstadt.de nakano@belle.shiga-med.ac.jpHarvey.Coxson@hli.ubc.cagparraga@robarts.ca

INTRODUCTION

International clinical guidelines reinforce the overwhelming opinion that chronic obstructive

pulmonary disease (COPD) is a manageable and treatable disease for patients with all grades of

disease severity.1 This is a very hopeful and optimistic notion given that smoking cessation is

the most effective intervention for managing COPD to-date, and the only intervention yet shown

to slow the rate of lung function decline. While it has been reported that pharmacological

therapy can improve lung function and reduce COPD symptoms,4 COPD exacerbations and

hospitalizations, the effects are not strong and therapeutic developments still lag behind other

major chronic illnesses. Palliative treatments such as oxygen therapy7 and lung volume reduction

either surgically8 or using interventional bronchoscopic valve placement9 also independently

offer a survival advantage for some patients with very severe disease. Despite the modest

efficacy reported using existing management strategies, COPD hospitalization and mortality

rates continue to rise, most strikingly in women,10 and health economists warn that this upward

trend will continue and very soon overwhelm our healthcare systems.11 In light of this dire

situation, new COPD management strategies are being actively pursued and investigated to

combat this growing public health emergency.

With precision medicine methods on the horizon for all patients with chronic disease, there is a

renewed optimism for the possibility of finding new ways to reduce the burden of COPD.

Precision medicine depends on a deeper understanding of the individual patient’s genetic

background and the gene-environment interactions that lead to specific underlying biological

processes – the disease “endotype” (ie. inflammation). By understanding how these distinct

underlying biological processes manifest in the individual patient – the disease “phenotype” (ie.

emphysema) – it is thought that precision medicine will enable the “right” treatment to be

provided to the right patient at the right time. A better understanding of underlying disease

endotypes will not only allow for better patient management, but will also open doors for the

discovery of new drugs/interventions and clinical trials that target the underlying pathologies

embodied by chronic lung disease. Furthermore, linking the endotype to the phenotype will

enable the identification and stratification of patients with certain disease features for clinical

trials of targeted treatments, as well as evaluation of treatment efficacy. In light of this, there is

renewed interest in the development and evaluation of the COPD phenotype measurements. In

patients with COPD, such phenotype measurements currently include clinical and functional

biomarkers as well as recently discovered pulmonary imaging biomarkers.

The technical development, clinical and commercial translation of medical imaging technologies

and the more recent development of quantitative imaging methods, have led to their increased

use in the diagnosis and management of many chronic diseases. Until very recently however, for

patients with COPD, medical imaging has not played a major role outside the setting of acute

complications or lung cancer diagnosis. Therefore, in this review we highlight three pulmonary

imaging modalities: computed tomography (CT), magnetic resonance imaging (MRI) and optical

coherence tomography (OCT) imaging and the COPD biomarkers that may be helpful for

managing COPD patients. By summarizing how imaging is currently used in COPD patient

management and its potential to play a larger role in COPD phenotyping and treatment decisions,

we endeavour to answer the question: Is there a role for quantitative imaging in patients with

COPD?

X-RAY COMPUTED TOMOGRAPHY MEASUREMENTS AND BIOMARKERS OF COPD

For over a decade, the clinical standard for COPD patient imaging has been high resolution x-ray

computed tomography (CT). Pulmonary CT enables the direct and regional quantification of a

number of COPD pathologic features and biomarkers that are relevant to the evaluation and

management of COPD patients.

Parenchyma Microstructure: Emphysema

Regional tissue destruction or emphysema is very common in COPD patients and appear as

regions of low attenuation on CT images acquired at suspended full-inspiration. To quantify low

attenuating regions, the most commonly used approaches make use of the histogram of all CT

densitometry values in the lung and a CT density threshold cut-off or a percentile of the CT

density histogram to generate “CT density masks.” Although various threshold values and

percentile points have been investigated, the -950HU threshold12-14 and the 15th percentile on the

frequency distribution curve15 are the most widely accepted.

A limitation of CT histogram measurement approaches for quantifying the extent of emphysema

is that these do not take into account the size of the low attenuation areas in the lung or how they

cluster together, which may be an important consideration in COPD patient management. One

approach that does account for clusters of values is the analysis of low attenuation clusters

(LAC)16,17 which quantifies low attenuation regions as “emphysematous holes.” Studies show

that the LAC measurement is correlated with mathematical models of emphysematous cluster

formation and agrees with radiologist emphysema scores.18-20

Figure 1A shows a centre slice, coronal CT image with all lung voxels less than the -950HU

threshold highlighted in red. The three-dimensional reconstruction of the CT airway tree and the

low attenuation clusters (colour-coded for the different lung lobes) are shown below in Figure

1B. Although it is clear in the CT density mask for the subject with moderate and severe COPD

that emphysema is located predominantly in the upper and middle lobes, the LAC analysis

provides both a visual representation of the localized disease, depicted as large spheres in regions

corresponding to greater spatial clustering of the emphysematous lesions, as well as a

quantitative measurement.

Large Airways

Airways disease is another important pathological feature in patients with COPD, and the

proximal airways can be directly visualized and quantified using low dose thoracic CT methods.

Several approaches have been used to segment the airway tree and then quantify the airway wall

dimensions, such as threshold-based analysis21 and full-width at half maximum algorithm,22

subsequent to three-dimension reconstruction of the airway tree.23-25 While many of these

algorithms are still experimental, several have been incorporated into commercially-available

software including VIDA Diagnostics, Inc. (Coralville, IA, USA), FLUIDDA (North Brunswick,

NU, USA), Pulmo3D Software (Fraunhofer MEVIS, Bremen, Germany), and Airway Inspector

(Harvard Medical School, Boston, MA).

There is a wealth of quantitative airway dimension measurements that can be obtained from CT

images. One of the measurements used in clinical research is the mean airway wall area (WA),

or airway wall area expressed as a percentage (WA%), which was shown to be significantly

correlated with histological measures of the small airways.26 However, the anatomical locations

within the airway tree where WA measurements are obtained vary across studies, from a single

airway location (eg. RB127) to an average of all airways of a particular generation (eg. 5 th

generation airways28), and therefore it is difficult to compare measurements between studies.

Furthermore, it is difficult to compare this measurement between patients because there are

different distributions of airway sizes in different patients. Another common CT airway

measurement was introduced to avoid the potential bias that occurs due to the different

distribution of airway sizes is a standardized measure for airway wall thickness.29 This measure,

called the Pi10, is derived by plotting the square root of the airway wall area versus the internal

perimeter of each measured airway and using the regression line to calculate the square root of

the airway wall area for a representative airway with an internal perimeter of 10mm. More

recently, Smith and colleagues30 proposed another way to overcome this potential measurement

bias by only including measurements from airway segments that were anatomically matched for

all subjects.

Most commercially available lung analysis software, such as is shown in Figure 1C (Apollo,

VIDA Diagnostics Inc. Coralville, IA, USA), allows the three-dimensional reconstruction and

labeling of the airway branching structure. This approach allows investigators to select an

individual airway (e.g. RB1) and obtain specific regional airway measurements.

Functional Small Airways Disease

It has long been known that small airway obstruction may be identified using evidence of CT

lucency, or gas trapping, on CT expiratory images in the absence of emphysema.31 Although

numerous approaches have been evaluated, including the inspiratory-to-expiratory volume

change of voxels with attenuation values from -860 to -950 HU (RVC -860 to -950)32 and the

expiratory-to-inspiratory ratio of mean lung density in HU (E/I ratio), a more regionally

quantitative approach was recently introduced. This approach provides a way to quantify

regional gas trapping, or “functional small airways disease,” on a voxel-by-voxel basis by

registration of the full-inspiration and full-expiration thoracic CT images. While measurements

generated using this approach have not yet been histologically validated, such measurements

were shown to be reproducible over short periods of time,37 they correlated with pulmonary

function,31, 32 showed spatial agreement with functional abnormalities identified with other

imaging modalities,38 and were associated with longitudinal changes in FEV1.39

Functional Information: Dual-Energy CT

Conventional anatomical CT cannot provide direct information regarding which parts of the lung

receive ventilation or perfusion – functional information. Therefore, other approaches such as

dual-energy CT (DECT)40-42 are being developed and evaluated. A recent study investigating

DECT with combined xenon-enhanced ventilation and iodine contrast enhanced perfusion

imaging demonstrated the feasibility of obtaining regional and quantitative ventilation and

perfusion measurements and, importantly, the measurements of the ventilation-perfusion

relationship on a voxel-by-voxel basis.43 In patients with COPD, these DECT ventilation-

perfusion measurements were shown to be significantly associated with measures of pulmonary

function.43

MAGNETIC RESONANCE IMAGING MEASUREMENTS OF COPD

There is growing interest in the use of pulmonary MRI for research, not only because of the

relative speed in which images can be generated, and excellent safety and tolerability profile, but

also because of the unique combination of morphological and functional information MRI

provides. Morpho-functional pulmonary measurements can be achieved using conventional 1H

MRI using the inherent signal of the parenchyma. In addition, functional information may also

be derived using physiologic signal alterations such as pulsatile blood flow or ventilation

stemming from the 1H MRI signal itself or using intravenous or inhaled contrast agents.

Non-contrast Enhanced 1H MRI

Clinical non-contrast enhanced 1H MRI is used to visualize and detect tissue abnormalities in the

body. However, direct visualization of lung tissue is difficult due to the low proton density and

rapid signal decay due to the very short transverse magnetization relaxation time (T2*) of aerated

lung tissue.46 This results in multiple susceptibility artifacts at air-tissue interfaces of the alveoli.

Therefore, clinical MRI for pulmonary applications necessarily focuses on abnormalities with

high signal intensity against the dark background of the lung tissue.47 For example, fluid

accumulation or solid pathology and atelectasis are visualized with excellent signal-to-

background ratio. Lung tissue abnormalities in COPD patients such as focal emphysema,

bronchiectasis and consolidation may be detected and quantified/scored using MRI despite a

slightly lower spatial resolution compared to CT.48 Already in clinical use, steady state gradient

echo and T2-weighted fast spin echo techniques help to differentiate healthy lung tissue from

focal emphysema, bullae and pneumothorax. Beyond such “clinical” scans, experimental

pulmonary MRI offers tremendous flexibility in the type of images and information that can be

obtained.

One approach used to generate images with greater signal from the lung parenchyma involves

pulse sequences that incorporate ultra-short echo times or UTE MRI.49 Short echo times (TE) or

zero-echo times (where echo time is typically defined as the time between the end of

radiofrequency excitation and the beginning of data acquisition), effectively reduce 1H signal

loss from the lung parenchyma. As a result, UTE MR images may be used to identify many of

the same pulmonary pathological features as CT. Measurements generated using this approach

often quantify the 1H signal intensity directly52 or T2*.53 Longitudinal relaxation times (T1)

themselves may be used for direct quantitative imaging of lung parenchyma alterations in

COPD.54 Studies have determined that in COPD patients there are significantly shorter T1-

relaxation times and intra-individual regional differences of T1 correspond to lung parenchyma

differences on CT at the lobar or segmental level.

Regarding functional information, fast-time-resolved imaging (obtained with gradient echo or

steady state gradient echo based sequences) can be used to detect impaired respiratory

mechanics. Regional ventilation maps can also be generated using volumetric analysis of

inspiratory and expiratory 4D data sets.57 Moreover, the inherent signal alterations from blood

flow may be used to generate lung perfusion maps. Arterial spin labeling (ASL) uses the

intrinsic contrast of magnetized, inflowing blood into the imaging plane or volume. Since ASL

does not rely on exogenous contrast, it may be used for serial or longitudinal studies with

virtually unlimited repetition of measurements.58 However, pulmonary ASL imaging has not yet

made its way into clinical applications, and successful implementation in COPD patients has not

been reported.59

Fourier Decomposition MRI is another recently introduced 1H-based MRI approach. This

technique acquires images during free breathing and then co-registers the images such that the

changes in MR signal intensity over time in each voxel due to the breathing and cardiac cycle

can be decomposed to yield ventilation-weighted and perfusion-weighted MR images. In

ventilation-weighted images, measurements of the “ventilation defects” have been shown to

agree well with dynamic contrast enhanced MRI in patients with cystic fibrosis62 and with

hyperpolarized 3He MRI and CT emphysema measurements in patients with COPD.63

Oxygen-enhanced 1H MRI

Oxygen-enhanced MRI exploits the signal inherent to T1 signal alterations stemming from the

presence of O2 in tissue and air. Pulmonary 1H images obtained during inhalation of different

concentrations of O2 (i.e., room air and 100% of O2) can be subtracted to depict signal changes in

the ventilated lung related to a combination of ventilation, membrane function, and perfusion

effects. In COPD patients, Ohno and colleagues65 demonstrated that measurements of the mean

relative enhancement ratio (the difference in signal intensity between images acquired during

100% oxygen and room-air breathing, normalized to the signal intensity of room air breathing)

showed greater heterogeneity in subjects with emphysema compared to healthy volunteers.

1H MRI with Application of Intravenous Contrast Material

First pass perfusion contrast-enhanced imaging uses an intravenous bolus injection of a clinically

approved contrast medium based on gadolinium chelates. The bolus is injected during

continuous acquisition of time-resolved T1-weighted ultrashort TR and TE gradient-echo (GRE)

imaging. When incorporated with parallel imaging and view-sharing technology, dynamic

volume studies (one full lung volume/second) can be employed. The visual evaluation of the

image sets is facilitated by subtraction of the non-enhanced from the contrast-enhanced image

signal, which results in a bright display of the contrast-enhanced lung vessels and parenchyma.66

Lung perfusion deficits due to hypoxic vasoconstriction are used for the indirect visualization of

ventilation defects in obstructive lung disease. Furthermore, visualizing perfusion dynamically

during image acquisition not only allows for perfusion abnormalities to be detected at peak

enhancement, but also for other conditions such as delayed perfusion to be detected.67 Semi-

quantitative analysis of contrast-enhanced studies is based on the calculation of signal-time

curves, signal-to-noise ratios, and contrast-to-noise ratios with region-of-interest analysis of lung

tissue signal. Absolute quantification remains difficult due to the non-linear relationship of

contrast material concentration and T1-shortening effects; hence the injected amount of contrast

material has to be chosen carefully. Because perfusion MRI relies on contrast material that is

approved for clinical use and standard MRI scanner hardware, it is one of the most

straightforward and robust functional pulmonary MRI approaches.47 A nearly finalized multi-

centre trial of 600 COPD patients that evaluates first pass perfusion MRI and CT will certainly

provide insights into the clinical value and feasibility of first pass perfusion MRI.

Another method for assessing pulmonary ventilation uses gadolinium-DTPA that is typically

administered intravenously but can also be aerosolized and inhaled. Using this approach,

pulmonary ventilation appears as bright T1-signal in lung regions where contrast is retained. This

approach has been used for experimental imaging in large animals and human volunteers, but use

in COPD has not yet been documented.71

Multinuclear MRI with Inhaled Gas-Contrast

MRI with inhaled hyperpolarized 3He or 129Xe gas provides another way of obtaining pulmonary

structural and functional measurements.72 Parenchyma microstructural information can be

obtained using a variety of diffusion-weighted MR pulse sequences. During inspiration breath-

hold, the gas diffuses within the lung micro-structure and the “apparent” diffusion coefficient

(ADC) is derived such that it reflects the level of restriction of the gas to diffusion. Studies have

shown that 3He and 129Xe diffusion-weighted measurements are significantly correlated with

histology in ex-vivo lungs. In addition to lung microstructural information, ventilation images

that show the regional distribution of inhaled gas may be acquired. Static ventilation images

have been quantified in a number of ways, including ventilation defect counting or scoring, as

well as more automated measurements using manual segmentation,75 thresholding77 or more

complex algorithms.78-80 Finally, unlike hyperpolarized 3He gas, hyperpolarized 129Xe gas is

soluble in blood and tissues making it possible to measure regional perfusion81 and alveolar

transport kinetics.82 While there is still limited clinical experience using this approach in COPD

patients, measurements have been shown to be strongly associated with the diffusing capacity of

the lung for carbon monoxide (DLCO) in patients with idiopathic pulmonary fibrosis,83 and

therefore there is strong potential to phenotype COPD patients using hyperpolarized 129Xe MRI.

3He and 129Xe static ventilation images and diffusion-weighted images are shown in Figure 2 for

a single healthy volunteer and three COPD patients in whom 3He and 129Xe ventilation defects

are readily visualized. The 3He and 129Xe ADC maps are also brighter reflecting larger ADC

numbers for the COPD subjects indicative of airspace enlargement and/or emphysematous tissue

destruction.

Other approaches, such as MRI with inhaled fluorinated gases have also been investigated in

patients with COPD. Although 19F MRI has reduced image quality in comparison to

hyperpolarized noble gas MRI, 19F MRI has the advantage that it does not require specialized and

expensive polarization equipment, and therefore has potential for more widespread use. 19F MRI

is still very early in development and although quantitative measurements have yet to be

developed, 19F MRI images show more heterogeneous signal intensity with observable

ventilation defects in COPD compared to healthy volunteers.84

OPTICAL COHERENCE TOMOGRAPHY IMAGING MEASUREMENTS OF COPD

Optical coherence tomography (OCT) is most commonly used for coronary artery and retinal

imaging. Recently, investigators have begun to explore the use of OCT for airway diseases

because of the very high resolution (approximately 15μm axial resolution) images that may be

acquired. OCT imaging can be performed during bronchoscopic procedures and are limited only

by the diameter of airway that can be accessed using the OCT probe. Typical OCT probes are

1.5mm in diameter, and therefore the small airways that are the site of airflow limitation in

COPD85 can be evaluated. Figure 3 shows OCT images obtained in a proximal airway

(approximately 3mm in diameter) and a peripheral airway along the same airway path with the

same diameter as the 1.5mm diameter OCT probe for four ex-smokers with and without airflow

limitation. In comparison to the ex-smokers without COPD, who have notably thickened airway

walls, the ex-smokers with COPD have larger alveoli and greater destruction of the airway wall.

While OCT airway imaging has been used more extensively to evaluate cancer, in COPD

patients airway wall OCT measurements are highly reproducible88 and are significantly

associated with pulmonary function measurements and CT airway wall measurements.28

Although measuring treatment changes have not yet been demonstrated in COPD patients, OCT

has been shown to identify changes in the airway wall following bronchial thermoplasty in

asthma.89 It is also possible to obtain information related to the elastic properties of the airway

using OCT by measuring the airway diameter at various airway pressures and deriving airway

compliance measurements.90

PULMONARY IMAGING TO IMPROVE COPD PATIENT MANAGEMENT

How is Imaging Currently Used in COPD Patient Management?

In current clinical practice, imaging is not routinely acquired for COPD patients. Moreover,

chest imaging was not included in recommendations for COPD patient management from the

Canadian Thoracic Society91 or in European reports,92 and imaging was only noted as valuable in

excluding alternative diagnoses and establishing the presence of comorbidities in the GOLD

Executive Summary.1 In fact, one of the only instances where CT has been mentioned as useful

in COPD management decisions was for lung volume reduction therapies.93 The National

Emphysema Treatment Trial (NETT)8 demonstrated that patients with severe COPD with a

predominance of CT emphysema in the upper lobes and reduced exercise capacity may

experience improved outcomes and survival following lung volume reduction surgery compared

to standard medical therapy. Thoracic imaging is also used in COPD patient management when

planning endobronchial valve placement. Significant improvements in pulmonary function have

been shown in COPD patients with heterogeneous emphysema and intact interlobar fissures.

Potential Role of Quantitative Imaging?

The quantitative imaging field is rapidly developing, and in concert with emerging imaging

informatics tools, large quantities of imaging information can be mined to determine image

features that predict outcomes following interventions. For example, the NETT study

demonstrated that the regional distribution of emphysema is a clinically important feature to help

select patients who better respond to lung volume reduction surgery,8 however there have been

numerous others, particularly from the large, multicentre cohort studies, including COPDGene,95

ECLIPSE96 and SPIROMICS.97 For instance, the presence of CT emphysema predicts lung

function decline;94 CT emphysema extent predicts response to certain COPD treatments;98 CT

emphysema extent and airway dimension measurements predict mortality;99 CT emphysema

extent and airway dimension measurements,100 as well as MRI ventilation defect

measurements,101 may predict COPD exacerbations that require hospitalizations. Other

quantitative imaging measurements, namely pulmonary artery enlargement (PA:A ratio

measured on CT >1) were also shown to be associated with severe exacerbations of COPD.102

Importantly, the genetic determinants of these quantitative imaging phenotypes are also being

investigated, and large multi-centre cohort studies incorporating multi-modality imaging, such as

the German ASCONET study, will certainly help provide more insight into the clinical value and

feasibility of thoracic imaging in COPD patients.70

More research is required to investigate whether quantitative imaging may help identify

phenotypes that will enable effective COPD management to improve outcomes. Furthermore,

linking information from multiple scales, from disease endotype to imaging phenotype, may

improve how COPD patients are managed at all levels from drug discovery, such that there are

more treatment options for patients with COPD, to real-time incorporation of quantitative

imaging into patient management decisions.

CONCLUSIONS

The challenges inherent to managing patients with COPD and reducing the overall disease

burden demands more and better treatment options, and new approaches that can help to

individualize patient management strategies. Towards this goal, new imaging methods and

measurements have been developed over the last several decades. The development of low dose

CT and the commercialization of CT analysis tools have led to the incorporation of CT in large-

scale, multi-center studies, enabling new discoveries and a better understanding of COPD

pathogenesis and phenotypes at the population level. Furthermore, the number of MRI

techniques for pulmonary evaluation has experienced a large growth over the last two decades,

and continues to grow, as evidence is mounting that pulmonary MRI measurements provide very

sensitive COPD biomarkers. Finally, OCT provides unique information and remains the only

imaging modality that can directly visualize and quantify the small airways – an important

therapeutic target in COPD patients. Despite the advances in imaging methods and

measurements, the road towards precision medicine in COPD is still long and will require the

standardization of imaging protocols and methods, development and validation of imaging

biomarkers, and demonstrating efficacy in clinical trials. Initiatives such as the Quantitative

Imaging Biomarkers Alliance (QIBA) and the Xenon Clinical Trials Consortium are an

important step forward, and will be critical in achieving this goal. Furthermore, it is unlikely

there is a “one size fits all” imaging modality, and future studies must investigate how we can

incorporate multi-modality imaging information for individual patients. These challenges, once

surmounted can cement the role of quantitative imaging in guiding the management of patients

with COPD.

REFERENCES

[1] Vestbo J, Hurd SS, Agusti AG, et al. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Pulmonary Disease GOLD Executive Summary. American journal of respiratory and critical care medicine 2013;187(4):347-65.

[2] Fletcher C, Peto R. The natural history of chronic airflow obstruction. British medical journal 1977;1(6077):1645-8.

[3] Anthonisen NR, Skeans MA, Wise RA, et al. The effects of a smoking cessation intervention on 14.5-year mortality: a randomized clinical trial. Annals of internal medicine 2005;142(4):233-9.

[4] O'Donnell DE, Lam M, Webb KA. Spirometric correlates of improvement in exercise performance after anticholinergic therapy in chronic obstructive pulmonary disease. American journal of respiratory and critical care medicine 1999;160(2):542-9.

[5] Fabbri LM, Calverley PM, Izquierdo-Alonso JL, et al. Roflumilast in moderate-to-severe chronic obstructive pulmonary disease treated with longacting bronchodilators: two randomised clinical trials. Lancet 2009;374(9691):695-703.

[6] Calverley PM, Rabe KF, Goehring UM, et al. Roflumilast in symptomatic chronic obstructive pulmonary disease: two randomised clinical trials. Lancet 2009;374(9691):685-94.

[7] Stoller JK, Panos RJ, Krachman S, Doherty DE, Make B, Long-term Oxygen Treatment Trial Research G. Oxygen therapy for patients with COPD: current evidence and the long-term oxygen treatment trial. Chest 2010;138(1):179-87.

[8] Fishman A, Martinez F, Naunheim K, et al. A randomized trial comparing lung-volume-reduction surgery with medical therapy for severe emphysema. The New England journal of medicine 2003;348(21):2059-73.

[9] Davey C, Zoumot Z, Jordan S, et al. Bronchoscopic lung volume reduction with endobronchial valves for patients with heterogeneous emphysema and intact interlobar fissures (the BeLieVeR-HIFi trial): study design and rationale. Thorax 2015;70(3):288-90.

[10] Centers for Disease Control and Prevention. Surveillance Summaries 2002;51(No. SS-6).[11] Khakban A, Sin DD, FitzGerald JM, et al. Ten-Year Trends in Direct Costs of COPD: A

Population-Based Study. Chest 2015;148(3):640-6.[12] Muller NL, Staples CA, Miller RR, Abboud RT. "Density mask". An objective method to

quantitate emphysema using computed tomography. Chest 1988;94(4):782-7.[13] Gevenois PA, De Vuyst P, de Maertelaer V, et al. Comparison of computed density and

microscopic morphometry in pulmonary emphysema. American journal of respiratory and critical care medicine 1996;154(1):187-92.

[14] Gevenois PA, de Maertelaer V, De Vuyst P, Zanen J, Yernault JC. Comparison of computed density and macroscopic morphometry in pulmonary emphysema. American journal of respiratory and critical care medicine 1995;152(2):653-7.

[15] Dirksen A, Friis M, Olesen KP, Skovgaard LT, Sorensen K. Progress of emphysema in severe alpha 1-antitrypsin deficiency as assessed by annual CT. Acta radiologica 1997;38(5):826-32.

[16] Mishima M, Hirai T, Itoh H, et al. Complexity of terminal airspace geometry assessed by lung computed tomography in normal subjects and patients with chronic obstructive

pulmonary disease. Proceedings of the National Academy of Sciences of the United States of America 1999;96(16):8829-34.

[17] Gould GA, MacNee W, McLean A, et al. CT measurements of lung density in life can quantitate distal airspace enlargement--an essential defining feature of human emphysema. The American review of respiratory disease 1988;137(2):380-92.

[18] Gietema HA, Muller NL, Fauerbach PV, et al. Quantifying the extent of emphysema: factors associated with radiologists' estimations and quantitative indices of emphysema severity using the ECLIPSE cohort. Academic radiology 2011;18(6):661-71.

[19] Bankier AA, De Maertelaer V, Keyzer C, Gevenois PA. Pulmonary emphysema: subjective visual grading versus objective quantification with macroscopic morphometry and thin-section CT densitometry. Radiology 1999;211(3):851-8.

[20] Park KJ, Bergin CJ, Clausen JL. Quantitation of emphysema with three-dimensional CT densitometry: comparison with two-dimensional analysis, visual emphysema scores, and pulmonary function test results. Radiology 1999;211(2):541-7.

[21] McNitt-Gray MF, Goldin JG, Johnson TD, Tashkin DP, Aberle DR. Development and testing of image-processing methods for the quantitative assessment of airway hyperresponsiveness from high-resolution CT images. Journal of computer assisted tomography 1997;21(6):939-47.

[22] Nakano Y, Muller NL, King GG, et al. Quantitative assessment of airway remodeling using high-resolution CT. Chest 2002;122(6 Suppl):271S-5S.

[23] Reinhardt JM, D'Souza ND, Hoffman EA. Accurate measurement of intrathoracic airways. IEEE transactions on medical imaging 1997;16(6):820-7.

[24] Saba OI, Hoffman EA, Reinhardt JM. Maximizing quantitative accuracy of lung airway lumen and wall measures obtained from X-ray CT imaging. J Appl Physiol (1985) 2003;95(3):1063-75.

[25] King GG, Muller NL, Whittall KP, Xiang QS, Pare PD. An analysis algorithm for measuring airway lumen and wall areas from high-resolution computed tomographic data. American journal of respiratory and critical care medicine 2000;161(2 Pt 1):574-80.

[26] Nakano Y, Wong JC, de Jong PA, et al. The prediction of small airway dimensions using computed tomography. American journal of respiratory and critical care medicine 2005;171(2):142-6.

[27] Diaz AA, Bartholmai B, San Jose Estepar R, et al. Relationship of emphysema and airway disease assessed by CT to exercise capacity in COPD. Respiratory medicine 2010;104(8):1145-51.

[28] Coxson HO, Quiney B, Sin DD, et al. Airway wall thickness assessed using computed tomography and optical coherence tomography. American journal of respiratory and critical care medicine 2008;177(11):1201-6.

[29] Grydeland TB, Dirksen A, Coxson HO, et al. Quantitative computed tomography: emphysema and airway wall thickness by sex, age and smoking. The European respiratory journal 2009;34(4):858-65.

[30] Smith BM, Hoffman EA, Rabinowitz D, et al. Comparison of spatially matched airways reveals thinner airway walls in COPD. The Multi-Ethnic Study of Atherosclerosis (MESA) COPD Study and the Subpopulations and Intermediate Outcomes in COPD Study (SPIROMICS). Thorax 2014;69(11):987-96.

[31] Arakawa H, Webb WR. Air trapping on expiratory high-resolution CT scans in the absence of inspiratory scan abnormalities: correlation with pulmonary function tests and differential diagnosis. AJR American journal of roentgenology 1998;170(5):1349-53.

[32] Matsuoka S, Kurihara Y, Yagihashi K, Hoshino M, Watanabe N, Nakajima Y. Quantitative assessment of air trapping in chronic obstructive pulmonary disease using inspiratory and expiratory volumetric MDCT. AJR American journal of roentgenology 2008;190(3):762-9.

[33] O'Donnell RA, Peebles C, Ward JA, et al. Relationship between peripheral airway dysfunction, airway obstruction, and neutrophilic inflammation in COPD. Thorax 2004;59(10):837-42.

[34] Eda S, Kubo K, Fujimoto K, Matsuzawa Y, Sekiguchi M, Sakai F. The relations between expiratory chest CT using helical CT and pulmonary function tests in emphysema. American journal of respiratory and critical care medicine 1997;155(4):1290-4.

[35] Galban CJ, Han MK, Boes JL, et al. Computed tomography-based biomarker provides unique signature for diagnosis of COPD phenotypes and disease progression. Nature medicine 2012;18(11):1711-5.

[36] Kim EY, Seo JB, Lee HJ, et al. Detailed analysis of the density change on chest CT of COPD using non-rigid registration of inspiration/expiration CT scans. European radiology 2015;25(2):541-9.

[37] Boes JL, Hoff BA, Bule M, et al. Parametric response mapping monitors temporal changes on lung CT scans in the subpopulations and intermediate outcome measures in COPD Study (SPIROMICS). Academic radiology 2015;22(2):186-94.

[38] Capaldi DP, Zha N, Guo F, et al. Pulmonary Imaging Biomarkers of Gas Trapping and Emphysema in COPD: (3)He MR Imaging and CT Parametric Response Maps. Radiology 2016;279(2):597-608.

[39] Bhatt SP, Soler X, Wang X, et al. Association Between Functional Small Airways Disease and FEV Decline in COPD. American journal of respiratory and critical care medicine 2016.

[40] Chae EJ, Seo JB, Goo HW, et al. Xenon ventilation CT with a dual-energy technique of dual-source CT: initial experience. Radiology 2008;248(2):615-24.

[41] Lee CW, Seo JB, Lee Y, et al. A pilot trial on pulmonary emphysema quantification and perfusion mapping in a single-step using contrast-enhanced dual-energy computed tomography. Investigative radiology 2012;47(1):92-7.

[42] Park EA, Goo JM, Park SJ, et al. Chronic obstructive pulmonary disease: quantitative and visual ventilation pattern analysis at xenon ventilation CT performed by using a dual-energy technique. Radiology 2010;256(3):985-97.

[43] Hwang HJ, Seo JB, Lee SM, et al. Assessment of Regional Xenon Ventilation, Perfusion, and Ventilation-Perfusion Mismatch Using Dual-Energy Computed Tomography in Chronic Obstructive Pulmonary Disease Patients. Investigative radiology 2016;51(5):306-15.

[44] Lutey BA, Lefrak SS, Woods JC, et al. Hyperpolarized 3He MR imaging: physiologic monitoring observations and safety considerations in 100 consecutive subjects. Radiology 2008;248(2):655-61.

[45] Shukla Y, Wheatley A, Kirby M, et al. Hyperpolarized 129Xe magnetic resonance imaging: tolerability in healthy volunteers and subjects with pulmonary disease. Academic radiology 2012;19(8):941-51.

[46] Wild JM, Marshall H, Bock M, et al. MRI of the lung (1/3): methods. Insights into imaging 2012;3(4):345-53.

[47] Biederer J, Beer M, Hirsch W, et al. MRI of the lung (2/3). Why ... when ... how? Insights into imaging 2012;3(4):355-71.

[48] Barreto MM, Rafful PP, Rodrigues RS, et al. Correlation between computed tomographic and magnetic resonance imaging findings of parenchymal lung diseases. European journal of radiology 2013;82(9):e492-501.

[49] Mayo JR, MacKay A, Muller NL. MR imaging of the lungs: value of short TE spin-echo pulse sequences. AJR AmJRoentgenol 1992;159(5):951-6.

[50] Lutterbey G, Gieseke J, von Falkenhausen M, Morakkabati N, Schild H. Lung MRI at 3.0 T: a comparison of helical CT and high-field MRI in the detection of diffuse lung disease. European radiology 2005;15(2):324-8.

[51] Bianchi A, Tibiletti M, Kjorstad A, et al. Three-dimensional accurate detection of lung emphysema in rats using ultra-short and zero echo time MRI. NMR in biomedicine 2015;28(11):1471-9.

[52] Ma W, Sheikh K, Svenningsen S, et al. Ultra-short echo-time pulmonary MRI: Evaluation and reproducibility in COPD subjects with and without bronchiectasis. Journal of magnetic resonance imaging : JMRI 2015;41(5):1465-74.

[53] Ohno Y, Koyama H, Yoshikawa T, et al. T2* measurements of 3-T MRI with ultrashort TEs: capabilities of pulmonary function assessment and clinical stage classification in smokers. AJR American journal of roentgenology 2011;197(2):W279-85.

[54] Stadler A, Jakob PM, Griswold M, Stiebellehner L, Barth M, Bankier AA. T1 mapping of the entire lung parenchyma: Influence of respiratory phase and correlation to lung function test results in patients with diffuse lung disease. Magnetic resonance in medicine : official journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine 2008;59(1):96-101.

[55] Alamidi DF, Morgan AR, Hubbard Cristinacce PL, et al. COPD Patients Have Short Lung Magnetic Resonance T1 Relaxation Time. Copd 2016;13(2):153-9.

[56] Jobst BJ, Triphan SM, Sedlaczek O, et al. Functional lung MRI in chronic obstructive pulmonary disease: comparison of T1 mapping, oxygen-enhanced T1 mapping and dynamic contrast enhanced perfusion. PloS one 2015;10(3):e0121520.

[57] Kolb C, Wetscherek A, Buzan MT, et al. Regional Lung Ventilation Analysis Using Temporally Resolved Magnetic Resonance Imaging. Journal of computer assisted tomography 2016.

[58] Mai VM, Hagspiel KD, Christopher JM, et al. Perfusion imaging of the human lung using flow-sensitive alternating inversion recovery with an extra radiofrequency pulse (FAIRER). Magnetic resonance imaging 1999;17(3):355-61.

[59] Molinari F, Puderbach M, Eichinger M, et al. Oxygen-enhanced magnetic resonance imaging: influence of different gas delivery methods on the T1-changes of the lungs. Investigative radiology 2008;43(6):427-32.

[60] Deimling M, Jellus V, Geiger B, Chefd'hotel C. Time Resolved Lung Ventilation Imaging by Fourier Decomposition. Proceedings of the 16th Annual Meeting of ISMRM, Toronto, Canada 2008:2639.

[61] Bauman G, Puderbach M, Deimling M, et al. Non-contrast-enhanced perfusion and ventilation assessment of the human lung by means of fourier decomposition in proton MRI. Magn ResonMed 2009;62(3):656-64.

[62] Bauman G, Puderbach M, Heimann T, et al. Validation of Fourier decomposition MRI with dynamic contrast-enhanced MRI using visual and automated scoring of pulmonary perfusion in young cystic fibrosis patients. European journal of radiology 2013;82(12):2371-7.

[63] Capaldi DP, Sheikh K, Guo F, et al. Free-breathing pulmonary 1H and Hyperpolarized 3He MRI: comparison in COPD and bronchiectasis. Academic radiology 2015;22(3):320-9.

[64] Edelman RR, Hatabu H, Tadamura E, Li W, Prasad PV. Noninvasive assessment of regional ventilation in the human lung using oxygen-enhanced magnetic resonance imaging. NatMed 1996;2(11):1236-9.

[65] Ohno Y, Hatabu H, Takenaka D, Van CM, Fujii M, Sugimura K. Dynamic oxygen-enhanced MRI reflects diffusing capacity of the lung. Magn ResonMed 2002;47(6):1139-44.

[66] Biederer J, Heussel CP, Puderbach M, Wielpuetz MO. Functional magnetic resonance imaging of the lung. Seminars in respiratory and critical care medicine 2014;35(1):74-82.

[67] Risse F, Eichinger M, Kauczor HU, Semmler W, Puderbach M. Improved visualization of delayed perfusion in lung MRI. European journal of radiology 2011;77(1):105-10.

[68] Puderbach M, Risse F, Biederer J, et al. In vivo Gd-DTPA concentration for MR lung perfusion measurements: assessment with computed tomography in a porcine model. European radiology 2008;18(10):2102-7.

[69] Ohno Y, Murase K, Higashino T, et al. Assessment of bolus injection protocol with appropriate concentration for quantitative assessment of pulmonary perfusion by dynamic contrast-enhanced MR imaging. Journal of magnetic resonance imaging : JMRI 2007;25(1):55-65.

[70] Karch A, Vogelmeier C, Welte T, et al. The German COPD cohort COSYCONET: Aims, methods and descriptive analysis of the study population at baseline. Respiratory medicine 2016;114:27-37.

[71] Haage P, Karaagac S, Spuntrup E, Truong HT, Schmidt T, Gunther RW. Feasibility of pulmonary ventilation visualization with aerosolized magnetic resonance contrast media. Investigative radiology 2005;40(2):85-8.

[72] Albert MS, Cates GD, Driehuys B, et al. Biological magnetic resonance imaging using laser-polarized 129Xe. Nature 1994;370(6486):199-201.

[73] Woods JC, Choong CK, Yablonskiy DA, et al. Hyperpolarized 3He diffusion MRI and histology in pulmonary emphysema. Magnetic resonance in medicine : official journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine 2006;56(6):1293-300.

[74] Thomen RP, Quirk JD, Roach D, et al. Direct comparison of Xe diffusion measurements with quantitative histology in human lungs. Magnetic resonance in medicine : official journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine 2016.

[75] Parraga G, Ouriadov A, Evans A, et al. Hyperpolarized 3He ventilation defects and apparent diffusion coefficients in chronic obstructive pulmonary disease: preliminary results at 3.0 Tesla. Investigative radiology 2007;42(6):384-91.

[76] Samee S, Altes T, Powers P, et al. Imaging the lungs in asthmatic patients by using hyperpolarized helium-3 magnetic resonance: assessment of response to methacholine and exercise challenge. The Journal of allergy and clinical immunology 2003;111(6):1205-11.

[77] Kauczor HU, Markstaller K, Puderbach M, et al. Volumetry of ventilated airspaces by 3He MRI: preliminary results. Investigative radiology 2001;36(2):110-4.

[78] Tustison NJ, Altes TA, Song G, de Lange EE, Mugler JP, 3rd, Gee JC. Feature analysis of hyperpolarized helium-3 pulmonary MRI: a study of asthmatics versus nonasthmatics. Magnetic resonance in medicine : official journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine 2010;63(6):1448-55.

[79] Kirby M, Heydarian M, Svenningsen S, et al. Hyperpolarized He-3 Magnetic Resonance Functional Imaging Semiautomated Segmentation. Academic radiology 2012;19(2):141-52.

[80] Guo F, Yuan J, Rajchl M, et al. Globally optimal co-segmentation of three-dimensional pulmonary H and hyperpolarized He MRI with spatial consistence prior. Medical image analysis 2015;23(1):43-55.

[81] Cleveland ZI, Cofer GP, Metz G, et al. Hyperpolarized Xe MR imaging of alveolar gas uptake in humans. PloS one 2010;5(8):e12192.

[82] Patz S, Hersman FW, Muradian I, et al. Hyperpolarized (129)Xe MRI: a viable functional lung imaging modality? European journal of radiology 2007;64(3):335-44.

[83] Kaushik SS, Freeman MS, Yoon SW, et al. Measuring diffusion limitation with a perfusion-limited gas--hyperpolarized 129Xe gas-transfer spectroscopy in patients with idiopathic pulmonary fibrosis. J Appl Physiol (1985) 2014;117(6):577-85.

[84] Halaweish AF, Moon RE, Foster WM, et al. Perfluoropropane gas as a magnetic resonance lung imaging contrast agent in humans. Chest 2013;144(4):1300-10.

[85] Hogg JC, Macklem PT, Thurlbeck WM. Site and nature of airway obstruction in chronic obstructive lung disease. The New England journal of medicine 1968;278(25):1355-60.

[86] Lam S, Standish B, Baldwin C, et al. In vivo optical coherence tomography imaging of preinvasive bronchial lesions. Clin Cancer Res 2008;14(7):2006-11.

[87] Whiteman SC, Yang Y, Gey van Pittius D, Stephens M, Parmer J, Spiteri MA. Optical coherence tomography: real-time imaging of bronchial airways microstructure and detection of inflammatory/neoplastic morphologic changes. Clinical cancer research : an official journal of the American Association for Cancer Research 2006;12(3 Pt 1):813-8.

[88] Kirby M, Ohtani K, Nickens T, et al. Reproducibility of optical coherence tomography airway imaging. Biomedical optics express 2015;6(11):4365-77.

[89] Kirby M, Ohtani K, Lopez Lisbona RM, et al. Bronchial thermoplasty in asthma: 2-year follow-up using optical coherence tomography. The European respiratory journal 2015.

[90] Williamson JP, McLaughlin RA, Noffsinger WJ, et al. Elastic properties of the central airways in obstructive lung diseases measured using anatomical optical coherence tomography. AmJRespirCrit Care Med 2011;183(5):612-9.

[91] O'Donnell DE, Hernandez P, Kaplan A, et al. Canadian Thoracic Society recommendations for management of chronic obstructive pulmonary disease - 2008 update - highlights for primary care. Canadian respiratory journal : journal of the Canadian Thoracic Society 2008;15 Suppl A:1A-8A.

[92] Miravitlles M, Vogelmeier C, Roche N, et al. A review of national guidelines for management of COPD in Europe. The European respiratory journal 2016;47(2):625-37.

[93] Celli BR, Decramer M, Wedzicha JA, et al. An Official American Thoracic Society/european Respiratory Society Statement: Research questions in chronic obstructive pulmonary disease. American journal of respiratory and critical care medicine 2015;191(7):e4-e27.

[94] Vestbo J, Edwards LD, Scanlon PD, et al. Changes in forced expiratory volume in 1 second over time in COPD. The New England journal of medicine 2011;365(13):1184-92.

[95] Regan EA, Hokanson JE, Murphy JR, et al. Genetic epidemiology of COPD (COPDGene) study design. Copd 2010;7(1):32-43.

[96] Vestbo J, Anderson W, Coxson HO, et al. Evaluation of COPD Longitudinally to Identify Predictive Surrogate End-points (ECLIPSE). The European respiratory journal 2008;31(4):869-73.

[97] Couper D, Lavange LM, Han M, et al. Design of the Subpopulations and Intermediate Outcomes in COPD Study (SPIROMICS). Thorax 2013.

[98] Lee JH, Lee YK, Kim EK, et al. Responses to inhaled long-acting beta-agonist and corticosteroid according to COPD subtype. Respiratory medicine 2010;104(4):542-9.

[99] Johannessen A, Skorge TD, Bottai M, et al. Mortality by level of emphysema and airway wall thickness. American journal of respiratory and critical care medicine 2013;187(6):602-8.

[100] Han MK, Kazerooni EA, Lynch DA, et al. Chronic obstructive pulmonary disease exacerbations in the COPDGene study: associated radiologic phenotypes. Radiology 2011;261(1):274-82.

[101] Kirby M, Pike D, Coxson HO, McCormack DG, Parraga G. Hyperpolarized He-3 Ventilation Defects Used to Predict Pulmonary Exacerbations in Mild to Moderate Chronic Obstructive Pulmonary Disease. Radiology 2014;273(3):887-96.

[102] Wells JM, Washko GR, Han MK, et al. Pulmonary arterial enlargement and acute exacerbations of COPD. The New England journal of medicine 2012;367(10):913-21.

[103] Cho MH, Castaldi PJ, Hersh CP, et al. A Genome-Wide Association Study of Emphysema and Airway Quantitative Imaging Phenotypes. American journal of respiratory and critical care medicine 2015;192(5):559-69.

[104] Castaldi PJ, Cho MH, San Jose Estepar R, et al. Genome-wide association identifies regulatory Loci associated with distinct local histogram emphysema patterns. American journal of respiratory and critical care medicine 2014;190(4):399-409.