You are on page 1of 332

Sang-Do Lee


Heterogeneity and
Personalized Treatment

Sang-Do Lee

Heterogeneity and Personalized
Sang-Do Lee
Department of Pulmonary and Critical Care Medicine
Asan Medical Center University
of Ulsan College of Medicine
South Korea

ISBN 978-3-662-47177-7    ISBN 978-3-662-47178-4 (eBook)

DOI 10.1007/978-3-662-47178-4

Library of Congress Control Number: 2017947875

© Springer-Verlag Berlin Heidelberg 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, express or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer-Verlag GmbH Germany
The registered company address is: Heidelberger Platz 3, 14197 Berlin, Germany

Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity

and mortality worldwide. Over the last few decades, the study of COPD has
become one of the most rapidly developing fields in medicine. The recent years
have provided clinicians and researchers with major advances in the under-
standing of underlying mechanisms in COPD. In the past decades, COPD was
classified solely on the basis of the degree of airflow limitation. Nowadays,
COPD is regarded as a heterogeneous disease, with multiple etiological factors,
clinical phenotypes, and comorbidities. One of the main reasons for poor
understanding and poor treatment is the heterogeneity of COPD. The strategy
for the management of COPD is moving toward a more personalized approach
compared with the historical approach. Dissecting the heterogeneity would
lead to a better understanding and effective personalized treatment of COPD.
Airway Vista, also known as Chronic Obstructive Airway Diseases
Symposium, has been hosted by the Obstructive Lung Disease Research
Foundation in South Korea since 2008. This academic event is designed to offer
respiratory health professionals new horizons in their understanding of COPD
and asthma. The scientific program of the symposium includes the most signifi-
cant advances in the researches of chronic airway diseases, COPD, asthma, and
pulmonary functional imaging. We have held Airway Vista successfully every
year, featuring more than 50 world-renowned speakers respectively. This year
(2017) has marked the 10th anniversary of Airway Vista. To celebrate the
achievements of this 10-year-old symposium, we decided to publish a textbook
by gathering the contents of previous symposium programs. We have tried to
provide readers with an overview of COPD, the current understanding of its
pathobiology, and a contemporary approach to diagnosis and treatment. With
this goal in mind, a group of experts took the task of developing this publica-
tion, focusing on essential issues that all providers should be aware of.
The first chapter of this book covers overviews of COPD which include
the current definition, epidemiology, risk factors, and pathogenesis of COPD.
The second chapter is comprised of diagnosis and assessment given to COPD
patients. In Chap. 3, COPD heterogeneity was described in a clinical pheno-
type as well as radiological and genetic aspects. Various pharmacological and
nonpharmacological management strategies are reviewed based on evidence
in the fourth chapter. The final chapter outlines a future perspective on COPD.
This book presents state-of-the-art knowledge on issues related to heteroge-
neity, such as phenotypes (clinical, physiological, radiological, etc.), geno-
types, tools to be used for dissecting heterogeneity (CT/MRI/Scan, Biomarkers

vi Preface

etc.), and tailored treatment strategies in each subgroup of patients. Especially,

radiologic imaging is a new promising tool for this issue and will be presented
in detail with numerous figures. A further key feature is presentation about the
current and future treatment strategies for tailored medicine including broncho-
scopic lung volume reduction, pulmonary hypertension, and comorbidity man-
agement. This textbook will become a great asset in clinical practice and
research to all who are involved or interested in COPD.
I would like to acknowledge the work done by the members of the Korean
Obstructive Lung Disease (KOLD) Cohort Study who contributed to the
preparation of this book. We are especially grateful to all contributing authors
from abroad: Norbert Voelkel, Edwin Silverman, Meilan Han, Paul Jones,
Rubin Tuder, and Nurdan Kokturk. Finally we wish to thank our families for
their patience and consistent support during our academic lives.
I hope that all readers will find these chapters as helpful and insightful as
we have.

Seoul, South Korea Sang-Do Lee


Part I  Overview

1 Definition and Epidemiology of COPD��������������������������������������������  3

Young Sam Kim
2 Risk Factors: Factors That Influence Disease
Development and Progression����������������������������������������������������������  9
Ji Ye Jung
3 Pathology of Chronic Obstructive Pulmonary Diseases��������������  17
Rubin M. Tuder
4 Pathogenesis of COPD��������������������������������������������������������������������  35
Ji-Hyun Lee

Part II  Assessment

5 Pathophysiology of COPD��������������������������������������������������������������   57
Eun Kyung Kim
6 Diagnosis and Assessment of COPD����������������������������������������������  65
Yong Bum Park
7 Symptomatic Assessment of COPD������������������������������������������������  75
Paul W. Jones
8 Imaging of COPD����������������������������������������������������������������������������  87
Sang Min Lee, Sang Min Lee, Song Soo Kim, Hye Jeon Hwang,
and Joon Beom Seo
9 Biomarkers of COPD��������������������������������������������������������������������  129
Ho Il Yoon

Part III  Heterogeneity

10 Phenotypes of  COPD ��������������������������������������������������������������������  147

Jamie Sheth and MeiLan Han
11 Genetics of  COPD��������������������������������������������������������������������������  169
Woo Jin Kim

viii Contents

12 Imaging Heterogeneity of COPD��������������������������������������������������  179

Sang Min Lee and Joon Beom Seo
13 Asthma-COPD Overlap Syndrome����������������������������������������������  189
Chin Kook Rhee
14 The Spectrum of Pulmonary Disease in COPD��������������������������  195
Norbert F. Voelkel, Shiro Mizuno,
and Carlyne D. Cool

Part IV  Management

15 Prevention of  COPD����������������������������������������������������������������������  211

HyoungKyu Yoon
16 Pharmacologic Management��������������������������������������������������������  219
Seong Yong Lim
17 Non-pharmacologic Management: LVR,
Rehabilitation, and Nutrition��������������������������������������������������������  243
Sei Won Lee and Eun Mi Kim
18 Exacerbation of  COPD������������������������������������������������������������������  261
Jin Hwa Lee
19 Comorbidities: Assessment and Treatment��������������������������������  267
Nurdan Kokturk, Ayse Baha, and Nese Dursunoglu
20 Personalized Treatment in COPD������������������������������������������������  299
Jae Seung Lee and Sang-Do Lee

Part V  Prospectives

21 Cohort Study in COPD: Introduction to COPD

Cohorts (The KOLD and COPDGene Study)
and Collaborative Approaches������������������������������������������������������  313
Deog Kyeom Kim
22 Big Data and Network Medicine in COPD����������������������������������  321
Edwin K. Silverman
Index���������������������������������������������������������������������������������������������������������� 333
Part I
Definition and Epidemiology
of COPD 1
Young Sam Kim

Definition of COPD lung function impairment and prognosis [2].

Traditionally, COPD has been classified as
Chronic Obstructive Pulmonary Disease (COPD) chronic bronchitis (CB) and emphysema. CB is
is a common disease and prevalence is increasing defined as the presence of a chronic productive
worldwide. It is characterized by persistent air- cough for 3 months in each of two consecutive
way obstruction that is partially reversible but it years. Emphysema is defined as the destruction
is considered preventable and treatable disease of alveolar walls and permanent enlargement of
now. Airflow limitation is associated with chronic the airspaces distal to the terminal bronchioles.
and abnormal inflammatory response in the air- Current GOLD guidelines do not include the use
ways and the lung to noxious stimuli [1]. Airway of these terms in the definition of COPD. Asthma
obstruction is defined by a reduction of expira- and COPD represent different disease entity with
tory airflow. Generally, forced expiratory volume different pathogeneses and risk factors.
in 1 s/forced volume capacity (FEV1/FVC) ratio Sometimes clinical manifestations of both dis-
of less than 70% after bronchodilator has been eases may overlap in a patient with airway
used to identify COPD patient. The use of lower obstruction and cannot be classified as COPD or
limit of normal (LLN) values has been proposed asthma only. Large population studies show that
to define airflow limitation by spirometry, but some of the patients with airway obstruction are
current Global initiative for chronic Obstructive classified with more than one diagnosis.
Lung Disease (GOLD) and American Thoracic Therefore, overlapping diagnoses of asthma and
Society/European Respiratory Society guidelines COPD has been proposed and it is called COPD
continue to recommend the fixed ratio criteria and Asthma Overlap Syndrome (ACOS) [1].
instead of an LLN for the diagnosis of COPD [1].
Patients with COPD have shown a great deal of
heterogeneity and can be classified according to Epidemiology of COPD
their clinical and radiologic parameters, biomarkers,
COPD is a leading cause of morbidity and
mortality worldwide. The prevalence and bur-
den of COPD is increasing now. It is due to
continued exposure of risk factors especially
Y.S. Kim smoking and aging population. Estimate of
Division of Pulmonology, Department of Internal
Medicine, Severance Hospital, Yonsei University
prevalence and incidence of COPD is different
College of Medicine, Seoul, South Korea according to the study population and diagnos-
e-mail: tic criteria [2, 3].

© Springer-Verlag Berlin Heidelberg 2017 3

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_1
4 Y.S. Kim

Prevalence stage II or higher COPD was 1–10% overall,

8–11% for men, and 5–8% for women [7].
Prevalence of COPD shows remarkable variation In Asia, Nippon COPD Epidemiology (NICE)
due to differences in study population, survey Study was performed to estimate prevalence of
method, and diagnostic criteria [4]. Meta-analysis COPD in Japanese adults. Prevalence of airflow
of 62 studies published between 1990 and 2004 limitation was 10.9%. Among them, 56% of
that included prevalence estimates from 28 differ- cases were classified to mild, 38% moderate, 6%
ent countries reported a pooled prevalence of severe degree of airway obstruction. Airflow lim-
COPD of 7.6%. The prevalence estimate increased itation was more common in males [8]. In South
to 8.9% from epidemiologic studies using spirom- Korea, nationwide epidemiologic survey called
etry data. Consistent with previous observations, Korean National Health and Nutrition
COPD prevalence was higher among studies using Examination Survey III (KNHANES III) was
GOLD criteria to define COPD compared with performed in 2001. The prevalence of airflow
other classification methods. Prevalence was low limitation by GOLD criteria was 17.2% (men,
when it is calculated from self-reporting or physi- 25.8%; women, 9.6%) among adults older than
cian diagnosis of COPD. 45 years. Most of these cases were mild in degree,
In the USA, data from the Third National and only a minority of these subjects had received
Health and Nutrition Examination Survey physician diagnosis or treatment [9]. According
(NHANES III) estimated that 23.6 million adults to the data from the fourth Korean National
(13.9%) met GOLD definition of COPD (stage 1 Health and Nutrition Survey, prevalence of air-
or higher) in 2000 [1]. According to NHANES way obstruction was detected in 8.8% of subjects
data from 2007 to 2010, the prevalence of airway over age 19 and 13.4% of adults older than
obstruction was 13.5% for adults aged 40 years (19.4% of men and 7.9% of women)
20–79 years old, comprising 28.9 million people. [10]. According to population-based survey data
Among them, 15.9 million had mild degree of from seven China provinces/cities, overall preva-
obstruction and 12.9 million showed moderate to lence of COPD over 40 years old was 8.2% (men,
severe obstruction [5]. 12.4%; women, 5.1%). COPD was more com-
The Latin American Project for the Investigation mon in rural residents, elderly patients, smokers,
of Obstructive Lung Disease (PLATINO) exam- and in those who were exposed to occupational
ined the prevalence of post-­bronchodilator airflow dusts or biomass fuels [11].
limitation among persons over age 40 in five major In Africa, median prevalence of COPD based
Latin American cities. The prevalence of COPD on spirometry in persons aged 40 years or older
ranged from 7.8 to 20.0%. The prevalence is was 13.4%. When applied to the appropriate age
higher in men, in older people, and in those with group (40 years or more), which accounted for
lower education, lower body-mass index, and 196.4 million people in Africa in 2010, the esti-
greater exposure to smoking [6]. mated prevalence translates into 26.3 million
In 2002, the Burden of Obstructive Lung (18.5–43.4 million) cases of COPD [12].
Disease (BOLD) project has been proposed to BOLD and PLATINO study which applied
estimate prevalence globally. This is standardized standardized survey methods and used same defi-
and population-based epidemiologic studies. nition of COPD demonstrate a variable preva-
Post-bronchodilator FEV1/FVC ratio of less than lence estimates ranging from 5.1% in Chinese
0.7 was used to define the presence of COPD [1]. women to 22.2% in South African men [2]. In
Participants from 12 countries included in the developed countries, prevalence of COPD is
BOLD study and performed post-bronchodilator 8–10% among adults 40 years of age and older;
spirometry testing and questionnaire survey. The whereas, in developing countries, prevalence
prevalence of COPD that was GOLD stage I or varies significantly among c­ ountries and is diffi-
higher varied across countries and was generally cult to estimate. Recent studies which estimate
greater in men than in women. The prevalence of prevalence change of COPD revealed that
1  Definition and Epidemiology of COPD 5

p­ revalence has been decreased or stabilized in risk factors of COPD were confirmed such as
some developed countries. But most of the world smoking status, male gender, and increasing age
population is still exposed to smoking, biomass [14]. Incidence rate of COPD is reported from
fuels, and other environmental risk factors, and 2000 to 13,500/100,000 person-year worldwide
prevalence is still high and increasing in the later (Table 1.1) [15–18].
part of last century [1]. The pooled prevalence of
COPD was 7.6% from 37 studies. Prevalence of
CB alone was 6.4% and of emphysema alone was Mortality
1.8%. The pooled prevalence from spirometric
survey data was 8.9% [4]. According to NHANES Due to inconsistent COPD coding at the report of
III study of USA, 70% of adults with airflow death and different use of diagnostic criteria,
obstruction had never received the diagnosis of mortality data must be interpreted cautiously.
COPD. The IBERPOC study in Spain also Mortality may be underestimated because of
reported that there was no previous diagnosis of underdiagnosis problem. However, it is clear that
COPD in 78% of identified cases. Underdiagnosis COPD is one of the most important causes of
and undertreatment is still a significant problem death in most countries [13]. According to the
worldwide [13]. World Health Organization, COPD is the fourth
leading cause of death in the world. Approximately
2.7 million deaths from COPD occurred in 2000,
Incidence half of them in the Western Pacific Region espe-
cially in China. Annually 400,000 deaths occur in
In this large population-based cohort study of developed countries [3]. In Europe, mortality
the general Dutch population of 40 years and rates are variable ranging from 20 to 80 per
older, the overall incidence rate of physician- 100,000 population [19]. A report of global bur-
diagnosed COPD was 2.92/1000 person-year. den of disease study that included mortality
Based on these data, the risk to be diagnosed between 1990 and 2010 demonstrates that COPD
with COPD in the coming 40 years was 12.7% is now the third leading cause of mortality in the
for a 40-year-old male and 8.3% for a 40-year- world, although the number of deaths attributed
old female. The incidence increased with age, to COPD decreased from 3.1 to 2.9 million
and was higher in men than in women. Known annually.

Table 1.1  Incidence rate of COPD

Cohort Follow-up of COPD Age
Author Nation (City) Year size period case (years) Incidence rate
Van Durme et al. Netherlands 1990–2004 7983 11 648 ≥55 9.2/1000
(Rotterdam) person-year
Krzyzanowski Poland 1968–1981 4612 13 1864 19–70 5.0/1000
et al. (Cracow) person-year
Huhti et al. Finland 1961–1971 1476 10 1163 40–64 2.0/1000
(Harjavalta) person-year
and 10.0/1000
for smokers
Lindberg et al. Sweden 1996–2003 963 7 45 46–77 6.7/1000
(Norrbotten) (>Gold person-year
II), 91 for >Gold II
(>Gold I) and 13.5/1000
for >Gold I
6 Y.S. Kim

In the period of 1965–1998, death rates from COPD patients, US $5037 for patients with mod-
coronary heart disease, strokes, and other cardio- erate COPD, and US $10,812 for severe COPD
vascular diseases decreased but deaths from patient. COPD exacerbations account for the
COPD increased by 163% [13]. However, accord- greatest proportion of the total cost. In a phar-
ing to the data from the NHANES I and NHANES maco-economic study of COPD patient treated in
III follow-up studies, mortality rate is decreased the outpatient clinic, the average direct cost of
by 15.8% for participants with moderate or acute exacerbation was US $159 [13]. In the
severe COPD and 25.2% for those with mild European Union, the total direct costs of respira-
COPD. Overall mortality of COPD in the USA tory disease are estimated to be about 6% of the
may be decreasing recently [1]. In China, COPD total cost. Medical cost of COPD accounts for
ranks as the fourth leading cause of death in urban 56% of this cost of respiratory disease. Direct
areas and third leading in rural areas. Both crude cost of COPD tends to increase in the elderly age
and age-adjusted COPD mortality rates have fluc- above 65 years old because of frequent use of
tuated but have displayed a decreasing trend from acute healthcare services due to COPD exacerba-
1990 [20]. The World Health Organization (WHO) tions [23]. The DALYs for a specific condition
has predicted that COPD will become the third are the sum of years lost because of premature
most common cause of death in the world by 2030 mortality and years of life lived with disability,
[1]. Other study reports that COPD will become adjusted for the severity of disability. In 1990,
the fourth leading cause of death and 7.8% of COPD was the 12th leading cause of DALYs lost
death worldwide [21]. Mortality was high espe- in the world, responsible for 2.1% of the total.
cially in very severe COPD patients in whom 26% According to the projects, COPD will be the sev-
died after 1 year of follow-up, whereas 2.8% died enth leading cause of DALYs lost worldwide in
among the non-­ COPD subjects [14]. This 2030. In the Global Burden of Disease Study
increased mortality of COPD is mainly caused by 2013 (GBD 2013), migraine, hearing loss,
worldwide epidemic of smoking and aging of the COPD, anxiety, and diabetes are included in the
world population [13]. top ten cause of DALYs lost [24].

Economic and Social Burden References

COPD is associated with significant economic 1. Diaz-Guzman E, Mannino DM. Epidemiology and

burden. In the USA, economic burden of COPD prevalence of chronic obstructive pulmonary disease.
Clin Chest Med. 2014;35(1):7–16.
was estimated to be US $15.5 billion in 1993. 2. Rosenberg SR, Kalhan R, Mannino DM. Epidemiology
Some studies have shown that the cost of hospital of chronic obstructive pulmonary disease: prevalence,
stay represents 40–57% of the total direct costs morbidity, mortality, and risk factors. Semin Respir
generated by patients with COPD, reaching up to Crit Care Med. 2015;36(4):457–69.
3. Lopez AD, et al. Chronic obstructive pulmonary
63% in severe patients. In the USA, the mean disease: current burden and future projections. Eur
cost of hospital admission by COPD in a cohort Respir J. 2006;27(2):397–412.
of patients with severe COPD was estimated to 4. Halbert RJ, et al. Global burden of COPD: system-
be US $7100 [13]. In terms of direct medical atic review and meta-analysis. Eur Respir J. 2006;
costs of COPD in 2005, the cost per patient was 5. Ford ES, et al. Trends in the prevalence of obstruc-
estimated at US $2700–5900 for attributable tive and restrictive lung function among adults in the
costs and to US $6100–6600 for excess costs United States: findings from the National Health and
[22]. Recently estimated direct costs of COPD nutrition examination surveys from 1988-1994 to
2007-2010. Chest. 2013;143(5):1395–406.
are $29.5 billion and the indirect costs $20.4 bil- 6. Menezes AM, et al. Chronic obstructive pulmonary dis-
lion. In a cohort study of 413 patients with COPD, ease in five Latin American cities (the PLATINO study):
direct healthcare costs were US $1681 for mild a prevalence study. Lancet. 2005;366(9500):1875–81.
1  Definition and Epidemiology of COPD 7

7. Buist AS, et al. International variation in the preva- function and the development of chronic obstruc-
lence of COPD (the BOLD study): a population-based tive pulmonary disease in a 13-year follow-up of the
prevalence study. Lancet. 2007;370(9589):741–50. Cracow study. Risk of chronic obstructive pulmonary
8. Fukuchi Y, et al. COPD in Japan: the Nippon COPD disease. Am Rev Respir Dis. 1986;134(5):1011–9.
epidemiology study. Respirology. 2004;9(4):458–65. 17. Huhti E, Ikkala J, Hakulinen T. Chronic respiratory
9. Kim DS, et al. Prevalence of chronic obstructive disease, smoking and prognosis for life. An epidemio-
pulmonary disease in Korea: a population-based spi- logical study. Scand J Respir Dis. 1977;58(3):170–80.
rometry survey. Am J Respir Crit Care Med. 2005; 18. Lindberg A, et al. Seven-year cumulative incidence of
172(7):842–7. COPD in an age-stratified general population sample.
10. Yoo KH, et al. Prevalence of chronic obstructive pul- Chest. 2006;129(4):879–85.
monary disease in Korea: the fourth Korean National 19. Raherison C, Girodet PO. Epidemiology of COPD. 
Health and nutrition examination survey, 2008. Eur Respir Rev. 2009;18(114):213–21.
Respirology. 2011;16(4):659–65. 20. Fang X, Wang X, Bai C. COPD in China: the bur-
11. Zhong N, et al. Prevalence of chronic obstructive den and importance of proper management. Chest.
pulmonary disease in China: a large, population-­ 2011;139(4):920–9.
based survey. Am J Respir Crit Care Med. 2007; 21. Mathers CD, Loncar D. Projections of global mortal-
176(8):753–60. ity and burden of disease from 2002 to 2030. PLoS
12. Adeloye D, et al. An estimate of the prevalence of Med. 2006;3(11):e442.
COPD in Africa: a systematic analysis. COPD. 2015; 22. Mannino DM, Buist AS. Global burden of COPD:
12(1):71–81. risk factors, prevalence, and future trends. Lancet.
13. Chapman KR, et al. Epidemiology and costs of
chronic obstructive pulmonary disease. Eur Respir 23. Bustacchini S, et al. The economic burden of chronic
J. 2006;27(1):188–207. obstructive pulmonary disease in the elderly: results
14. Afonso AS, et al. COPD in the general population: from a systematic review of the literature. Curr Opin
prevalence, incidence and survival. Respir Med. 2011; Pulm Med. 2011;17(Suppl 1):S35–41.
105(12):1872–84. 24. Vos T, et al. Global, regional, and national inci-

15. van Durme YM, et al. Prevalence, incidence, and life- dence, prevalence, and years lived with disability
time risk for the development of COPD in the elderly: for 301 acute and chronic diseases and injuries in
the Rotterdam study. Chest. 2009;135(2):368–77. 188 countries, 1990–2013: a systematic analysis for
Krzyzanowski M, Jedrychowski W, Wysocki the global burden of disease study 2013. Lancet.
M. Factors associated with the change in ventilatory 2015;386(9995):743–800.
Risk Factors: Factors That
Influence Disease Development 2
and Progression

Ji Ye Jung

Genes Gender

The most well-known genetic factor related with COPD has been far more frequent in men than in
COPD is a severe hereditary deficiency of alpha-1 women in regard to patterns of smoking and
antitrypsin (AAT). AAT is the prototypic member occupational exposures. However, COPD-related
of the serine protease inhibitor superfamily of deaths among women continued to increase and
proteins, which have a major role in inactivating it surpassed the number among men by 2000 in
neutrophil elastase and other proteases to main- the United States [6]. Several studies suggested
tain protease–antiprotease balance. Smoking is that women are more susceptible to smoking-­
most important risk factor for accelerating the related decline in lung function than men [7–12],
airflow obstruction and the onset of dyspnea in and women were at a higher risk of hospitaliza-
those with deficiency of AAT [1]. In nonsmokers tion for COPD [10]. Globally, women are more
with AAT deficiency, lung function declined exposed to biomass fuels related with cooking
faster in male and those with increasing age over open fires compared with men, and women
(especially after 50 years old), asthmatic symp- exposed to smoke for cooking had a higher risk
toms, and occupations exposure to airway irri- of COPD [13–17].
tants [2].
Familial aggregation of COPD has been
reported in a few studies [3, 4]. In Danish and Lung Growth and Development
Swedish Twin Registry, genetic factor was related
with approximately 60% of the individual sus- Lung growth and development starts from the
ceptibility to develop severe COPD [5]. Various period of gestation, at the birth, and during the
other genes are being investigated in relation to childhood and adolescence. Airflow limitation
development and progression of COPD in differ- persisted from mid-childhood to adulthood after
ent ethnicities. extreme preterm birth, most evident in those with
neonatal bronchopulmonary dysplasia. They may
experience an earlier and steeper decline in lung
function during adulthood [18, 19]. In relation to
birth weight, a meta-analysis reported positive
J.Y. Jung association between birth weight and FEV1 [20].
Division of Pulmonary, Department of Internal Low birth weight was associated with worse
Medicine, Severance Hospital, Yonsei University
College of Medicine, Seoul, South Korea adult lung function and higher rate of death from
e-mail: COPD in adult life [21]. Poor airway function

© Springer-Verlag Berlin Heidelberg 2017 9

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_2
10 J.Y. Jung

shortly after birth was a risk factor for airflow in the contemporary cohorts between 2000 and
obstruction in young adults [22]. People with 2010 [39]. The hazard ratio for mortality in the
early life disadvantages (e.g., maternal asthma, usual care group compared with the smoking
paternal asthma, childhood asthma, maternal cessation program intervention group was 1.18
smoking, and childhood respiratory infections) (95% CU, 1.02–1.37) during 14.5 years of fol-
have permanently lower lung function and low-up of COPD patients in the Lung Health
showed no catch-up with age with slightly larger Study [40].
decline in lung function increasing the risk of
COPD [23].
 ccupational Exposures to Dusts,
Chemical Agents, Fumes
Exposure to Particles
Occupational exposure contributed to the
Cigarette Smoking development and influenced clinical course of
COPD. According to ecological analysis of
Among the smokers, the proportion of patients international data using BOLD (Burden of
with COPD varies from 12 to 35% with dose Obstructive Lung Disease), PLATINO (Project
response to smoking amount [24, 25]. However, for Investigation of Obstructive Lung Disease),
recent data reported development of COPD in and ECRHS (European Community Respiratory
up to 50% of elderly smokers in Sweden [26]. Health Survey follow-up study), 0.8% of
Among the patients with COPD, ever smokers COPD prevalence increased as 10% of expo-
account for two-third of the prevalence glob- sure prevalence increased [41]. The model pre-
ally [24, 27–30]. Other type of tobacco such as dicted 20% relative reduction in COPD
water pipe negatively affects lung function and prevalence (i.e., 3.4–2.7%) by 8.8% reduction
marijuana is associated with increased respira- in prevalence of occupational exposure. The
tory symptoms suggestive of obstructive lung occupational effect was higher in women than
disease [31, 32]. Children whose mothers in males [41]. Self-reported exposure to vapors,
smoked during pregnancy had significantly gas, dust, or fumes on the longest held job was
lower lung function than did children whose associated with an increased risk of COPD
mothers never smoked. Moreover, effects of (OR = 2.11) [42]. Biological dust increased
exposure to tobacco smoking by the mother risk of chronic obstructive bronchitis
during pregnancy and/or environmental (OR = 3.19), emphysema (OR = 3.18), and
tobacco smoke exposure in the first few years COPD (OR = 2.70). The risk was higher in
of life persist into childhood and may affect the women than in men, and no significant
pulmonary function attained throughout the increased risks for COPD were found for min-
child’s life [33, 34]. eral dust or gases/fumes [43]. Joint exposure to
Smoking cessation brought a small recovery both smoking and occupational factors mark-
in pulmonary function, but ceased to low pulmo- edly increased the risk of COPD (OR 14.1)
nary function at an accelerated rate [35–38]. [42]. Besides causing COPD, occupational
However, reduction in smoking amount did not exposure affected decline of lung function in
demonstrate linear relationship in reduction in COPD. In men with early COPD, continued
the rate of lung function decline in continuing fume exposure was associated with a 0.25%
smokers in the Lung Health Study [37]. predicted reduction in FEV1% predicted every
In the study of 50-year trend in smoking-­ year [44]. Occupational exposure is also related
related mortality in the United States, male and with mortality in COPD. Construction workers
female current smokers with 55 years of age or exposed to inorganic dusts demonstrated
older showed similar relative risks for death increased mortality compared to other unex-
from COPD (25.6 for men and 22.4 for women) posed construction workers [45].
2  Risk Factors: Factors That Influence Disease Development and Progression 11

Indoor Air Pollution In children, changes in air quality (PM10) caused

by relocation and urban traffic/pollutant exposure
Burning biomass fuel (wood, animal dung, and during adolescent growth years have a measur-
crop waste) for cooking and heating in poorly able and potentially important effect on lung
ventilated homes is the major source of indoor air function growth and performance [55–57].
pollution. In rural area where the smoking is less According to cross-sectional study in Germany,
common, indoor pollutants from biomass fuels is 55-year-old women living less than 100 m from a
an important risk factor for COPD [46, 47]. busy road were at the higher risk of developing
Exposure to wood smoke could equal up to 20 COPD than those living farther away (OR = 1.79,
pack-years of active exposure to cigarette smoke 95% CI 1.06–3.02) [58]. However, to determine
[48]. According to meta-analysis, consistent evi- the relationship between chronic exposure to out-
dence was found that exposure to indoor air pol- door air pollution and COPD risk, more precise
lution is a risk for COPD (OR = 2.80) and chronic measurement of pollutants and longer duration of
bronchitis (OR = 2.32), with at least a doubling study are needed.
of risk, despite of marked heterogeneity by both
county and fuel type [49]. At present, a dose–
response relationship and differential toxicity for Socioeconomic Status
different fuel types cannot be defined because of
insufficient information although this analysis The low socioeconomic status is one of the risk
shows with wood smoke being associated with factors for COPD and it is also associated with
the greatest effect [49]. In contrast to that bio- less COPD-related health care utilization [24, 25,
mass fuel exposure is well-known risk factor for 59]. However, its impact on symptoms, lung
COPD in the developing countries with low function, and other indices of COPD such as
socioeconomic status, association between wood morbidity and mortality seems to be second only
and charcoal exposure (OR = 4.5) and COPD was to smoking. Moreover, it is not clear yet whether
reported in European societies, such as Spain indoor/outdoor air pollutants, poor nutrition,
[50]. Higher level of indoor particulate matter infections related with low socioeconomic status
less than 2.5 μm was associated with worse health are the risk factors or low socioeconomic status
status of patients with severe COPD [51]. itself is the significant factor for COPD [60].

Outdoor Air Pollution Asthma/Bronchial Hyperactivity

A few cross-sectional studies consistently Despite distinctive clinical and pathophysiologic

reported that acute increases in air pollution was characteristics at initial diagnosis, epidemiologic
related with acute exacerbation of COPD [52]. studies of asthma and COPD have demonstrated
Increased mortality and higher rates of hospital- that similar feature may develop in two diseases
ization or admission to emergency departments [61]. Asthmatic patient is known to be suscepti-
were observed [52]. Association of air pollution ble to rapid lung function decline. In a longitudi-
with the development of COPD has not been nal Copenhagen City Heart Study of general
established clearly. However, in large samples of population, adults with self-reported asthma had
representative of the English population, increase substantially greater declines in FEV1 over time
in particulate matter less than 10 μm (PM10) level, than those without (38 mL/year vs. 22 mL/year)
nitrogen dioxide (NO) and sulfur dioxide (SO2) [62]. Airway hyperresponsiveness and irrevers-
of 10 μg/m3 was associated with 3 and 0.7% ible airway obstruction are cardinal features of
reduction in adult FEV1 [53]. Similar relationship asthma. In European Community Respiratory
was also observed in Switzerland where PM10, Health Survey of young adults (20–44 years), air-
NO2, and SO2 affected both FEV1 and FVC [54]. way hyperresponsiveness was the second
12 J.Y. Jung

s­ trongest attributable factor (15% of population) childhood reduced adult lung function and was a
with fourfold greater risk of developing COPD risk factor for COPD [21, 63]. The incidence of
[63]. Irreversible airway obstruction (FEV1 < 80% COPD was 20.3 per 1000 person-years among
predicted and reversibility <9% predicted) was HIV-infected patients compared with 17.5 per
developed in 16% of subjects with asthma and 1000 person-years among HIV-uninfected patients
23% had a reduced postbronchodilator transfer [69]. The pathogenesis of COPD and other chronic
coefficient (carbon monoxide transfer factor/ lung diseases in HIV remains unclear. Multiple
alveolar volume < 80% predicted) during interacting factors including increased systemic
21–33 years of follow-up [64]. Among non- and lung oxidative stress, recurrent respiratory
smoker males with asthma during 10-year fol- tract infections and colonization in the setting of
low-­ up study, 23% fulfilled the criteria for aging are likely to be involved [69–73]. According
irreversible airway obstruction and had a steeper to meta-analysis, despite marked heterogeneity,
decline in FEV1 than those without irreversible past history of tuberculosis was associated with
airway obstruction (53 mL/year vs. 36 mL/year) chronic airflow obstruction independent of ciga-
[65]. Asthmatic patients with incomplete revers- rette smoking (OR 1.37–3.13) [74–77]. Delay in
ibility of airflow obstruction (FEV1 ≤ 75% pre- antituberculosis treatments was associated with
dicted despite optimal corticosteroid treatment) higher risk for COPD [78]. Moreover, patients
show more severe asthma and asthma of longer with COPD treated with inhaled corticosteroid are
duration than asthmatic subjects with complete at risk of tuberculosis and NTM pulmonary dis-
reversibility of airflow obstruction (FEV1 > 80% ease [79, 80].
of predicted) suggesting that incomplete revers-
ibility of airflow obstruction may result from
long-standing airway inflammation and associ- References
ated structural changes [66]. In Tucson long-term
cohort study of airway obstructive disease, active 1. Janus ED, Phillips NT, Carrell RW. Smoking, lung
function, and alpha 1-antitrypsin deficiency. Lancet.
asthmatics were ten times higher risk for acquir- 1985;1:152–4.
ing symptoms of chronic bronchitis, 17 times 2. Piitulainen E, Tornling G, Eriksson S. Effect of
higher risk for being diagnosed with emphysema, age and occupational exposure to airway irritants
and 12.5 times higher risk for fulfilling COPD on lung function in non-smoking individuals with
alpha 1-­ antitrypsin deficiency (PiZZ). Thorax.
criteria during 20-year follow-up study [61]. The 1997;52:244–8.
degree of airflow limitation and hyperinflation is 3. Silverman EK, Chapman HA, Drazen JM, et al.
related to the duration of asthma [66, 67]. Genetic epidemiology of severe, early-onset chronic
However, despite similar fixed airflow obstruc- obstructive pulmonary disease. Risk to relatives
for airflow obstruction and chronic bronchitis. Am
tion, those with a history of asthma and those with a J Respir Crit Care Med. 1998;157:1770–8.
history of COPD show different functional and 4. McCloskey SC, Patel BD, Hinchliffe SJ, Reid ED,
pathologic airway inflammation suggesting that Wareham NJ, Lomas DA. Siblings of patients with
asthmatic airway inflammation does not change and severe chronic obstructive pulmonary disease have
a significant risk of airflow obstruction. Am J Respir
does not become similar to the airway inflammation Crit Care Med. 2001;164:1419–24.
characteristics of COPD after development of fixed 5. Ingebrigtsen T, Thomsen SF, Vestbo J, et al. Genetic
airflow obstruction. Therefore, asthma and COPD influences on chronic obstructive pulmonary dis-
should be identified and treated separately [68]. ease—a twin study. Respir Med. 2010;104:1890–5.
6. Mannino DM, Homa DM, Akinbami LJ, Ford ES,
Redd SC. Chronic obstructive pulmonary disease
surveillance—United States, 1971-2000. MMWR
Infections Surveill Summ. 2002;51:1–16.
7. Chen Y, Horne SL, Dosman JA. Increased susceptibil-
ity to lung dysfunction in female smokers. Am Rev
The contribution of infection on development and Respir Dis. 1991;143:1224–30.
progression of airflow limitation is becoming more 8. Gan WQ, Man SF, Postma DS, Camp P, Sin
important. Respiratory infection in infancy or DD. Female smokers beyond the perimenopausal
2  Risk Factors: Factors That Influence Disease Development and Progression 13

period are at increased risk of chronic obstructive 24. Kim DS, Kim YS, Jung KS, et al. Prevalence of

pulmonary disease: a systematic review and meta-­ chronic obstructive pulmonary disease in Korea: a
analysis. Respir Res. 2006;7:52. population-based spirometry survey. Am J Respir Crit
9. Langhammer A, Johnsen R, Gulsvik A, Holmen TL, Care Med. 2005;172:842–7.
Bjermer L. Sex differences in lung vulnerability to 25. Yoo KH, Kim YS, Sheen SS, et al. Prevalence of
tobacco smoking. Eur Respir J. 2003;21:1017–23. chronic obstructive pulmonary disease in Korea: the
10. Prescott E, Bjerg AM, Andersen PK, Lange P, Vestbo fourth Korean National Health and nutrition examina-
J. Gender difference in smoking effects on lung func- tion survey, 2008. Respirology. 2011;16:659–65.
tion and risk of hospitalization for COPD: results 26. Lundback B, Lindberg A, Lindstrom M, et al. Not 15
from a Danish longitudinal population study. Eur but 50% of smokers develop COPD?—Report from
Respir J. 1997;10:822–7. the obstructive lung disease in northern Sweden stud-
11. Xu X, Weiss ST, Rijcken B, Schouten JP. Smoking, ies. Respir Med. 2003;97:115–22.
changes in smoking habits, and rate of decline in 27. Celli BR, Halbert RJ, Nordyke RJ, Schau B. Airway
FEV1: new insight into gender differences. Eur obstruction in never smokers: results from the third
Respir J. 1994;7:1056–61. National Health and nutrition examination survey.
12. Sorheim IC, Johannessen A, Gulsvik A, Bakke PS, Am J Med. 2005;118:1364–72.
Silverman EK, DeMeo DL. Gender differences in 28. Bridevaux PO, Probst-Hensch NM, Schindler C,

COPD: are women more susceptible to smoking et al. Prevalence of airflow obstruction in smok-
effects than men? Thorax. 2010;65:480–5. ers and never-smokers in Switzerland. Eur Respir
13. Dossing M, Khan J, Al-Rabiah F. Risk factors for J. 2010;36:1259–69.
chronic obstructive lung disease in Saudi Arabia. 29. Lamprecht B, McBurnie MA, Vollmer WM, et al.
Respir Med. 1994;88:519–22. COPD in never smokers: results from the population-­
14. Dennis RJ, Maldonado D, Norman S, et al. Wood based burden of obstructive lung disease study. Chest.
smoke exposure and risk for obstructive airways dis- 2011;139:752–63.
ease among women. Chest. 1996;109:55S–6S. 30. Zhou Y, Wang C, Yao W, et al. COPD in Chinese non-
15. Perez-Padilla R, Regalado J, Vedal S, et al. Exposure smokers. Eur Respir J. 2009;33:509–18.
to biomass smoke and chronic airway disease in 31. Raad D, Gaddam S, Schunemann HJ, et al. Effects
Mexican women. A case-control study. Am J Respir of water-pipe smoking on lung function: a systematic
Crit Care Med. 1996;154:701–6. review and meta-analysis. Chest. 2011;139:764–74.
16. Ekici A, Ekici M, Kurtipek E, et al. Obstructive air- 32. Tetrault JM, Crothers K, Moore BA, Mehra R,

way diseases in women exposed to biomass smoke. Concato J, Fiellin DA. Effects of marijuana smok-
Environ Res. 2005;99:93–8. ing on pulmonary function and respiratory com-
17. Peabody JW, Riddell TJ, Smith KR, et al. Indoor air plications: a systematic review. Arch Intern Med.
pollution in rural China: cooking fuels, stoves, and 2007;167:221–8.
health status. Arch Environ Occup Health. 2005; 33. Cunningham J, Dockery DW, Speizer FE. Maternal
60:86–95. smoking during pregnancy as a predictor of lung func-
Carraro S, Filippone M, Da Dalt L, et al. tion in children. Am J Epidemiol. 1994;139:1139–52.
Bronchopulmonary dysplasia: the earliest and per- 34. Gilliland FD, Berhane K, McConnell R, et al.

haps the longest lasting obstructive lung disease in Maternal smoking during pregnancy, environmental
humans. Early Hum Dev. 2013;89(Suppl 3):S3–5. tobacco smoke exposure and childhood lung function.
19. Vollsaeter M, Roksund OD, Eide GE, Markestad T, Thorax. 2000;55:271–6.
Halvorsen T. Lung function after preterm birth: devel- 35. Xu X, Dockery DW, Ware JH, Speizer FE, Ferris BG
opment from mid-childhood to adulthood. Thorax. Jr. Effects of cigarette smoking on rate of loss of pul-
2013;68:767–76. monary function in adults: a longitudinal assessment.
20. Lawlor DA, Ebrahim S, Davey SG. Association of Am Rev Respir Dis. 1992;146:1345–8.
birth weight with adult lung function: findings from 36. Sherrill DL, Holberg CJ, Enright PL, Lebowitz MD,
the British women’s heart and health study and a Burrows B. Longitudinal analysis of the effects of
meta-analysis. Thorax. 2005;60:851–8. smoking onset and cessation on pulmonary function.
21. Barker DJ, Godfrey KM, Fall C, Osmond C, Winter Am J Respir Crit Care Med. 1994;149:591–7.
PD, Shaheen SO. Relation of birth weight and child- 37. Simmons MS, Connett JE, Nides MA, et al.

hood respiratory infection to adult lung function and Smoking reduction and the rate of decline in FEV(1):
death from chronic obstructive airways disease. BMJ. results from the lung health study. Eur Respir
1991;303:671–5. J. 2005;25:1011–7.
22. Stern DA, Morgan WJ, Wright AL, Guerra S,
38. Scanlon PD, Connett JE, Waller LA, et al. Smoking ces-
Martinez FD. Poor airway function in early infancy sation and lung function in mild-to-­moderate chronic
and lung function by age 22 years: a non-selective obstructive pulmonary disease. The Lung Health
longitudinal cohort study. Lancet. 2007;370:758–64. Study. Am J Respir Crit Care Med. 2000;161:381–90.
23. Svanes C, Sunyer J, Plana E, et al. Early life origins 39. Thun MJ, Carter BD, Feskanich D, et al. 50-Year
of chronic obstructive pulmonary disease. Thorax. trends in smoking-related mortality in the United
2010;65:14–20. States. N Engl J Med. 2013;368:351–64.
14 J.Y. Jung

40. Anthonisen NR, Skeans MA, Wise RA, Manfreda J, 56. Nicolai T, Carr D, Weiland SK, et al. Urban traffic and
Kanner RE, Connett JE. The effects of a smoking ces- pollutant exposure related to respiratory outcomes
sation intervention on 14.5-year mortality: a random- and atopy in a large sample of children. Eur Respir
ized clinical trial. Ann Intern Med. 2005;142:233–9. J. 2003;21:956–63.
41. Blanc PD, Menezes AM, Plana E, et al. Occupational 57. Sugiri D, Ranft U, Schikowski T, Kramer U. The
exposures and COPD: an ecological analysis of inter- influence of large-scale airborne particle decline and
national data. Eur Respir J. 2009;33:298–304. traffic-related exposure on children’s lung function.
42. Blanc PD, Iribarren C, Trupin L, et al. Occupational Environ Health Perspect. 2006;114:282–8.
exposures and the risk of COPD: dusty trades revis- 58. Schikowski T, Sugiri D, Ranft U, et al. Long-term
ited. Thorax. 2009;64:6–12. air pollution exposure and living close to busy roads
43. Matheson MC, Benke G, Raven J, et al. Biological are associated with COPD in women. Respir Res.
dust exposure in the workplace is a risk factor for 2005;6:152.
chronic obstructive pulmonary disease. Thorax. 2005; 59. Jung JY, Kang YA, Park MS, et al. Chronic obstructive
60:645–51. lung disease-related health care utilisation in Korean
44. Harber P, Tashkin DP, Simmons M, Crawford L,
adults with obstructive lung disease. Int J Tuberc
Hnizdo E, Connett J. Effect of occupational exposures Lung Dis. 2011;15:824–9.
on decline of lung function in early chronic obstruc- 60. Prescott E, Vestbo J. Socioeconomic status and

tive pulmonary disease. Am J Respir Crit Care Med. chronic obstructive pulmonary disease. Thorax. 1999;
2007;176:994–1000. 54:737–41.
45. Bergdahl IA, Toren K, Eriksson K, et al. Increased mor- 61. Silva GE, Sherrill DL, Guerra S, Barbee RA. Asthma
tality in COPD among construction workers exposed as a risk factor for COPD in a longitudinal study.
to inorganic dust. Eur Respir J. 2004;23:402–6. Chest. 2004;126:59–65.
46. Liu S, Zhou Y, Wang X, et al. Biomass fuels are 62. Lange P, Parner J, Vestbo J, Schnohr P, Jensen

the probable risk factor for chronic obstructive pul- G. A 15-year follow-up study of ventilatory func-
monary disease in rural South China. Thorax. 2007; tion in adults with asthma. N Engl J Med. 1998;
62:889–97. 339:1194–200.
47. Salvi SS, Barnes PJ. Chronic obstructive pulmonary 63. de Marco R, Accordini S, Marcon A, et al. Risk fac-
disease in non-smokers. Lancet. 2009;374:733–43. tors for chronic obstructive pulmonary disease in a
48. Smith KR, Aggarwal AL, Dave RM. Air pollution and European cohort of young adults. Am J Respir Crit
rural biomass fuel in developing countries: a pilot vil- Care Med. 2011;183:891–7.
lage study in India and implications for research and 64. Vonk JM, Jongepier H, Panhuysen CI, Schouten

policy. Atmos Environ. 1983;17:2343–62. JP, Bleecker ER, Postma DS. Risk factors associ-
49. Kurmi OP, Semple S, Simkhada P, Smith WC, Ayres ated with the presence of irreversible airflow limi-
JG. COPD and chronic bronchitis risk of indoor air tation and reduced transfer coefficient in patients
pollution from solid fuel: a systematic review and with asthma after 26 years of follow up. Thorax.
meta-analysis. Thorax. 2010;65:221–8. 2003;58:322–7.
Orozco-Levi M, Garcia-Aymerich J, Villar J, 65. Ulrik CS, Backer V. Nonreversible airflow obstruc-
Ramirez-­Sarmiento A, Anto JM, Gea J. Wood smoke tion in life-long nonsmokers with moderate to severe
exposure and risk of chronic obstructive pulmonary asthma. Eur Respir J. 1999;14:892–6.
disease. Eur Respir J. 2006;27:542–6. 66. Hudon C, Turcotte H, Laviolette M, Carrier G, Boulet
51. Osman LM, Douglas JG, Garden C, et al. Indoor air LP. Characteristics of bronchial asthma with incom-
quality in homes of patients with chronic obstruc- plete reversibility of airflow obstruction. Ann Allergy
tive pulmonary disease. Am J Respir Crit Care Med. Asthma Immunol. 1997;78:195–202.
2007;176:465–72. 67. Cassino C, Berger KI, Goldring RM, et al. Duration of
52. Sunyer J. Urban air pollution and chronic obstruc- asthma and physiologic outcomes in elderly nonsmok-
tive pulmonary disease: a review. Eur Respir J. 2001; ers. Am J Respir Crit Care Med. 2000;162:1423–8.
17:1024–33. 68. Fabbri LM, Romagnoli M, Corbetta L, et al.

53. Forbes LJ, Kapetanakis V, Rudnicka AR, et al.
Differences in airway inflammation in patients with
Chronic exposure to outdoor air pollution and lung fixed airflow obstruction due to asthma or chronic
function in adults. Thorax. 2009;64:657–63. obstructive pulmonary disease. Am J Respir Crit Care
54. Ackermann-Liebrich U, Leuenberger P, Schwartz J, Med. 2003;167:418–24.
et al. Lung function and long term exposure to air pol- 69. Crothers K, Huang L, Goulet JL, et al. HIV infection
lutants in Switzerland. Study on air pollution and lung and risk for incident pulmonary diseases in the com-
diseases in adults (SAPALDIA) team. Am J Respir bination antiretroviral therapy era. Am J Respir Crit
Crit Care Med. 1997;155:122–9. Care Med. 2011;183:388–95.
55. Avol EL, Gauderman WJ, Tan SM, London SJ, Peters 70. George MP, Kannass M, Huang L, Sciurba FC, Morris
JM. Respiratory effects of relocating to areas of dif- A. Respiratory symptoms and airway obstruction in
fering air pollution levels. Am J Respir Crit Care HIV-infected subjects in the HAART era. PLoS One.
Med. 2001;164:2067–72. 2009;4:e6328.
2  Risk Factors: Factors That Influence Disease Development and Progression 15

71. Pacht ER, Diaz P, Clanton T, Hart J, Gadek

76. Menezes AM, Hallal PC, Perez-Padilla R, et al.

JE. Alveolar fluid glutathione decreases in asymp- Tuberculosis and airflow obstruction: evidence from
tomatic HIV-seropositive subjects over time. Chest. the PLATINO study in Latin America. Eur Respir
1997;112:785–8. J. 2007;30:1180–5.
72. Cole SB, Langkamp-Henken B, Bender BS, Findley 77. Lee SW, Kim YS, Kim DS, Oh YM, Lee SD. The
K, Herrlinger-Garcia KA, Uphold CR. Oxidative risk of obstructive lung disease by previous pul-
stress and antioxidant capacity in smoking and non- monary tuberculosis in a country with interme-
smoking men with HIV/acquired immunodeficiency diate burden of tuberculosis. J Korean Med Sci.
syndrome. Nutr Clin Pract. 2005;20:662–7. 2011;26:268–73.
73. Morris AM, Huang L, Bacchetti P, et al. Permanent 78. Lee CH, Lee MC, Lin HH, et al. Pulmonary tuber-
declines in pulmonary function following pneumonia culosis and delay in anti-tuberculous treatment are
in human immunodeficiency virus-infected persons. important risk factors for chronic obstructive pulmo-
The pulmonary complications of HIV infection study nary disease. PLoS One. 2012;7:e37978.
group. Am J Respir Crit Care Med. 2000;162:612–6. 79. Andrejak C, Nielsen R, Thomsen VO, Duhaut P,

74. Allwood BW, Myer L, Bateman ED. A systematic Sorensen HT, Thomsen RW. Chronic respiratory
review of the association between pulmonary tubercu- disease, inhaled corticosteroids and risk of non-­
losis and the development of chronic airflow obstruc- tuberculous mycobacteriosis. Thorax. 2013;68:
tion in adults. Respiration. 2013;86:76–85. 256–62.
75. Lam KB, Jiang CQ, Jordan RE, et al. Prior TB, smok- 80. Kim JH, Park JS, Kim KH, Jeong HC, Kim EK,
ing, and airflow obstruction: a cross-sectional analy- Lee JH. Inhaled corticosteroid is associated with an
sis of the Guangzhou Biobank cohort study. Chest. increased risk of TB in patients with COPD. Chest.
2010;137:593–600. 2013;143:1018–24.
Pathology of Chronic Obstructive
Pulmonary Diseases 3
Rubin M. Tuder

Introduction impact of socioeconomic development and their

ensuing environmental impact that have occurred
In this modern age of in-depth molecular and over the past 500 years. Smoking is a critical deter-
genetic focus on diseases, it is pressing that we minant of COPD development in more than 90% of
revisit the fundamental pathology underlying patients; environmental pollution and, infrequently,
chronic pulmonary obstructive diseases—forget- genetic causes account for a growing number of
ting the past limits our ability to make the best patients. Within this background of epidemiologi-
from the present. This review seeks to integrate, cal and clinical complexity, COPD reflects intricate
when possible, what is known about the pathol- structural alterations within the lung, often the
ogy of chronic obstructive pulmonary diseases focus of pathologists over decades. These patho-
(COPD) with key pathogenetic data. However, to logical descriptions have contributed to forming
move the field forward, investigators dedicated to the foundation of our attempts to understand the
COPD are required to understand the normal and disease. We seek to provide a timely and necessary
diseased lung structure (qualitatively and quanti- review of the pathology of COPD, as many of
tatively), including on how best determine these today’s scholars have limited understanding of the
key parameters; there was a time, approximately scope of the pathological data accumulated in the
more than a half a decade ago, in which these past decades. Investigations in the broad angles of
were the most exciting and hopeful developments COPD require an understanding of the structural
to understand COPD. They form the foundation lung alterations in COPD, the structure of the nor-
to better appreciate the challenge to understand mal and aged lungs, and on how quantitative mea-
COPD and, most importantly, give proper credit sures aid in describing both the normal and COPD
to key studies that, in the past 50 years, shaped lung. However, the “bar” for the description of the
our current understanding of this highly chal- pathology of COPD is high: William Thurlbeck
lenging disease. provides a superlative assessment of role of patho-
COPD, refers to a complex disease, with vary- logical changes underlying chronic airway obstruc-
ing clinical phenotypes, largely resulting from the tion, with a particular emphasis on how they relate
function [1]. We will refer to this publication often,
as it provides valuable insights into the pathology
R.M. Tuder, M.D. of COPD, with reference to studies covering inves-
Program in Translational Lung Research, Division tigations initiated in the 1950s and extending by the
of Pulmonary Sciences and Critical Care Medicine,
time of its publication in 1985. The readers are
University of Colorado School of Medicine,
Aurora, CO 80045, USA strongly encouraged to read this summary to better
e-mail: appreciate the advances made in the field by the

© Springer-Verlag Berlin Heidelberg 2017 17

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_3
18 R.M. Tuder

mid-1980s and the challenges that still remain up stereological methods provide the most accurate
to this date. data not only on the human lung, but that also is
It is highly instructive to go over the extensive required for proper experimental modeling. We
literature of COPD, particularly of the pathology then review studies in the pathology of COPD.
of the disease. While in the late 1950s and early
1960s, there was a more qualitative attempt to
describe the pathological alterations of the lung Lung Volumes in Lung Stereology
aimed at understanding the clinical manifesta-
tions of the disease [2]. This early effort translated Determination of lung volumes is a key parame-
into the need to better quantification of structural ter in lung stereology and to properly interpret
parameters, as means to relate more closely the quantitative lung pathology. It allows to express
alterations in lung structure with the clinical man- quantities in relation to the whole lung rather
ifestations, ultimately providing a clear rationale than fractions (like percentages), correct the val-
for treatment and insights into natural history. ues for the “real volume,” and minimize impor-
Stereology was then incorporated into the analy- tant biases (errors) that most histological
sis of the normal lung, led by Weibel, and estimates impose. Lung volumes can be esti-
expanded to the COPD field by Dunnill and mated from pulmonary function tests when avail-
Thurlbeck. The accumulated knowledge of quan- able. However, in most studies involving human
titative pathology of COPD eventually came to a disease and animal modeling, the lungs are
standstill, as it became apparent that the age of removed and the lung volumes estimated by
tissue quantification of COPD would not lead to water displacement or the Cavaliere’s method.
breakthroughs in the understanding of the disease, An extension of water displacement method is
as it lagged behind the development of pathobio- the determination of volume (mL) based on
logical insights—the field was largely tied up by weight in air − weight in water (g) (which is bet-
the protease/antiprotease hypothesis; this is ter suited for the lung, which may float and not
starkly delineated in the historical publication by displace water correctly). Another alternative is
Snider and colleagues about the definition of the determination of the weight of the lung in
emphysema in the middle eighties [3]. In the late water, which when divided by the specific den-
1990s and clearly into the new century, the field of sity of tissue of 0.96 (g/mL) should provide the
pathogenesis of COPD witnessed a “revolution,” lung volume; the Cavaliere’s method involves
breaking the conceptual constrains provided by slicing through the lung at specific thickness, cal-
the protease/antiprotease imbalance. This is culating the sum of the area of opposing sides of
apparent in several chapters of this book. But once the sectioned slices and multiplying this sum by
again, quantitative pathological examination of the thickness.
the diseased human lung, with the benefit of One of the most instructive studies referenc-
improved lung imaging (like computer tomogra- ing lung volumes obtained by water displace-
phy) and a growing body of knowledge regarding ment is that of Thurlbeck [4]. He studied 25
inflammation, cell signaling, cell death, among individuals with ages ranging from 25 to 79 years.
others, has provided a major step forward in our The lung volumes ranged between 3.3 and 7.5 L
understanding of COPD. As outlined below, we (mean of 4.9 L ± 1.4 SD). They correlated very
also owe James Hogg for his vision, leadership, closely with body length (Pearson’s r: .772), but
creativity, and ability to interface through all these not with age. However, Thurlbeck also used the
domains, the largest contribution in the evolving predicted total lung capacity (TLC), estimated on
insights into the pathology of the disease, which age, gender, and body length. As the lung volume
span almost 5 decades. depends on body length, total lung volume (TLV)
It is our goal in this review to revisit what is and TLC correlated very closely. In his study,
known about the normal lung that informs the TLC ranged between 3.3 and 7.5 L, with a mean
reader about COPD; we underscore which of 5 L. Thurlbeck underscores that it is difficult to
3  Pathology of Chronic Obstructive Pulmonary Diseases 19

achieve accurate and reproducible inflation in lage, are lined by pseudostratified epithelium
different lungs; however, inflation with formalin overlying a submucosa with connective tissue
offers several advantages, with a close correla- and outside rim of smooth muscle cells. They
tion with lung volume and measurements give rise to intermediate structures with alveoli in
obtained with different methods. We discuss their wall called respiratory bronchioles; each
below key quantitative parameters to describe respiratory bronchiole branches 1–3 times prior
normal and diseased alveolar structure, including to giving rise to alveolar ducts (Fig. 3.1d–e)
the mean linear intercept (Lm) and internal sur- (alveolated conduits, flanked proximally and dis-
face area (ISa), which are defined largely reliant tally by other alveolar ducts), and ending in a
on the measured lung volumes. single alveolar sac (which has a blunt end, and is
Overall, lung volumes obtained after inflation lined by alveoli). The primary lobule is consid-
with 10% formalin at 25–30 cmH2O pressure ered to represent the alveolar duct and alveoli it
approximate closely those obtained by X-ray at supplies in the alveolar sac; the acinus corre-
total lung capacity [5], with a correlation coeffi- sponds to the respiratory bronchiole, alveolar
cient of .82. On average, lung volumes are 8% ducts, and alveoli (Fig. 3.1c–e). The secondary
higher than those obtained by X-ray. lobule, which is largely relevant to radiological
imaging, consists of 15–150 primary lobules,
measures 1–3 cm, and is supplied by a terminal
Normal Lung bronchiole, which branches 5–6 times with its
accompanying pulmonary artery. It is often delin-
Overall Structure eated peripherally by connective tissue project-
ing from the pleura. The respiratory zone
The lung can be broadly divided in large, carti- contributes to 90% of the lung volume, with con-
laginous airways, dividing from mainstem bron- ductive airways and large blood vessels (often
chus for six generations and usually larger than hilar) making up the remaining 10% [6]. Table 3.1
2 mm in diameter (Fig. 3.1a, b); terminal bron- summarizes key structural characteristics of the
chioles (Fig. 3.1c–e), which do not have carti- normal lung, detailed below.

a b

d e

Fig. 3.1  Normal adult human lung. (a) Bronchus (B) cartilage and contains epithelial lining and a layer of
with submucosal glands (arrow), seen between the lumi- muscle. Note the partly collapsed normal alveoli. (d, e)
nal lining and inner surface of cartilage (arrowheads). Transition between TB, respiratory bronchiole (RB), and
(b) Intraparenchymal bronchus (B) flanked by a similar alveolar duct (AD in e). RB is lined by an interrupted
size pulmonary artery (PA). (c) Periphery of the lung, airway epithelial cell layer with alveolated tissue within
with a terminal bronchiole (TB), which does not have its walls
20 R.M. Tuder

Table 3.1  Key structural characteristic of normal lung down to the precapillary level of approximately
Lung volumes 15–25 μm in diameter. A summary of the struc-
3.3–7.5 L (mean of 4.9 L ± 1.4 SD) (inflation with ture and branching pattern of pulmonary arteries
10% formalin, Ref. [4]) is available in reference [8].
Number of alveoli The mean linear intercept (Lm) is determined
250–450 × 106 Ref. [6] by the calculation of number of alveolar intercepts
480 × 106 Ref. [7] crossing a linear grid system. As it reflects the
Mean linear intercept interalveolar septal distance, Lm is the most used
Mean 289 μm (Ref. [4]) tool to express and quantify airspace enlargement
226–350 μm (mean 271 ± 33 μ; Ref [6]) in emphysema (see below). It correlates with age
Internal alveolar surface area and does not correlate with body length. However,
Approximately 80 m2 (range 40–100 m2, Ref. [9])
Weibel has pointed out that Lm is affected by infla-
Mean 65 ± 16.5 SD m2 (range 40–100 m2, Ref. [13])
tion pressures and the intercept score includes alve-
Number of bronchiole profiles (per cm2 lung tissue)
olar ducts and small airways [9]. However, the data
0.89 (Ref. [6])
presented in the subsequent sections argue strongly
0.84 (Ref. [15])
for the validity of Lm to assess human emphysema.
Verbeken et al. studied normal lungs, with a mean
Alveolar Tissue age of 49 years [10]. They found an emphysema
score of 1.2 (minimal airspace enlargement in
It is important to revisit some of the key findings of selected cases). Other interesting measures from
Weibel and Gomez in their classic reporting of lung their study consisted of delineation of Lm of
structure using stereology [6]. They summarized 289 μm with a mean of 5.49 intercepts. The coef-
the relative volume contribution of alveoli to ficient of variation of intercepts was 60%. This
approximately 60% of total lung volume, air ducts study provided quantification of the structural com-
to approximately 26%, and tissue to about 4%. ponents that contribute to Lm; the airspace proper
They studied five lungs, ranging in age from 8 to measured on average 265 μm, while the septal wall
74 years of age, with lung volumes between 2.5 and measured 24 μm. Verbeken et al. counted alveolar
7 L. The lung contains approximately 300 × 106 attachments of 6.72/mm of airways; the number of
(range 250–450 × 10 [6], pending stature) alveoli; terminal bronchioles per surface tissue was 0.85,
alveolar ducts and sacs would account for approxi- with 800 μm diameter in average. In this study, Lm
mately 14 × 106 structures. More recently and using correlated inversely with height (which is in con-
newly developed stereological approaches to trast to data from Thurlbeck, see below), and posi-
directly count alveolar structures, Ochs and collab- tively with weight, with added regression power
orators estimated that an adult (between 18–41 years when combined with age [11].
of age) has 480 × 106 alveoli [7]. Interestingly, the Lm according to Thurlbeck has a dispersion of
overall alveolar diameter was lower for younger about 20% around the mean in a normal lung, i.e., in
lungs (around 200 μm) and close to 290 μm in the excess of this limit, it would be considered abnor-
older lungs (these measurements correspond to lung mal. His study of 25 nonemphysematous lungs
inflation to 75% of total lung capacity). (described in more detail below) showed an Lm
The average length of alveolar capillaries ranging from 226 to 350 μm, with a mean of
ranged between 8.2 and 13 μm; there are approxi- 271  μM ± 33. The upper limit, based on the
mately 277 billion capillaries with an average mean + 2 SD, would be 337 μm (or 333 if corrected
diameter of 8 μm. The capillary exchange area for predicted lung capacity) [4]. Verbeken and col-
would be 10% lower than the alveolar area, laborators established this upper limit of normality
accommodating 140 mL of blood. Overall, the to 380 μm. Importantly, Lm correlates with age but
pulmonary arteries follow closely the airway not with body length [4]. Based on data from Weibel
branching (for approximately 23 branches), but and Gomez, the size of alveoli was estimated to be
extending further with a total of about 38 generations in the order of 250–290 μm (i.e., close to the range
3  Pathology of Chronic Obstructive Pulmonary Diseases 21

seen with Lm, pending correction for lung volumes airways or terminal bronchioles have less than
and contraction of tissue after formalin inflation and 2 mm, do not have cartilage in their walls, and are
contraction of tissue after paraffin embedding) [6]. completely surrounded by an epithelium, basal
An important measurement derived from the lamina, and bundles of muscle (Fig. 3.1c–e).
studies by Weibel and Gomez was the alveolar Given the orientation dependency of airways
internal surface area (ISa), representing the over- and pulmonary arteries, elucidation of these
all gas exchange area of the lung provided by the structures (branching, size, etc.) requires either
interface of alveolar septa and alveolar space. casting or imaging in three dimensions after a
The ISa can be determined by the number of radiopaque substance is injected [14]. This
intercepts with a grid of lines. Weibel and Gomez approach allows to divide the lung into three
determined alveolar surface area of approxi- compartments pending their role in gas transport
mately 80 m2 (range 40–100 m2 [9]), or akin the and potential for gas exchange: a conducting
size of a tennis court [12]; it is therefore related to zone, a transition zone, responsible for conduc-
Lm, with a relationship defined by the formula tion and gas exchange, and a respiratory zone,
ISa = 4 × volume alveolar parenchyma/Lm. involved in gas exchange. While Weibel assumed
Thurlbeck reassessed the ISa in 25 nonemphy- the airways branching symmetrically from the
sematous (possibly normal) lungs, which were trachea, Horsfield proposed that they could be
inflated with formalin at 25 cmH2O pressure [4]. asymmetric as well. Weibel counted 16 genera-
Like in the work targeted at emphysematous lungs tions, leading to 216 or 65,536 terminal bronchi-
[13], Thurlbeck introduced some adjustments to oles and 131,071 conducting airways [14];
measurements of Lm and ISa. The total lung vol- however, Horsfield predicted half of this number
umes (TLV) were determined after inflation, by based on asymmetric arrangement of airway
water displacement, therefore including both aer- branching (rather than a given more proximal
ated and tissue parenchyma of the lung. He also branch giving rise consistently to two symmetri-
corrected the TLV by the total lung capacity cal branches). The concept of asymmetry is
(TLC), which represents (only) the aerated vol- important as it can explain a lower dead space
ume of the lung based on predicted values, derived than the symmetrical dichotomous branching
from data that included age, gender, and height. pattern, with gases reaching alveolar units located
The corrective factor was the ratio of TLC/TLV, at at much shorter distances from the trachea. The
2/3 power for ISa and 1/3 power for Lm. average distance for the gas to travel from the lar-
Nonemphysematous lungs showed a wide varia- ynx to the gas exchange units is approximately
tion of ISa, ranging from 40 to 100 m2 (mean 25 cm, largely covered by convective gas move-
65 ± 16.5 SD m2) largely derived from the scatter ment. The final 2.6 mm from the alveolar ducts to
of height. This meant that it is anticipated that a alveoli is covered by oxygen diffusion (in nitro-
tall individual may have a large ISa while a short gen, estimated to be 0.25 cm2/s). The final step
individual may have an ISa of 40 m2. In contrast to involves the passage of oxygen from the alveolar
Lm, ISa correlates closely with height, with an R2 space to capillaries, largely within water (diffu-
of 0.83, adding a potent confounder when to be sion constant for oxygen of 0.00193/cm2/s, i.e.,
used in studies involving emphysema. Lung vol- 1.3 log slower than in nitrogen). Despite the
umes have a greater impact on ISa as MLI does not slower diffusion rate, oxygen would be required
correlate with ISa (Pearson’s r: −0.07). to traverse a shorter distance in the distal lung
(respiratory bronchioles, alveolar ducts, and
alveolar sacs) before reaching the capillaries.
Small Airways Matsuba and Thurlbeck investigated the num-
ber and internal diameter of membranous air-
Airways can be broadly divided based on struc- ways, less than 2 mm, in twenty normal lungs
ture and diameter. Bronchi have cartilage in their [15]. The number of small airways was approxi-
walls and have diameters larger than 2 mm. Small mately 0.84/cm2 of lung tissue and the internal
22 R.M. Tuder

diameter was 0.756 mm (please compare below proposing that alveolar expansion would occur
with the measurements in 12 COPD individuals, during the first 8 years, followed by a significant
with number of airways of 0.638/cm2 and internal decrease afterwards. However, pending the
diameter of 0.738). The number of airways/area height of the child, additional significant alveolar
of lung tissue decreases with height, consistent number increase occurs into early adulthood. A
with the conclusion that the total number of air- recent report [17] using stringent stereological
ways is constant in different height individuals. approaches as recommended by the American
Indeed, Matsuba and Thurlbeck, based on several Thoracic Society (ATS) [18] re-examined
studies, stated that terminal airways would not be whether alveolar numbers increase into adult-
affected by overall distension of the lung (i.e., hood. Their approach was based on randomly
increased TLC in COPD or with age). (done systematically using methods to give all
regions the same probability to be chosen for
analyses) selecting approximately eight blocks
Lung Cellular Composition representative of the right or left lung of 11 sub-
jects, with ages ranging from 1 month to 15 year
The study of Crapo and collaborators continues and 11 months. Alveoli were determined and
being the seminal reporting of key stereological counted by defining their openings in two paral-
data regarding the composition of the alveolated lel sections of predetermined thickness (as
lung [16]. The study was based on eight autop- reported by Ochs et al. [7]). Between 2 and
sied lungs; they were fixed in glutaraldehyde and 4 months of age, the number of alveoli increased
sampled for electron microscopy. Type I cells progressively from approximately 100 × 106 to
covered 93% of the alveolar surface, with a cell around 200 × 106 (n = 8 individuals). The log of
surface area of 5000 μm [2]. Alveolar type II number of alveoli increased with a two-­parameter
cells are 14-fold thicker than type I cells with power function with a fast increase in the first
183 μm [2] surface area (i.e., approximately 1.5 two years of age; it tapered off by adolescence,
log less than type I cells), i.e., covering 7% of the but with continual expansion to the age of about
alveolar surface area. Both type I and II accounted 16 years (an individual with 583 × 106 alveoli).
for 24% of all cells in the lung parenchyma and The log number of alveoli correlated closely with
21% of the total alveolar tissue volume. Capillary weight and height.
endothelial cells are much smaller than type I
cells; they each cover an equivalent 27% of the
alveolar surface area covered by individual type I Aged Lung
cells; the total number of capillary endothelial
cells is 3.6-fold higher than type I cells, collec- The increase in alveoli associated with age has
tively covering almost a similar surface area. been interpreted as “single alveolar enlarge-
Overall, capillary endothelial cells account for ment.” This interpretation stems from the per-
30% of the cells in the alveolus and just 14% of ceived lack of inflammation or other marks of
alveolar tissue volume. The remaining interstitial destructive components, including marked frag-
cells (fibroblasts, macrophages, pericytes, inflam- mentation of elastic fibers, or remodeling of
matory cells, etc.) accounted for 37% of all alve- small airway, or disorganization of alveolar air-
olar cells. way attachments [3].
Weibel and Gomez’s assessments of alveolar
surface area were dependent on age, as the older
Lung Growth individuals had a decrease in the fractional vol-
ume of alveoli from 57% in the younger vs. 52%
A central aspect of lung structure is the develop- in the older lungs, while air ducts increased from
mental growth of the lung. A newborn lung con- 27% in younger vs. 32% in older lungs [6]; the
tains approximately 20 × 106 alveoli, with Dunnill aged lungs had a decrease in alveolated lung.
3  Pathology of Chronic Obstructive Pulmonary Diseases 23

Importantly, this change is paralleled by a ma/TLC and those of airflow did not correlate
decrease in surface area with loss of alveoli. Of with morphological parameters.
interest, this was ascribed to the loss of capillar- As with centrilobular emphysema, there is a
ies by some investigators [19], a concept redis- reduction of membranous bronchiole density in
covered 25 years later to explain the pathogenesis senile lung [10].
of emphysema [20].
Verbeken and collaborators studied group of
senile lungs of individuals aging 69 years. COPD
These lungs showed an enlargement of Lm by
60% over controls, with an emphysema score of There was an extensive focus of investigations in
9.1 (vs. 1.2 in controls) [11]. The Lm exceeded the pathology of COPD in the period extending
the upper normal limit of 380 μm (defined in from the 1950s through the early 1990s. It is
control lungs). The coefficient of variation of apparent that this effort followed in the footsteps
Lm increased to 60%; the mean septal compo- of the studies by Weibel and Gomez revealing
nent of Lm also increased by about 50%; no structural investigations of the human lung (out-
difference in alveolar attachments was noted, lined above [6]) and the need to better understand
tough there was a decrease in the numerical a frequent and complex disease. Driving this
density of small airway/area of lung. In their endeavor was the hope that assessments of struc-
aging group, the expansion of airspaces was tural alterations underlying the pathology of dis-
uniform. Moreover, it appears that this enlarge- ease would provide key explanations regarding
ment occurs without a decrease in alveolar on how best diagnose and treat COPD. Despite
attachments to the bronchiolar wall, which is this intense effort, Thurlbeck recognized the dis-
often seen with more advanced destructive crepancy between structure and function of
(centri- and panlobular) emphysema. Thurlbeck COPD, in particular in regards to airflow limita-
determined that Lm of 25 normal lungs was tion [1], or in other words, inability to explain the
275 μm ± 32, confirming that it increases with chronic airflow limitation with structural data. He
age (Pearson’s r: 0.575) [13]. listed several strong reasons behind this realiza-
Data concerning alveolar internal surface area tion, particularly those related to difficulties
(ISa) in normal individuals is discussed in regard involving the pathological nature of the studies.
to emphysema below [21]. Of note is the progres- In fact, he postulated that every known pathologi-
sive loss of ISa with age after early adulthood, at cal alteration in COPD can explain or contribute
an estimated rate of 2.7 m2/year [13]. These data to chronic airway obstruction, most notably
are largely confirmatory of Thurlbeck’s study on mucus gland enlargement, intraluminal mucus
nonemphysematous lungs [4], which correlated accumulation, alterations in terminal bronchioles,
inversely with age (Pearson’s r: −0.5). and emphysematous destruction of the respira-
Of interest is the correlation of physiological tory units, the acini; however, how each of these
parameters (obtained after death in isolated specific components limits lung function remains
lungs) with structural endpoints [22]. In senile unknown. We provide a summary of key points
emphysema, there is a marked increase in mini- below, emphasizing some key publications.
mal air (ma, air remaining after the air has been
removed from the lung)/Total Lung Capacity
(TLC), though less than in emphysema. Chronic Bronchitis
Moreover, there is a shift for the pressure volume
curves, being intermediate between normals and The increase in mucus gland mass has been
emphysematous. The measures of FEV1, FEV1/ linked to COPD for more than 50 years and semi-
FVC, and airflow were not different between nor- quantitatively assessed by the Reid index [23]:
mals and the lungs with senile emphysema. normal individuals would have mucus glands in
Moreover, the key physiological parameters of less than 50% of the bronchial surface area
24 R.M. Tuder

a b c

d e

f g

h i

Fig. 3.2 COPD lung. (a) Large bronchus (B) with The adjacent alveolar duct (AD) shows enlargement char-
chronic bronchitis. Note the expansion of glands (glands) acteristic of emphysema. (e) Respiratory bronchiolitis
beyond the outer rim of cartilage (arrowheads). (b) Low with thickened terminal bronchiole (TB) with clusters of
magnification of mild emphysema with notable airspace pigmented intra-alveolar macrophages (arrows). (f)
enlargement in the subpleural region (arrows). (c) Subpleural bullae (arrow) with absence of alveolar septa.
Centrilobular emphysema with enlargement of alveolar (g) Marked subpleural airspace enlargement (arrow). (h)
duct (AD) (arrows). (d) Chronic bronchiolitis in a termi- Characteristic thinning of alveolar septa in severe centri-
nal bronchiole (TB) (arrow). Note the increased thicken- lobular emphysema, which appears virtually avascular
ing of the airway wall, largely due to chronic inflammatory (arrows). (i) Pulmonary artery thickening in severe COPD
cells. There is a focal loss of epithelial lining (arrowhead). (arrows)

(Fig.  3.2). In her classic description of chronic Thurlbeck suggested that these airways con-
bronchitis, Lynne Reid underscored that in early tribute significantly to chronic airflow limitation.
cases, there is hypertrophy of mucus elements The pathological counterpart of the clinical char-
and airway luminal accumulation [24]; in acteristics of chronic bronchitis represents the
advanced cases, she highlighted the finding of excessive mucus production. He proposed that
purulent inflammation in large and small air- airflow limitation would (also) ensue due to
ways, with associated dilation of airways and hypertrophy of mucus glands and luminal accu-
sometimes obliterative changes. mulation of mucus. These would correlate with
3  Pathology of Chronic Obstructive Pulmonary Diseases 25

clinical history of chronic bronchitis and sputum obstructing small airways (Fig. 3.2). Additional
production. However, the Reid index follows a contributors consist of goblet cells, which might
modal distribution, with a right shift in normal undergo hyperplasia. An overall replacement of
and left shift in COPD; but, there is significant luminal contents, displacing surfactant, would
overlap. The extremes of both curves are infor- result in airway instability. However, distortion
mative in that Reid index less than 0.36 would be of the small airways and obliteration would also
only seen in normal while higher than .55 only in contribute to severe airflow limitation; these
COPD specimens [1]. would follow extensive emphysema. The discus-
The interaction between mucus gland enlarge- sion that follows largely confirms Thurlbeck’s
ment and emphysema is of interest, possibly predictions of the main contributors for chronic
reflecting that cigarette smoke triggers both airflow limitation, including some more recent
events. The increase in mucus gland mass studies involving COPD lungs.
occurred with age in COPD patients and age Hogg and collaborators provided key physio-
matched controls (mean age of 65 years) to a logical data that supported that the main site of
similar extent; the mean percentage thickness in increased airway resistance in COPD (seven
the control group was approximately 8% vs.12% emphysematous, one with bronchiectasis, and
in the aged groups. There is evidence that the fre- one with bronchiolitis) was the small airways, in
quency of chronic mucus hypersecretion the range of 2 mm in diameter. The increase in
increases with emphysema severity score peripheral resistance was about fourfold when
(obtained pathologically). However, it is unclear compared with control lungs [28]. The authors
whether the extent of mucus gland hypertrophy concluded that the increase in resistance in COPD
relates to the clinical symptoms of chronic bron- lungs could be derived from mucus-obstructed
chitis, like mucus production or cough. A prior small airways, narrowing, or occlusion by fibro-
study of 353 patients at autopsy showed, based sis, as emphasized by Thurlbeck.
on point counting morphometry, that mucus Inflammation of airways has been recognized
glands amounted to 17.6% in smokers and 14.5% since the early description of the pathology of
in nonsmokers [25]; however, no differences COPD by Leopold and Gough in 1957 [29].
were noted with age. Also interestingly, the Inflammation in small airways correlates with
authors state a lack of direct correlation between mild alterations in pulmonary function [1], being
mucus gland hypertrophy and emphysema in this perhaps more important in anteceding airway
cohort of 219 lungs yet the percent mucus gland fibrosis and squamous metaplasia. With worsen-
was higher in emphysematous (18.3%) vs. non- ing COPD, the number of airways with inflam-
emphysematous lungs (14.8%). These findings matory cells including polymorphonuclear cells,
led the investigators to question whether chronic macrophages, eosinophils, CD4, CD8, and B
bronchitis might result from alterations other cells increases [30]. Many airways contain lym-
than mucus gland hypertrophy [26]. phoid follicles, most notably in GOLD stage 3
Overall, mucus gland enlargement shows and 4 (most severe disease), contributing to over-
some degree of correlation with flow rates, but all airway thickness by 3–4-fold when compared
the correlation is weak at best, if not existent. with lungs with GOLD0/1 [30].
Studies by Matsuba and Thurlbeck addressed
the question whether there was a change in
Small Airways ­numbers and internal diameter of small airways,
defined as those less than 2 mm in diameter. This
Thurlbeck proposed that small airway disease study involved 12 individuals with mild COPD
contributes to mild airflow limitation [27]. He based on the Ryder score of 17.8 ± 1.2. As com-
also raised the contribution of inflammation, pos- pared with a control cohort [15], they found a
sibly leading to collapse of bronchioles; fibrosis significant decrease in numbers per unit area and
and muscle hypertrophy could also have a role in when corrected for anticipated lung volume at
26 R.M. Tuder

age of 20 years. When both cohorts were adjusted like airways). No studies have used casting or
to body length, there were no significant differ- three-dimension reconstruction to define how a
ences. Moreover, there were no differences in specific segment behaves in COPD lungs (or
internal diameter. Also, they did not find any cor- branching, as outlined by Horsfield studies) of
relation between measurements of small airways the normal lung. However, narrowing of terminal
with those of emphysema or flow rates. Of inter- bronchioles, assessed by multiple approaches
est, the authors noted a small shift of terminal including volume proportion, bronchiolar diam-
bronchioles measuring 200–400 μm in diameter eter, or frequency of airways less than 350 μm in
while there was a deficit of small airways between diameter, correlates with airflow limitation. This
400 and 600 μm in diameter. However, when correlation is however not as strong as between
they analyzed the ratio of small airway to total degree of emphysema and chronic airway
lung volumes (small airway luminal volume den- obstruction [1].
sity), they found a significant reduction in The topic of small airway pathology in COPD
emphysema vs. normal. This decrement was was more recently revisited by Hogg and collab-
accounted by a decrease in the number of small orators. Using lung resection specimens aimed at
airways and reduction of their size. However, removal of tumors or from patients enrolled in
Matsuba and Thurlbeck considered these small the National Emphysema Therapy Trial (NETT),
airway changes not to have a significant role in Hogg et al. found that with worsening of COPD
increased airway resistance and air flow (assessed by the GOLD score, reflecting worse
limitation. FEV1 [31]) correlated inversely with small air-
The reduction of the density of membranous way (less than 2 mm in diameter) occlusion by
bronchioles was noted by Verbeken and refer- mucus and debris (R = 0.5) [30]. In line with ear-
enced as noted previously. They also noted that lier studies on the behavior of small airway in
with increased MLI in emphysema, there is a COPD and their association with airflow limita-
negative relation with airway diameter (as there tion, Hogg et al. showed that total airwall thick-
is an increase in density of airway less than ness was strongly associated with worsening of
600 μm particularly in the lower lobes). Verbeken COPD [30].
and collaborators also verified a decrease in In a recent study, Hogg and collaborators used
numerical densities (per unit area of lung tissue) a sophisticated CT-based approach to study 2 mm
of terminal bronchioles (0.85–0.51/cm2) [10]. and smaller airways, while relying on histology
The airspace is more heterogeneous than due to to further validate their data [32]. They found a
aging, with increased alveolar septal thickening, progressive decrease in the number of airways of
usually with mild fibrosis. Small airway density 2–2.5 mm with worsening of GOLD stage.
decreased in emphysema lungs; there was a neg- Moreover, there was a dramatic reduction of ter-
ative interaction between MLI and alveolar minal bronchiole cross-sectional area and a
attachments, i.e., the higher the MLI, with more decrease in 89% of terminal bronchiole number.
severe emphysema, the lower the number of This reduction also happened in regions with Lm
alveolar attachments. This supports a causal rela- of less than 489 μm, the upper limit of the normal
tion between small airway remodeling and degree Lm values obtained in this study. The residual
of emphysema. airways had increased wall thickness. The authors
This reduction in selective diameter size suggest that emphysema might in reality start
ranges in COPD lungs (vs. normal lungs) might with the disappearance of terminal bronchioles
reflect progressive airway narrowing; however, [32], which might ultimately account for the
there are important pitfalls in most of the mea- increase in 4–40-fold the peripheral airway resis-
surements performed thus far, as they relied tance observed previously by the authors [28].
largely in planimetric assessments, or via stereol- Mucus in terminal bronchioles is increased
ogy (which cannot resolve measures of changes approximately 15-fold in lungs of patients with
in diameter and numbers of fractal structures, chronic bronchitis with severe emphysema, while
3  Pathology of Chronic Obstructive Pulmonary Diseases 27

it is increased only fourfold in lungs of bronchitis there are forms of alveolar enlargement that are
with no emphysema [1]. In combination with not “destructive” like in age-related enlargement
exudates, mucus plugging can contribute to or overdistension after unilateral pneumectomy
chronic airflow limitation, which was confirmed (referenced as “simple airspace enlargement”).
more recently by Hogg and collaborators [30]. There was an attempt to better define the term
Another potential contributor for chronic flow “destruction” as a reduction in amount of a spe-
limitation could be bronchiole tortuosity, poten- cific tissue; others interpreted destruction as dis-
tially leading to stenotic lesions. The basis of this organization of the alveolar attachments to the
finding could be related to inflammation or terminal and respiratory bronchiole [34]
decrease in alveolar attachments (aa) to airways. (Fig. 3.2d, e). As discussed below, more refined
Based on the study of 41 lungs, Nagai et al. cor- attempts to characterize “tissue destruction”
related aa (both as absolute numbers and in refer- involved the introduction of destructive index
ence to airway diameter) to emphysema [35] or alveolar septal holes [36]. These some-
parameters (score and Lm) and measures of air- what rudimentary and overly simplistic defini-
flow [33]. In summary, they delineated that aa/ tions probably reflect the knowledge of the times,
airway diameter were directly related to the prior to clarification on how cell and tissues can
degree of emphysema, which would be the most “disappear”; in the present days, these processes
proximal cause of airflow limitation. They also have been linked to necrosis, apoptosis, and
determined that the aa correlated with airway autophagy, all of which have been shown to be
deformity. No associations were detected with involved in emphysema.
airway inflammation. This has been confirmed in The etiology of emphysema has remained
more recent studies, with the finding that aa largely undetermined, though recognized in the
decreased from 6.72 to 5.76 [10]. The diameter of 1960s that it involved a unique form of tissue
membranous bronchioles correlates with FEV1/ destruction, labeled as “necrosis” [37]. It was
FVC in the emphysema group. also recognized that, in distinction to other forms
of lung necrosis or injury, in emphysema there
was mild inflammation and, importantly, a scar-
Emphysema ring process. The purported etiological agent
could arrive to the centrilobular regions via the
The present definition of emphysema was intro- airflow (like documented at the time with nitro-
duced in 1985 in a report of a National Heart, gen dioxide) or via the blood vessels. The latter
Lung, and Blood Institute workshop [3]. The was considered despite a lack of morphological
authors’ brief introduction referenced the early evidence at the time of precapillary or capillary
definitions by the World Health Organization and occlusion [37] (Fig. 3.2h). In advanced COPD,
American Thoracic Society, which stated that large areas of alveolar destruction lead to
emphysema involved enlargement of the acinus increased subpleural airspaces, which can form
(anatomic unit formed a respiratory bronchiole, subpleural bullae (Figs. 3.2f, g).
3–4 alveolar ducts, and the alveolar sac) In the time spanning the 1960s, 1970s, and
(Fig. 3.2). The committee recognized the impor- into the 1980s, there was an apparent impetus to
tance of the concept of “destruction” in the defi- introduce quantitative measures of alterations in
nition, which was however not defined in their lungs of COPD patients so to relate structural
prior statement (note the thinning of alveolar changes to clinical presentation and, hopefully, to
septa in emphysema, Fig. 3.2h). Moreover, the pathophysiology of the disease.
document expresses concern with the frequent
finding of increase in airspaces in processes asso-  uantification of Emphysema
ciated with prominent fibrosis, like granulomas We recommend several excellent introductory
(rather than in emphysema, where there is mini- texts regarding stereology [6, 9, 18, 38–40], which
mal fibrosis). Importantly, the report states that are necessary for all investigators interested in
28 R.M. Tuder

the lung, including COPD. A critical requirement Table 3.2  Key structural characteristic of emphysema-
tous lung
is the randomization for unbiased selection of
regions for analysis. This means that all fields Lung volumes
have to be given an equal chance of being repre- Mean for centrilobular emphysema: 6.3 L (range
4.8–7 L); mean for panlobular emphysema: 7.6 L
sented, which is accomplished by specifically
(range 6–10 L), (Ref. [43])
designed sampling approaches. The systematic
Number of alveoli
uniform random sampling (SURS) may provide Mean in centrilobular emphysema 218 × 106 (CI:
the best and most stringent sampling design [38]. 126–310); mean in panlobular emphysema
The main approaches to quantify emphysema in 96 × 106 ± 23 (Ref. [43])
lung slices involved two main methods. The first Parenchyma density
consisted of stereological determination of the Centrilobular emphysema: 50% alveolar density;
relative contribution of enlargement of air ducts 18.5% alveolar duct density; 19.5% centrilobular
space density, 12% tissue and blood vessel density, in
and sacs to the overall lung volume [39] and, the (Ref. [43])
second, involved grading of severity of emphy- Mean linear intercept
sema on paper mounts of lung slices based on Mean 598 μm (range: 472–791 μm; Ref. [44])
comparisons with a range of severities of emphy- Mean 279 μm, 304 μm, 369 μm, and 517 μm in
sema [41]. Two methods are available based on normal, mild, moderate, and severe emphysema lungs,
the latter approach: The Thurlbeck system respectively (Ref. [13])
involves radiating segments from a center point Internal surface area
positioned in the major fissure, usually in the third Mean 52 m2 ± 16.2 SD (range: 28–105); ISA corrected
5 L: 50.8 m2 ± 13.2 (range: 25–96) (Ref. [13])
sagittal slice of lung; each segment is scored
Small airways
between 0 and 3 pending the severity of emphy-
Number of bronchiole profiles (per cm2 lung tissue):
sema, with an overall score ranging between 0 and 0.638 (Ref. [15]), 0.51 (Ref. [10]), 90% reduction over
30. The method by Ryder involves matching a controls (Ref. [32])
paper mount slice to standards ranging from 0 to Internal diameter: 0.738 μm (Ref. [15])
100. The advantages of the point counting Increase in terminal bronchioles measuring 200–
approach consist of its accuracy, simplicity of use, 400 μm in diameter
and independence of shape or complexity of the Deficit of small airways between 400 and 600 μm in
counting objects. It is important to keep in mind diameter
that there is need to increase sampling if there is
significant variability of the parameter in ques-
tion; this applies in particular to centrilobular lungs volumes averaged 6.3 L; alveolar and duct
emphysema. These three approaches have been air amount to a mean of about 68% with centri-
compared for reproducibility [41]; Thurlbeck’s lobular spaces occupying 20% of lung paren-
and Ryder’s scoring are fast and provide a low chyma. All these lungs had mucus gland
intraobserver variation but with a wide interob- enlargement with mucus mass of approximately
server variation [42]. The point counting is the 0.42. In 18 lungs with panlobular emphysema, the
one that takes the longest (3–4 min per read), but lung volumes averaged 7.6 L (range 6.2–10.5 L);
with difficulties of calling a point hit with mild the emphysema volume density was 47% (range
lesions. 30–60%), largely at the expense of a reduction of
alveoli and ducts [43]. Emphysema was found in
Microscopic Assessments 219 of 353 autopsied lungs subjected to point
of Emphysema counting morphometry [25]. Emphysema was
A summary of key changes in emphysema is present in 21/73 nonsmokers, with a percentage
included in Table 3.2. The use of the point count volume of parenchyma involved by emphysema
method, as performed by Dunnill, provided inter- being 1.7% vs. 10.8% in smokers.
esting data [43]. In five lungs with severe centri- One of the most popular methods of measur-
lobular emphysema (three with cor pulmonale), ing airspaces is the mean linear intercept or Lm.
3  Pathology of Chronic Obstructive Pulmonary Diseases 29

Lm may provide the best measurement related to well with a coefficient around 0.5. Only the ISa
panlobular emphysema as its Lm was 598 μm for 5 L showed improved correlations, around
(range: 472–791 μm) in 11 lungs with severe 0.83 [44]. It is remarkable that of nine lungs with
respiratory failure [44]. However, as Thurlbeck mild emphysema based on semiquantitative scor-
recognized in a 1991 report, MLI is insensitive in ing, eight had normal ISa 5 L. Thurlbeck sum-
measuring emphysema, being generally normal marized that ISa is significantly reduced in the
in mild and even moderate emphysema [33]. On severe emphysema group, to levels below 80% of
the other hand, in panlobular emphysema (10 predicted. In fact, the data provided in the tables
lungs [43]), the Lm was 592 μm ± 23.7 (i.e., regarding this study demonstrate that Lm varies
increased about twofold over control value). more in tune with the emphysema score, register-
Internal surface area (ISa): Given the potential ing 279 μm, 304 μm, 369 μm, and 517 μm in nor-
importance of alveolar surface area for gas mal, mild, moderate, and severe emphysema
exchange, it is reasonable to propose that this is a lungs, respectively [13]. Thurlbeck agreed with
key measurement of emphysema. In an early Dinnill’s postulate that ISa may not accurately
study of five lungs with severe centrilobular reflect centrilobular emphysema as the lung vol-
emphysema [43], the surface area averaged umes increase pari passu with increases in Lm,
62.2 m2, close to the normal range, which was possibly due to loss of elastic recoil. Interestingly,
surprising given the severity of the disease. These based on the balance of data, Thurlbeck recom-
five lungs had a somewhat increased lung vol- mended the use of Lm, because of ease of use,
ume (mean of 6.3 L ± 0.85, vs. normal of 6 L). reproducibility, and independence of height and
This compensation of ISa is therefore probably age [13], though recognizing the merits (and
due to the increased lung volumes (suggesting accuracy—despite the lack of a gold standard for
that the ratio of surface/volume may be more emphysema) of so-called semiquantitative
accurate in measuring milder forms of emphy- (called by him as subjective) assessments [13].
sema). In ten lungs with panlobular emphysema, Number of alveoli is an infrequently used
the ISa was 48.7 m2 ± 9.6, therefore reduced by parameter in emphysema. Dunnill counted a
about 50% vs. control. mean number of 218 × 106 (CI: 126–310) [43].
ISa was determined in a study of 29 pairs of Interestingly, only one lung would have a higher
normal lungs. The lungs were inflated at number than the normal established by Weibel
25 cmH2O and the lung volume determined by and Gomez [6], yet still lower than the alveolar
volume displacement, after correction for infla- number of 480 × 106 of Ochs et al. In ten cases of
tion, or corrected by antemortem total lung panlobular emphysema, Dunnill reported
capacity assessed by pulmonary testing, or a 96 × 106 ± 23 alveoli (i.e., reduced by a 60% over
fixed processed lung volume of 5 L. In normal control numbers); the mean alveolar volume
lungs, the ISa ranged between 40 and 100 m2.. increased by twofold vs. normal values [43].
When the total lung capacity and volume of tis- Some other parameters of interest might
sue was set at 5 L (ISa corrected or ISa 5 L), then involve the number of centrilobular spaces and
the effect of body length was decreased [13]. In their diameters. Dunnill found that these numbers
44 pairs of emphysematous lungs, the ISa and ranged between 10,600 and 35,000 (mean of
corrected ISa were significantly decreased, down 18,600), with 3 logs increased volume when
to 28 m2. Point counting correlated very closely compared with alveoli in the same lungs. Their
with semiquantitative assessments of emphy- average diameter was 3.7 mm (i.e., tenfold larger
sema based on paper lung mounts, either scored than a normal alveolus).
by the Thurlbeck method (scores ranging 0–30) Destructive index (DI): The DI originated
or average of the grading between 0 and 3 by from the need to better define the concept of alve-
eight pathologists. The correlation of Lm was olar destruction in emphysema [35]. DI was
also very good, around Pearson’s coefficient of defined as interruptions of the alveolar septa; two
0.8. ISa (or if corrected by TLC) did not correlate or more disruptions qualified as a destroyed
30 R.M. Tuder

a­ lveolus. The original study reported that the DI variant, have been referenced above. More
was higher in smokers’ lungs and correlated with recently, panlobular emphysema has been
pulmonary function testing; DI correlated with described in intravenous Ritalin drug abusers,
Lm only in smokers with an “r” of 0.61 [35]. This possibly related to alteration of pulmonary arter-
assessment appears to be reproducible and of ies occluded by tablet compounds [45].
similar extent in upper and lower lobes. In a sec- Panlobular emphysema predominantly
ond study led by Thurlbeck and collaborators, involves the lower zones, with an overall sym-
they also found that overall, DI increases with metrical expansion of both lungs. When com-
Lm (correlation coefficient of 0.64). When the DI pared with centrilobular emphysema, panlobular
is in areas with frank emphysema, the score cor- emphysema has less bronchiolar abnormalities,
relates well with emphysema scoring. DI including less muscle and fibrosis [46]; when
increases but not significantly in mild emphy- extreme examples of centrilobular vs. panlobular
sema; however, DI increases significantly in are compared, the centrilobular has a lower com-
moderate and severe emphysema [34]. DI corre- pliance; the increased compliance is more appar-
lated with lung volumes at 30 cm water transpul- ent when the Lm is in excess of 360 μm in
monary pressure. The concordance between DI panlobular emphysema. Given the extent and
and Lm in nonemphysematous and mild emphy- uniformity of loss of alveolar tissue, it is conceiv-
sematous lungs suggests that these two parame- able that panlobular emphysema has a more sus-
ters change concordantly. tained and reproducible loss of elastic recoil
A potentially related finding to DI in emphy- (than centrilobular emphysema), therefore
sema is the presence of “holes” detected by scan- account for airflow limitation in this group of
ning electron microscopy [36]. In normal lung, patients.
the holes are largely represented by pores of Other forms include distal acinar emphysema.
Kohn, measuring less than 10 μm in diameter. In It is also called paraseptal, possibly leading to
emphysema, these spaces increase in size and spontaneous pneumothoraces in younger indi-
occurrence, reflecting the formation of fenestrae. viduals. Irregular emphysema is a common path-
Interestingly, normal regions in between areas of ological finding associated with the lung
emphysema have an increase in the diameter and parenchyma adjacent to fibrotic processes.
frequency of holes. These holes may be the early
event of destruction in emphysema. I nvolvement of Cellular Compartments
in Emphysema
 ther Forms of Emphysema
O More detailed studies of the relative contribution
In simple pneumoconiosis of coal workers, there of type I, type II, endothelial cell, and interstitial
is heavy accumulation of coal dust around the collagen and elastin in emphysema became avail-
bronchioles, forming the spidery macula. Its rela- able on two decades later than these earlier studies
tionship to centrilobular emphysema is unclear as [47]. Vlahovic et al. studied lobes obtained from
miners may also have COPD. One interpretation cancer resection, including mild and moderate
is that simple pneumoconiosis may be due to emphysema. Five random blocks were processed
enlargement while centrilobular emphysema for morphometric assessment of these compart-
involves tissue destruction [1]. This is supported ments, with an overall 35 blocks being analyzed.
by the observation that pneumoconiosis usually Lm in the normal lungs was between 200 and
involves the respiratory bronchiole while centri- 260  μm (13 blocks); in mild emphysema, it
lobular emphysema extends distally to involve increased to 260–390 μm, and in moderate emphy-
third-order terminal bronchioles. sema, the Lm was in excess of 390 μm. The key
Panlobular emphysema, characteristic of findings involved a decrease in alveolar and capil-
α1-antitrypsin deficiency, involves uniformly the lary surface area (normalized by basement mem-
lobule; some of the quantifiable pathological brane surface area). The dropout of alveolar
characteristics, as compared with centrilobular epithelial (about 50% in moderate emphysema
3  Pathology of Chronic Obstructive Pulmonary Diseases 31

vs. control lungs) and endothelial cells (reduced Notwithstanding the limitations described
by 66% in moderate vs. normal lung) appeared to above, there is evidence that most patients with
be equivalent with an increase in Lm, consistent chronic airflow limitation have severe emphy-
with a synchronous loss of alveolar septal struc- sema. However, several patients with moderate/
tures; the remaining septa in emphysematous severe emphysema do not show severely impaired
lungs remained constant when compared with nor- airflow, and there is no clear parallel between
mal lungs. Interestingly, the volume of type I and degrees of emphysema and severity of airflow
endothelial cells did not differ in all three groups limitation.
(after normalization with basement membrane Occurrence of cor pulmonale or hypertrophy
surface area). The most dramatic change between of the right ventricle also correlates with emphy-
control vs. emphysema lungs was the thickening sema severity; Thurlbeck describes that less than
of interstitium (from 0.8 volume density for nor- 1% of patients without emphysema have cor pul-
mals vs. 3.1 in moderate emphysema); the thick- monale, while the complication occurs in 5%,
ening involved both elastin and collagen associated 15%, and 40% in patients with mild, moderate,
with increase in volume density of fibroblasts and and severe emphysema, respectively [1].
macrophages. These findings suggest that the loss Assessments of ISa 5 L correlate with DLCO
of alveolar septa in emphysema involves the and to some degree with the ratio of residual vol-
simultaneous disappearance of epithelial and ume/TLC. The concordance of ISa 5 L and
endothelial cells [20] and that residual septa DLCO is apparent with measurements in the
undergo some form of scarring (a finding not reg- range of cutoffs of 75% for ISa 5 L. However,
istered by early pathological studies). there are more exceptions in regard to the antici-
pated correlation when DLCO is less than 80% of
predicted with up to 40% of patients showing
Physiological–Pathological relatively preserved ISa 5 L [3].

Investigations in the past 5 decades have shown a  ulmonary Vascular Structure

low correlation between pulmonary physiology and Function
and degree of emphysema assessed by Lm or
emphysema score. This is particularly apparent It is apparent that in normal smokers and patients
on the basis of a study of 48 well-characterized with COPD, the pulmonary arteries undergo
patients from the NIH Intermittent Positive remodeling (Fig. 3.2I). Intima thickening and
Pressure Clinical Trial [48]. Perhaps illustrating medial hypertrophy in COPD is more prominent
the limitations imposed by studying lung lobes vs. normals, however to a limited extent [50];
resected during cancer resection in smokers, an lungs of patients with mild to moderate COPD,
extensive study of 407 lungs failed to correlate with no evidence of pulmonary hypertension, had
FEV1 with number of alveolar septa anchored on mostly intima remodeling, possibly due to thin-
peripheral airways, small airway remodeling, and ning of the media. These data confirm a prior
airway inflammation [49]. Not surprisingly, lungs study that focused on pulmonary arteries of
from patients with less than 50% predicted FEV1 100 μm in diameter or less [51], which found an
had a significant increase in Lm (which ranged increase in muscularized pulmonary arteries in
from 125 to 175 μm (in severe COPD). However, COPD lungs based on replicated elastic layers in
macroscopic assessment of emphysema, based vessels containing a double layer elastic tissue.
on emphysema score, failed to correlate with the The percentage of thick pulmonary arteries was
extent of decrease in FEV1. These data imply greater in COPD lungs of patients with right ven-
that more subtle anatomical and pathophysiologi- tricular hypertrophy vs. no hypertrophy, and
cal alterations in small airways can explain their extent of centri- and panlobular emphysema [51].
contribution to increased airway resistance [28]. Using angiograms of patients with and without
32 R.M. Tuder

COPD, Horsfield and Thomas reconstructed the 3. Snider GL, Kleinerman LJ, Thurlbeck WM, Bengali
ZH. The definition of emphysema: report of a National,
pulmonary vascular tree larger than 1 mm vascu-
Heart, Lung and Blood Institute. Division of Lung
lar segments. The hierarchical branching order Diseases Workshop. Am Rev Respir Dis. 1985:182–5.
was then established, for three-order generations 4. Thurlbeck WM. The internal surface area of nonemphy-
[52] (for a review of pulmonary artery branching sematous lungs. Am Rev Respir Dis. 1967;95:765–73.
5. Thurlbeck WM. Post-mortem lung volumes. Thorax.
in normal pulmonary arteries, please refer to the
review in Ref. [8]). The most significant pulmo- 6. Weibel ER, Gomez DM. Architecture of the human
nary vascular changes occurred in segments of lung. Use of quantitative methods establishes fun-
orders 2–4, with significantly decreased pulmo- damental relations between size and number of lung
structures. Science. 1962;137:577–85.
nary artery diameters. Interestingly, this study
7. Ochs M, Nyengaard JR, Jung A, Knudsen L, Voigt
confirmed a prior finding of increased diameter M, Wahlers T, Richter J, Gundersen HJ. The number
of the first order pulmonary artery in the hilum. of alveoli in the human lung. Am J Respir Crit Care
Interestingly, the more severe the pulmonary Med. 2004;169:120–4.
8. Tuder RM, Stacher E, Robinson J, Kumar R, Graham
artery remodeling, the less responsive are the
BB. Pathology of pulmonary hypertension. Clin Chest
arteries to supplemental oxygen [53]. Med. 2013;34:639–50.
9. Weibel ER. A retrospective of lung morphometry:
Conclusions from 1963 to present. Am J Physiol Lung Cell Mol
Physiol. 2013;305:L405–8.
We owe to several past and current investiga-
10. Verbeken EK, Cauberghs M, Mertens I, Clement J,
tors for detailed insights into the pathology of Lauweryns JM, van de Woestijne KP. The senile lung.
COPD. These investigations developed in the Comparison with normal and emphysematous lungs.
footsteps of the highly innovative and needed 1. Structural aspects. Chest. 1992;1992(101):793–9.
11. Verbeken EK, Cauberghs M, Lauweryns JM, van de
developments in lung stereology. The fast
Woestijne KP. Anatomy of membranous bronchioles
speed of research in the present days may obvi- in normal, senile and emphysematous human lungs.
ate the need to grasp these data and incorporate J Appl Physiol. 1994;1994(77):1875–84.
state-of-­the-art methods in the assessment of 12. Weibel ER. Design and structure of human lung. In:
Fishman AP, editor. Pulmonary diseases and disor-
pathology in human and experimental disease,
ders. New York: McGraw-Hill; 1980. p. 224–81.
particularly related to COPD. As apparent by 13. Thurlbeck WM. Internal surface area and other mea-
the ATS statement on using stereology for surements in emphysema. Thorax. 1967;22:483–96.
measuring parameters in lung tissue using sec- 14. Horsfield K. Quantitative morphology and structure:
functional correlations in the lung. Monogr Pathol.
tions [18], it is timely that these methods be
used by current investigators. This chapter 15. Matsuba K, Thurlbeck WM. The number and dimen-
sought to highlight several key morphological sions of small airways in emphysematous lungs. Am
parameters as they relate to the pathogenesis of J Pathol. 1972;67:265–75.
16. Crapo JD, Young SL, Fram EK, Pinkerton KE, Barry
COPD. They serve as a guide on how best to
BE, Crapo RO. Morphometric characteristics of cells
interpret them and incorporate the most signifi- in the alveolar region of mammalian lungs. Am Rev
cant alterations present in humans in our Respir Dis. 1983;128:S42–6.
understanding of the disease and how to best 17. Herring MJ, Putney LF, Wyatt G, Finkbeiner WE,
Hyde DM. Growth of alveoli during ­ postnatal
approach it in animal models.
development in humans based on stereological
estimation. Am J Physiol Lung Cell Mol Physiol.
References 18. Hsia CC, Hyde DM, Ochs M, Weibel ER. An official
research policy statement of the American Thoracic
Society/European Respiratory Society: standards
1. Thurlbeck WM. Chronic airflow obstructions: corre- for quantitative assessment of lung structure. Am
lation of structure and function. In: Petty TL, editor. J Respir Crit Care Med. 2010;181:394–418.
Chronic obstructive pulmonary disease. New York: 19. Liebow AA, Gough J. Pulmonary emphysema with
Marcel Decker; 1985. p. 129–203. special reference to vascular changes. Am Rev Respir
2. Mitchell RS, Toll G, Filley GF. The early lesions Dis. 1959;80:67–93.
in pulmonary emphysema. Am J Med Sci. 20. Tuder RM, Yoshida T, Fijalkowka I, Biswal S,

1962;243:409–19. Petrache I. Role of lung maintenance program in the
3  Pathology of Chronic Obstructive Pulmonary Diseases 33

heterogeneity of lung destruction in emphysema. Proc emphysema in humans. Am J Respir Crit Care Med.
Am Thorac Soc. 2006;3:673–9. 1994;150:1411–5.
21. Thurlbeck WM. Internal surface area of normal and 37. Wright GW, Kleinerman J. The J. Burns Amberson
emphysematous lungs. Aspen Emphysema Conf. Lecture: A consideration of the etiology of emphy-
1967;10:379–93. sema in terms of contemporary knowledge. Am Rev
22. Verbeken EK, Cauberghs M, Mertens I, Lauweryns Respir Dis 1963; 88: 605–620
JM, van de Woestijne KP. Tissue and airway imped- 38.
Hyde DM, Harkema JR, Tyler NK, Plopper
ance of excised normal, senile, and emphysematous CG. Design-based sampling and quantitation of the
lungs. J Appl Physiol. 1992;72:2343–53. respiratory airways. Toxicol Pathol. 2006;34:286–95.
23. Reid LM. Pathology of chronic bronchitis. Lancet. 39. Dunnill MS. Evaluation of a simple method of sam-
1956:275–81. pling for quantitative histological analysis. Thorax.
24. Reid LM. Pathology of chronic bronchitis. Lancet. 1964;19:443–8.
1954;266:274–8. 40. Dunnill MS. Quantitative methods in the study of pul-
25. Dunnill MS, Ryder R. A quantitative study of the rela- monary pathology. Thorax. 1962;17:320–8.
tionship between chronic bronchitis, emphysema and 41. Thurlbeck WM, Dunnill MS, Hartung W, Heard BE,
smoking. Chest 1971;59: Suppl 35S+. Heppleston AG, Ryder RC. A comparison of three methods
Bedrossian CW, Anderson AE Jr, Foraker of measuring emphysema. Hum Pathol. 1970;1:215–26.
AG. Bronchial morphometry in emphysema and 42. Dunnill MS. The recognition and measurement of
senescence. Exp Mol Pathol. 1977;27:44–50. pulmonary emphysema. Pathol Microbiol (Basel).
27. Thurlbeck WM. The pathology of small airways in 1970;35:138–45.
chronic airflow limitation. Eur J Respir Dis Suppl. 43. Dunnill MS. Quantitative observations on the anat-
1982;121:9–18. omy of chronic non-specific lung disease. Med
28. Hogg JC, Macklem PT, Thurlbeck WM. Site and
Thorac. 1965;22:261–74.
nature of airway obstruction in chronic obstructive 44. Thurlbeck WM. Measurement of pulmonary emphy-
lung disease. N Engl J Med. 1968;278:1355–60. sema. Am Rev Respir Dis. 1967;95:752–64.
29. Leopold JG, Gough J. The centrilobular form of
45. Schmidt RA, Glenny RW, Godwin JD, Hampson NB,
hypertrophic emphysema and its relation to chronic Cantino ME, Reichenbach DD. Panlobular emphy-
bronchitis. Thorax. 1957;12:219–35. sema in young intravenous Ritalin abusers. Am Rev
30. Hogg JC, Chu F, Utokaparch S, Woods R, Elliott Respir Dis. 1991;143:649–56.
WM, Buzatu L. The nature of small-airway obstruc- 46. Kim WD, Eidelman DH, Izquierdo JL, Ghezzo H, Saetta
tion in chronic obstructive pulmonary disease. N Engl MP, Cosio MG. Centrilobular and panlobular emphy-
J Med. 2004;350:2645–53. sema in smokers. Two distinct morphologic and func-
31. Pauwels RA, Buist AS, Calverley PM, Jenkins CR, tional entities. Am Rev Respir Dis. 1991;144:1385–90.
Hurd SS. Global strategy for the diagnosis, man- 47. Vlahovic G, Russell ML, Mercer RR, Crapo JD.

agement, and prevention of chronic obstructive pul- Cellular and connective tissue changes in alveolar
monary disease. NHLBI/WHO Global Initiative septal walls in emphysema. Am J Respir Crit Care
for Chronic Obstructive Lung Disease (GOLD) Med. 1999;160:2086–92.
Workshop summary. Am J Respir Crit Care Med. 48.
West WW, Nagai A, Hodgkin JE, Thurlbeck
2001;163:1256–76. WM. The National Institutes of Health Intermittent
32. McDonough JE, Yuan R, Suzuki M, Seyednejad N, Positive Pressure Breathing trial—pathology studies.
Elliott WM, Sanchez PG, Wright AC, Gefter WB, III. The diagnosis of emphysema. Am Rev Respir Dis.
Litzky L, Coxson HO, Pare PD, Sin DD, Pierce RA, 1987;135:123–9.
Woods JC, McWilliams AM, Mayo JR, Lam SC, 49. Hogg JC, Wright JL, Wiggs BR, Coxson HO, Opazo
Cooper JD, Hogg JC. Small-airway obstruction and SA, Pare PD. Lung structure and function in cigarette
emphysema in chronic obstructive pulmonary dis- smokers. Thorax. 1994;49:473–8.
ease. N Engl J Med. 2011;365:1567–75. 50. Magee F, Wright JL, Wiggs BR, Pare PD, Hogg JC.
33. Nagai A, Yamawaki I, Takizawa T, Thurlbeck
Pulmonary vascular structure and function in chronic
WM. Alveolar attachments in emphysema of human obstructive pulmonary disease. Thorax. 1988;43:183–9.
lungs. Am Rev Respir Dis. 1991;144:888–91. 51. Scott KW. Quantitation of thick-walled peripheral

34. Saito K, Cagle P, Berend N, Thurlbeck WM. The lung vessels in chronic airways obstruction. Thorax.
"destructive index" in nonemphysematous and 1976;31:315–9.
emphysematous lungs. Morphologic observations 52. Horsfield K, Thomas M. Morphometry of pulmonary
and correlation with function. Am Rev Respir Dis. arteries from angiograms in chronic obstructive lung
1989;1989(139):308–12. disease. Thorax. 1981;36:360–5.
35. Saetta M, Shiner RJ, Angus GE, Kim WD, Wang NS, 53. Barbera JA, Riverola A, Roca J, Ramirez J,

King M, Ghezzo H, Cosio MG. Destructive index: Wagner PD, Ros D, Wiggs BR, Rodriguez-Roisin
a measurement of lung parenchymal destruction in R. Pulmonary vascular abnormalities and ventilation-­
smokers. Am Rev Respir Dis. 1985;131:764–9. perfusion relationships in mild chronic obstructive
36. Nagai A, Inano H, Matsuba K, Thurlbeck WM.
pulmonary disease. Am J Respir Crit Care Med.
Scanning electronmicroscopic morphometry of 1994;149:423–9.
Pathogenesis of COPD
Ji-Hyun Lee

COPD Is an Inflammatory Disease nodes [1]. On activation, these antigen-specific

CD4+ and CD8+ cells and antibody-producing B
COPD has been traditionally viewed as a chronic cells are drawn to the lungs to neutralize the anti-
inflammatory disease, which develops in response gens. CD8+ T cells and natural killer cells con-
to noxious particles or gases, most commonly tribute to cytotoxicity of lung tissue cells through
from tobacco smoking. Inflammation related to the release of the proteolytic enzymes perforin
COPD includes cells and mediators of both and granzyme B [3, 4]. As the disease progresses,
innate and adaptive immunity. tertiary lymphoid aggregates including an oligo-
Exposure to cigarette smoke leads to activa- clonal selection of the B and T cells develop
tion of several pattern recognition receptors around the small airways [5, 6].
(PRRs), either directly or indirectly by damage-­ Even though smoking elicits an inflammatory
associated molecular patterns (DAMPs) released response in the lungs of all smokers, this response
from injured epithelial cells. Activation of PRRs is enhanced and exaggerated in those who
such as Toll-like receptors (TLRs) and receptor develop COPD. This suggests that there is an
for advanced glycation end products (RAGE) in abnormal amplification of the inflammatory
airway epithelium and alveolar macrophages response in the lungs of smokers who develop
leads to release of proinflammatory cytokines COPD, and the intensity of infiltration with acti-
and to attract circulating neutrophils, monocytes, vated inflammatory cells correlates with the
and lymphocytes into the lung [1, 2]. severity of COPD [7–10]. In addition, this inflam-
From these cells, several types of proteases mation persists for several years even after smok-
and oxidants are released and, if not sufficiently ing cessation, suggesting that there was
counterbalanced by antiproteases and antioxi- self-perpetuating mechanisms.
dants, further damage will occur [1, 2]. Immature
dendritic cells pick up antigens released from
damaged tissue and foreign pathogens and pres- Inflammatory Cells in COPD
ent them to naive T cells in the draining lymph
 irway Epithelial Cells
The normal differentiated airway epithelium is
composed of ciliated cells, undifferentiated
J.-H. Lee columnar cells, secretary cells, and basal cells.
Department of Allergy, Pulmonary and Critical Care Ciliated and mucus-producing cells remove
Medicine, CHA Bundang Medical Center, CHA microbes and other foreign particles via muco-
University, Pocheon, South Korea
e-mail: ciliary clearance mechanism. Non-mucus

© Springer-Verlag Berlin Heidelberg 2017 35

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_4
36 J.-H. Lee

s­ecretory cells produce antimicrobial and anti- There is a lot of evidence that macrophages
inflammatory proteins, and basal cells function as play a key role in orchestrating the inflammation
stem/progenitor cells to constantly renew the dif- of COPD through the release of chemokines that
ferentiated cell populations. In addition, airway attract neutrophils, monocytes, and T cells, pro-
epithelial cells provide a physical barrier against viding a cellular mechanism that links smoking
the outside environment via tight- and adherence with inflammation in COPD (Fig. 4.1). Increased
junctions that keep adjacent epithelial cells phys- numbers of macrophages in the lungs of patients
ically connected to each other and prevent pas- with COPD and in the lungs of smokers may
sage of microbes and xenobiotics across the result from increased recruitment of monocytes
epithelial layer [11, 12]. Also, epithelial cells are from the circulation in response to monocyte che-
in fact a rich source of cytokines and chemokines motactic chemokines such as monocyte chemo-
molecules involved in modulating inflammation tactic peptide (MCP)-1, and other CXC
and lung defense mechanisms [13]. chemokines (CXCL1, CCL2) released by macro-
Smoking- and COPD-associated functional phages via interaction with the chemokine recep-
and architectural changes of the airway epithelial tor CCR2 and CXCR2 expressed on monocytes
barrier can also contribute to lung inflammation. [21]. Macrophages also attract neutrophils into
Smoking causes loss of Clara cells in the small the lung via CXCL1 and CXCL8 (also known as
airways, which leads to decreased production of IL-8), which act on CXCR2 expressed predomi-
anti-inflammatory protein secretoglobin 1A1 nantly by neutrophils [22]. Chemokines such as
(also known as Clara cell protein). Squamous CXCL9, CXCL10, and CXCL11 released from
metaplasia, a common histologic lesion in the macrophages are chemotactic for CD8+ T cyto-
airway epithelium of individuals with COPD toxic (Tc) cells and CD4+ Th1 cells, via interac-
[14], is associated with increased production of tion with the chemokine receptor CXCR3
proinflammatory cytokines, interleukin (IL)-1α expressed on these cells [23]. Macrophages also
and IL-1β [15] and decreased expression of anti- release transforming growth factor (TGF)-β and
microbial factors, such as secretory leukoprote- connective tissue growth factor (CTGF), which
ase inhibitor (SLPI) [16]. Further, disorganization stimulate fibroblast proliferation, resulting in
of the junctional barrier in the airway of COPD fibrosis in the small airways.
smokers results in increased permeability of the Alveolar macrophages secrete proteases,
airway epithelium [17, 18], which may allow including matrix metalloproteinase MMP-2,
microbial products or cigarette smoke to diffuse MMP-9, and MMP-12; cathepsins K, L, and S;
through the epithelial layer and activate inflam- and neutrophil elastase, taken up from neutro-
matory cells in the airway mucosa [11]. phils, which may contribute to emphysematous
alveolar destruction [2]. Alveolar macrophages
Alveolar Macrophages from patients with COPD are more activated,
Alveolar macrophages reside on the respiratory secrete more inflammatory proteins, and have
epithelial surface and thus are directly exposed to greater elastolytic activity than those from nor-
the outside environment. Macrophages are mal smokers, which is further enhanced by expo-
responsible for a broad set of host defense includ- sure to cigarette smoke [21, 24].
ing recognition and phagocytosis of pathogenic Mechanism of macrophage activation occurs
material and apoptotic cells. via oxidant-induced inactivation and reduction of
There is a five to tenfold increase in the num- histone deacetylase-2 (HDAC2), shifting the bal-
bers of macrophages in airways, lung paren- ance toward acetylated or loose chromatin,
chyma, and bronchoalveolar lavage (BAL) fluid exposing nuclear factor-κB (NF-kB) sites, and
in patients with COPD. Macrophage numbers in resulting in transcription of MMPs, proinflam-
the airways correlate with the severity of COPD matory cytokines [25]. Corticosteroid resistance
[19] and macrophage numbers in the alveoli cor- in COPD may be linked to the decreased HDAC
relate with the severity of emphysema [20]. activity [26, 27].
4  Pathogenesis of COPD 37




Fibroblast Th1 Tc1 Neutrophil Monocyte


Small airway fibrosis Emphysema Mucus hypersecretion

Fig. 4.1  Macrophages play a key role in orchestrating cells. Release of transforming growth factor (TGF)-β and
the inflammation of COPD through the release of several connective tissue growth factor (CTGF) from macro-
chemokines that attract neutrophils, monocytes, and T phage also stimulate fibroblast proliferation

Alveolar macrophages from COPD patients are induced sputum were correlated with the degree
defective in phagocytic removal of apoptotic cells of airflow limitation [32] and the rate of lung
(efferocytosis), and phagocytic uptake of bacteria, function decline in COPD [33].
which may contribute to the maintenance of Smoking stimulates the granulocyte produc-
chronic inflammation in COPD [28] and chronic tion/release from the bone marrow and prolongs
colonization of the lower airways by bacteria such the survival of neutrophils in the respiratory tract,
as Haemophilus influenzae or Streptococcus pneu- possibly mediated by granulocyte macrophage
monia [29]. The bacterial colonization of lower colony-stimulating factor (GM-CSF) and granulo-
airways may predispose to acute exacerbations or cyte colony-stimulating factor (G-CSF) released
pneumonia in patients with COPD [30]. from alveolar macrophages [34]. Smoking also
causes sequestration of neutrophils in the lung
Neutrophils capillaries by decreasing their deformability [35].
There is a lot of data supporting neutrophils as Neutrophils migrate and accumulate into the
the key effector cells in COPD. Activated neutro- lungs from the circulation under the direction of
phils were increased in number within the spu- various neutrophil chemotactic factors, including
tum of COPD patients [31]. The numbers of leukotriene B4, CXCL1, CXCL5 (ENA-78), and
neutrophils in bronchial biopsy specimen and CXCL8, which are increased in COPD airways [36].
38 J.-H. Lee

The migrating neutrophils release proteinases, tion, CD8+ Tc cells cause cytolysis and apoptosis
such as neutrophil elastase, cathepsin G, and pro- of alveolar epithelial cells through release of per-
teinase-3, as well as MMP-8 and MMP-9, which forins, granzyme B, and tumor necrosis factor-α
may contribute to alveolar destruction and mucus (TNF-α), and there is an association between
hypersecretion from submucosal glands and gob- CD8+ Tc cells and apoptotic alveolar cells in
let cells. emphysema [42].
Interestingly, neutrophils from patients with The numbers of CD4+ T cells are also increased
COPD show marked abnormalities in chemotac- in the airways and lungs of COPD patients. At
tic response with increased migration but adopted least two different types of effector CD4+ T cells
more circuitous pathways [37], which is likely to accumulate in the lungs of patients with stable
increase damage to bystander lung tissue and COPD: Th1 cells and Th17 cells [23, 43].
impede bacterial clearance [38]. Th1 cells produce interferon-γ and promote
accumulation of inflammatory cells to the lungs.
Dendritic Cells Lung lymphocytes isolated from the patients with
Dendritic cells are a specialized population of COPD have higher percentages of Th1 cells and
mononuclear cells responsible for recognition secrete more interferon-γ than in control smokers
and uptake of pathogenic materials. The airways [23]. IL-18, which promotes Th1 cell develop-
and lungs contain a rich network of immature ment, is strongly expressed in alveolar macro-
dendritic cells that are localized near the surface. phages, Tc cells, and epithelial cells in lungs of
Airway dendritic cells reside adjacent to epithe- severe COPD patients [44]. Cytokines of Th1 cells
lial cells and extend cytoplasmic protrusions to participate in perpetuating autoimmune responses
sample luminal antigens and interact with envi- and result in excessive proinflammatory responses
ronmental signals. Upon recognition of antigen, that can lead to uncontrolled tissue damage.
dendritic cells undergo a maturation process and Th17 cells are a distinct lineage of activated
migrate toward the local lymphoid tissues, such CD4+ T cells, mediate immunity against extracel-
as regional lymph nodes or mucosal lymphoid lular pathogens, but have also been implicated in
aggregates. Mature dendritic cells present the autoimmunity [45]. Th17 cells, which secrete
processed antigens to naive T lymphocytes, initi- IL-17A and IL-22, are also increased in airways
ating adaptive immune response to pathogens. of patients with COPD and may play a role in
There is an increased number of dendritic orchestrating neutrophilic inflammation [43, 46].
cells in the airways and alveolar walls of smokers Th17 cells may be regulated by IL-6 and IL-23
[39]. It is likely that the specific subset of den- released from alveolar macrophages.
dritic cells is activated in the lungs of patients Regulatory T cells (Treg) are another subset of
with COPD [40] and is linked to disease severity CD4+ T cells with immunoregulatory functions,
[41]. The role of dendritic cells in COPD is not which inhibit autoimmunity and suppress inflam-
yet defined, but they may form the crucial link mation. Tapering of the immune response by
between the innate and adaptive immune Treg cells protects against uncontrolled inflam-
responses in COPD. mation [47], and reduced Treg cell population
have been found in the lungs of patients with
Lymphocytes COPD [48, 49].
The only significant difference in the inflamma- In severe and very severe COPD patients, B
tory cell infiltrate in asymptomatic smokers and cells are found in large airways and bronchial-­
smokers with COPD is an increase in T cells, associated lymphoid tissues around small airways
mainly CD8+ Tc cells. [7] and the lung parenchyma [6]. B cells might be
The number of lung CD8+ T cells in COPD activated by bacterial or viral antigens as a conse-
increases substantially with the severity of air- quence of the chronic bacterial colonization or
flow limitation and emphysema [3]. On activa- latent viral infection in the airways of these
4  Pathogenesis of COPD 39

patients. The pathogenic role of the B cell response Inflammatory Mediators in COPD
is controversial; it might be protective against
microbial colonization and infection of the lower A lot of inflammatory cytokines and chemokines
respiratory tract or it could be directed against have been found to be associated with COPD [2,
lung tissue antigens, suggesting an autoimmune 59, 60]. Pulmonary epithelial cells when exposed
component in the pathogenesis of COPD. to noxious environmental substances release
TNF-α, IL-1β, GM-CSF, TGF-β1, MCP-1, leu-
Other Cells kotriene B4, and IL-8 [59, 61, 62]. Macrophages
Although mast cells and eosinophils have been also release many kinds of chemokines, such as
traditionally associated with the pathogenetic CCL2, CXCL1, CXCL8, CXCL9, CXCL10, and
mechanisms of allergic asthma, evidence sug- CXCL11, which are chemotactic to monocytes,
gests that mast cells and eosinophils could be neutrophils, and lymphocytes [62]. In addition,
implicated in the pathogenesis of COPD. proteolytic enzymes, e.g., MMP-2, MMP-9,
Histological studies of human lungs have MMP-12, and cathepsins, are also secreted in
shown that as COPD progresses to its severe response to reactive oxygen species (ROS)-rich
stages, mast cell populations undergo changes environment [63]. Transcription factor NF-κB
in density, morphology, and distribution, orchestrates these responses in the cells of
including an increase in the number in airway patients with COPD [64].
lumen [50]. Even though, not all patients with Which cytokines are the major players at each
stable COPD have shown the increased number time point along the evolving stage of COPD
of eosinophils in the airways [51, 52], many remains unclear [2, 60].
reports have consistently shown that the Similar inflammatory mediators increased in
increased numbers of eosinophils in bronchial the circulation of patients with COPD may under-
biopsies or BAL fluid during acute exacerba- lie and potentiate systemic comorbidities fre-
tions of COPD [53, 54]. The presence of eosin- quently seen in patients with COPD.
ophils in patients with COPD predicts a
response to corticosteroids and may indicate
coexisting asthma [52, 55]. Pathogenesis of Emphysema
Natural killer cells are classified as lympho-
cytes on the basis of their morphology, the Largely due to the greater structural similarity of
expression of lymphoid markers, and their origin animal air spaces than airways to human, we
from a common lymphoid progenitor cells. know more about mechanisms involved in
However, they are deemed components of innate emphysema than small airway obstruction.
immunity because they lack antigen-specific cell
surface receptors [56]. Natural killer cells act as
cytolytic effector lymphocytes, which can Protease-Antiprotease Imbalance
directly induce the death not only of virus-­ in the Pathogenesis of COPD
infected cells and tumor cells, but also of dam-
aged structural cells in the lungs [4, 57]. Increased The discovery of severe α1-antitrypsin deficiency
cytotoxic activity of natural killer cells express- in early-onset emphysema patients [65], and the
ing both perforin and granzyme B have been induction of emphysema by intratracheal instilla-
shown in induced sputum of patients with COPD tion of a proteolytic enzyme in experimental ani-
compared with healthy smokers [4]. mals [66, 67], led to the protease-antiprotease
Additionally, natural killer cells cross-talk imbalance hypothesis of emphysema. According
with dendritic cells and promote the maturation to this hypothesis, smoking induces an increased
of dendritic cells by producing interferon-γ and number of neutrophils and macrophages in the
TNF-α [58]. lung and the released proteases from these cells
40 J.-H. Lee

are not fully inhibited by antiproteases, which Neutrophil elastase is a potent elastolytic
lead to proteolysis of lung connective tissue, par- enzyme and its intratracheal injection in experi-
ticularly elastin and also stimulate inflammation mental animals induces emphysema [66, 67]. A
into the lung [68, 69]. pathogenic role of neutrophil elastase in
A large body of literature has been tested the α1-antitrypsin-deficient emphysema is supported
hypothesis that a protease-antiprotease imbal- by the correlation of increased neutrophil elastase
ance is the critical mechanism in the pathogene- concentration with severity of emphysema [72].
sis of emphysema in COPD. Increased elastase activity in patients with
COPD may contribute to inflammation because
Protease-Antiprotease Imbalance fragments of matrix proteins, generated by protease
in α1-Antitrypsin Deficiency activity, have chemotactic activity for neutrophils
There is strong evidence to support this hypothe- and monocytes and may also be proinflammatory
sis as the main pathogenic mechanism in emphy- [73]. Neutrophil elastase is also a potent stimulant
sema associated with severe α1-antitrypsin of mucus secretion in the airways [2].
deficiency. α1-antitrypsin is a 52-kDa single-­ The classic presentation for individuals with
chain glycoprotein with a sequence of 394 amino α1-antitrypsin deficiency is early-onset basal
acids that is synthesized predominantly in the panacinar emphysema. However, in recent years
liver and functions as the major inhibitor of neu- α1-antitrypsin deficiency testing has become
trophil elastase [69]. more widespread, and the variability of the age of
Several mechanisms are related to deficiency presentation has become more apparent as well
of α1-antitrypsin, including total absence of the as variations in the clinical phenotype. Patients
gene, frame shift mutations that lead to prema- may present with bronchiectasis and no emphy-
ture stop codons, as well as point mutations that sema [74], upper zone and centrilobular emphy-
may lead to no production or production of sema [75], as well as the classic lower zone
abnormal α1-antitrypsin phenotypes [70]. panacinar emphysema.
Abnormal α1-antitrypsin is accumulated in the
hepatocytes and has retained a tendency to form Protease-Antiprotease Imbalance
spontaneous polymers both in the serum and tis- in COPD Without Antitrypsin
sues, especially the lung. The most common phe- Deficiency
notype of α1-antitrypsin is the normal M form. Smoking may cause a protease-antiprotease
Affected individuals have two genes and these imbalance by reducing the functional activity of
are usually expressed in a codominant form, thus α1-antitrypsin and by increasing the amount of
heterozygotes are common. The level of elastolytic proteases released in the lung [76, 77].
α1-antitrypsin may be key to the susceptibility to However, most of the α1-antitrypsin in cigarette
develop pulmonary emphysema [69]. MZ hetero- smokers remains active and is therefore still
zygotes have partially reduced serum levels capable of protecting against the increased prote-
(approximately 60% of normal MM homozy- ase burden [78–80].
gotes), and SZ heterozygotes have levels approx- Therefore, further hypotheses have invoked a
imately 40% of the normal MM homozygotes. In contributory role for other proteases and antipro-
contrast, ZZ or Z-null types have levels less than teases imbalance, especially in COPD patients
15% of the normal value. without α1-antitrypsin deficiency.
Elastin is the principal component of elastic Pathological studies of accidentally dying
fibers. Insoluble elastin fibers are formed by com- young smokers reported an increased number of
plex mechanisms from the tropoelastin molecules macrophages in the respiratory bronchioles,
secreted from several cell types. Elastin is an where centrilobular emphysema develops in
important target for proteolytic enzymes, and its smokers without α1-antitrypsin deficiency [81],
destruction results in emphysematous destruction suggesting a potential role of macrophages espe-
and loss of elasticity in the lung parenchyma [71]. cially in centrilobular emphysema.
4  Pathogenesis of COPD 41

Alveolar macrophages may bind and internal- demonstrate progressive airspace enlargement
ize released neutrophil elastase in the lung [82] and enhanced collagen degradation or increased
and alveolar macrophage from smokers with low elastin breakdown without inflammation [98].
attenuation area on CT scan showed higher activ- While extracellular matrix proteolysis by
ity of neutrophil elastase [83]. inflammatory cell-derived proteolytic enzymes is
Alveolar macrophage also releases several a central event in emphysema, it is apparent that
elastolytic enzymes including MMP-2, MMP- it cannot explain the complexity of alveolar
9, MMP-12, cathepsins L and S [84–87], destruction in COPD.
which are not inhibited by α1-antitrypsin.
In addition, non-elastolytic enzyme, MMP-1
(collagenase) from macrophages, has been  xidative Stress in the Pathogenesis
implicated in the pathogenesis of emphysema of COPD
in transgenic mice [88, 89] by degrading
type III collagen [90]. Elastolytic activity of Oxidants include reactive molecules, including
cultured alveolar macrophage from patients free radicals and non-radical reactive species,
with emphysema was increased compared summarized as reactive species comprising ROS
with those of patients with bronchitis or other and reactive nitrogen species (RNS). Normal air-
lung diseases [91]. Emphysematous lung tis- ways are replete with antioxidants, such as gluta-
sue showed significantly higher levels of thione (GSH), urate, ascorbate, and extracellular
MMP-2 (gelatinase A) and MMP-9 (gelatin- superoxide dismutase (ECSOD). The balance
ase B) compared with control tissue [92]. A between oxidants and antioxidants is important
study using immunohistochemistry showed to maintain normal physiologic function in the
increases in MMP-1, MMP-2, MMP-8, and lung. Oxidative stress is a potentially harmful
MMP-9 in lung tissue from COPD patients imbalance between oxidants and antioxidants
compared with controls. MMP-9 is also and occurs when the burden of oxidants is not
known to activate the latent form of TGF-β well counterbalanced by the antioxidant defense
to its active form [93], which could provide a system [99], which may provoke diverse lung
link between enhanced elastolytic activity by pathologies.
MMP-9 and the simultaneous airway fibrosis Although inflammation and protease-­
by activation of TGF-β. antiprotease imbalance have been postulated to
In addition, extracellular matrix proteolytic be critical in cigarette smoke-induced emphy-
degradation fragments may act as chemokines sema, there is considerable evidence that oxida-
and promote inflammation [73, 94, 95]. MMP-9 tive stress plays an important role in COPD
generates a collagen fragment, N-acetyl-proline-­ [100, 101].
glycine-proline tripeptide, which is chemotactic
to neutrophils [73]. Laminin and fibronectin frag-  hysiological Roles of Radical Species
ments are chemotactic to human neutrophils and Experimental studies have identified mitochon-
monocytes [94]. In that way, proteolysis of extra- drial ROS as essential for O2 sensing. Superoxide
cellular matrix generates fragments that may per- radicals may aid in maintaining defined redox
petuate inflammation even after smoking potentials on the cellular level and so regulate
cessation. development and differentiation processes of the
Human alveolar macrophages also release organism. Tyrosine radicals are part of the sub-
endogenous inhibitors of MMPs, TIMP1 and unit of ribonucleoside diphosphate reductase in
TIMP2 [96]. Alveolar macrophages from COPD the synthesis of DNA. Reactive species also con-
patients release less TIMP1 in vitro than those tribute to the synthesis of prostaglandins and leu-
from smokers without COPD and nonsmokers kotrienes. Inflammatory cells not only use but
[97]. TIMP3 is the only TIMP that binds strongly also release oxidizing agents (e.g., myeloperoxi-
to the extracellular matrix. TIMP3 knockout mice dase from neutrophils or eosinophilic peroxidase
42 J.-H. Lee

from eosinophils) to destroy microbes and to pro- phages, resulting in amplification of the inflam-
tect the organs. Some of the reactive species are matory response. Oxidative stress also causes
important signaling mediators; ROS generated alveolar cell apoptosis in the setting of human
by macrophages in inflammatory processes may and experimental emphysema [107–109].
initiate intracellular signal transduction and then Moreover, oxidative stress underlies several of
express cytokines and other mediators of inflam- the mechanisms thought to participate in aging,
mation [102]. The function of nitric oxide is well which is another suggested mechanism of the
documented as a potent vasodilator, neurotrans- pathogenesis of COPD [110].
mitter, and bactericide. In addition, reactive spe- The defense mechanisms of the cell through
cies are closely correlated to cellular apoptosis. the activation of antioxidant system are also sig-
nificantly defected in COPD patients [111]. A
 xidative Stress in the Pathogenesis
O master antioxidant transcription factor, nuclear
of COPD erythroid-related factor 2 (Nrf2), controls the
With a large surface area of 80–100 m2 and a expression of more than 100 gene products,
daily breathing volume of 10,000–20,000 L, the including many important antioxidant enzymes.
lungs are very susceptible to oxidative stress. Disruption of the Nrf2 gene in mice led to earlier
Especially in smokers, the situation is drastic onset and more extensive cigarette smoke-­
because smoking leads to a significant exposure induced emphysema with increased inflamma-
to oxidants [103]. Besides cigarette smoke, urban tion and apoptosis in the lung [112].
(biomass fuel), industrial, and occupational air
pollution provide various exogenous oxidants
that probably are responsible for the high preva-  ell Death and Impaired Repair
lence rates of COPD in nonsmokers in develop- in COPD
ing countries [102].
The enhanced oxidizing environment can Excess oxidative stress causes cell death by non-
facilitate the binding of pathogens or antigens to physiological (necrotic) or regulated pathways
effector cells leading to an innate immune (apoptosis). Apoptotic death of alveolar struc-
system. tural cells (endothelial and epithelial cells) has
Increased expression of multiple inflamma- been shown to play a crucial role in the develop-
tory genes is regulated by acetylation of core his- ment of emphysema.
tones, and HDAC2 suppresses inflammatory Alveolar cell apoptosis was induced by the
gene expression. Oxidants modify HDAC pro- intratracheal instillation of active caspase-3 and
tein, induce proteosomal degradation of HDAC, caused emphysematous change [113]. Direct
and thus decrease HDAC activity [104]. It turns instillation of ceramide, oxidative stress-induced
on the redox-sensitive transcription factors (NF-­ proapoptotic molecule, to the lungs of mice cause
κB and activating protein-1 [AP-1]) resulting in apoptotic cell death and alveolar enlargement
the production of proinflammatory cytokines and [114]. Superoxide dismutase protected against
chemokines, which further aggravate inflamma- ceramide-induced alveolar cell apoptosis and
tion and oxidative stress [102]. alveolar enlargement [115]. These findings indi-
Oxidants inactivate antiproteases (such as cate that alveolar cell apoptosis and oxidative
α1-antitrypsin or secretory leukoprotease inhibi- stress mutually interact to mediate emphysema-
tor) [105] and activate MMPs [102], resulting in tous alveolar destruction.
a protease-antiprotease imbalance in the lungs, Also, vascular endothelial growth factor
and damage the lung matrix protein (e.g., elastin (VEGF) signaling seems to be important for the
and collagen) [106]. In addition, oxidants may maintenance of the integrity of alveolar structure.
interfere with elastin synthesis and repair [101]. Decreased VEGF or VEGF signaling could cause
In COPD, HDAC2 expression and activity are experimental emphysema with apoptotic cell death
reduced in peripheral lung and in alveolar macro- in animals [116, 117]. Decreased expression of
4  Pathogenesis of COPD 43

VEGF and VEGF-receptor 2 expressions were In alveolar tissue context, mesenchymal stem
demonstrated in human emphysema lungs [118] cell (MSC) also has been shown to contribute to
and cigarette smoke reduced the levels of VEGF- regeneration in cigarette smoke-induced emphy-
receptor 2 expression in rat lung and in the lungs sema in rat lung [134, 135]. Amniotic fluid-­
of smokers and COPD patients [119]. derived MSC has generated type II alveolar
Recently, autophagic cell death prior to apop- epithelial cells, integrated at the sites of destruc-
tosis has been implicated in the pathogenesis of tion, and induced local regeneration of the lung
emphysema. Autophagy is a pro-survival mecha- alveolar epithelium [134]. Administration of
nism responding to injury. Lung epithelial cells bone marrow-derived cell, MSC, cell free MSC-­
[120], endothelial cells [121], fibroblasts [122], conditioned media also induced a repair of
and alveolar macrophages [122, 123] initiate emphysema and increased the number of small
autophagy signaling responding to cigarette pulmonary vessels [135]. The reparative effects
smoke exposure. Current evidence suggests that of these stem cells are thought to be achieved due
the abnormal persistence of such signaling or the to paracrine effect rather than stem cell engraft-
inability to complete a physiological autophagic ment because stem cell engraftment rate was very
program may increase cellular stress such as low and cell-free MSC-conditioned media also
endoplasmic reticulum (ER) stress, leading to induced repair.
caspase activation and apoptosis in diseased Finally, lung fibroblasts are considered crucial
lungs [121, 123, 124]. In the specimens of COPD for maintenance of the integrity of alveolar struc-
patients, markers of autophagy were upregulated ture by producing extracellular matrix proteins,
in the early stage of disease progression, while including collagen, elastin, and fibronectin,
caspase activation was examined at the later which are required for attachment, structure, and
stages of COPD. This might indicate that autoph- function of alveolar epithelial cells [136].
agy is attempting to protect during the initial Cigarette smoke extract suppresses proliferation
stages from cigarette smoke-induced stress [27]. [137], apoptotic death [138], senescence [139],
Although, significant regeneration and repair and inhibition of collagen gel contraction [140]
are possible after physiologic insults, including in lung fibroblasts in vitro. Similarly, lung fibro-
pneumonectomy and severe respiratory infection blasts from patients with COPD showed similar
[125–128], the ability of the adult lung to repair effects [141, 142], suggesting that cigarette
damaged alveoli appears limited. It is unlikely smoke may interfere with reparative roles of lung
that the process of septation that is responsible fibroblasts.
for alveologenesis during lung development can The relentless lung injury due to oxidant
be reinitiated. Also, it appears difficult for an exposure, disruption of the balance between cell
adult human to completely restore an appropriate death and replenishment of structural cells along
extracellular matrix, particularly functional elas- with the potential exhaustion of lung mainte-
tic fibers. nance program ultimately leads to emphysema.
Some compensatory lung tissue regrowth is
possible, for which several factors, such as, thy-
roid transcription factor 1, VEGF, hypoxia-­  ther Pathogenetic Mechanism
inducible factor 1, and keratinocyte growth factor of Emphysema
are known to be responsible [129–132].
It is well known that the type II alveolar epi- Accelerated Aging
thelial cells have regenerative capacity. In a Oxidative stress induce a number of molecular
recent study, type II epithelial cells can be self-­ and cellular mechanisms associated with cellular
renewed in culture and differentiated into senescence including accumulation of DNA
alveolar-­ like structures “alveolospheres” con- damage [143], impairment of DNA repair [144]
taining type II and cells expressing type I epithe- epigenetic modifications in nuclear DNA [145],
lial cell markers. [133]. protein damage [146].
44 J.-H. Lee

There is growing evidence that emphysema dependent protein/histone deacetylase. SIRT-1

may be caused by accelerated aging of the lungs. also regulates NF-κB-dependent transcription
Emphysema and cellular senescence share some and cell survival in response to TNF-α [159].
features, inflammation and oxidative stress, Environmental stress, such as cigarette smoke
which are the main pathogenic mechanisms of exposure, reduced activity and expression of
COPD [110, 147]. SIRT-1 via increased expression of MMP-9
In addition, several aging animal models have [160]. SIRT-1 is decreased in lung cells from
been associated with emphysema. The klotho patients with COPD, compared with smokers
gene encodes a membrane protein that is a regu- without COPD as a result of post-translational
lator of oxidative stress and cell senescence oxidative modification [161]. This would accel-
[148]; mice with a defect of the klotho gene erate the process of aging and also enhance
develop a syndrome resembling aging including inflammation.
emphysema [149]. Senescent marker protein-30 In conclusion, cellular senescence induced by
(SMP-30) is expressed in early life and progres- ROS further impair tissue repair in response to
sively decreases with age [150]; SMP-30 knock- repetitive cigarette smoke-induced injury of the
out mice develop distal airspace enlargement lungs.
indicative of emphysema [151].
Telomeres are DNA caps located at the end of  utoimmunity and COPD
chromosomes, protecting DNA against degrada- Autoimmunity has been proposed as a late patho-
tion and remodeling, and preventing gene muta- genic event in the progressive course of
tion that may lead to cancerous changes [152]. COPD. Cigarette-induced lung injury may
Owing to the end-replication problem in mature uncover previously sequestered autoantigens, or
somatic cells, telomere repeats are lost with each cigarette smoke may damage lung interstitial and
replicative cycle until a critical length is reached, structural cells and make them antigenic [57].
at which point cells undergo apoptosis or other Several autoantibodies have been found in the
disruptive events. This entire process is acceler- circulation of patients with COPD, particularly in
ated by the presence of ROS or inflammation severe disease [162, 163]. Antibodies against pri-
[153], leading to short telomeres and telomere mary pulmonary epithelial cells, including anti-­
length is well-known biomarkers of biological Hep-­2 epithelial cell IgG autoantibodies, were
aging. From patients with COPD, telomeres of found more frequently in patients with COPD
blood leukocytes are shorter compared with con- than in controls [164]. Even though the data are
trol subjects [154]. conflicting, antielastin antibodies also have been
As other cellular senescence markers, expres- proposed to drive COPD progression [48, 165].
sion of senescence-associated β-galactosidase Lung vascular endothelial cells also can be
(SA–β-gal) and cyclin-kinase inhibitors p16 and exposed to autoimmune attacks. Xenogeneic
p21, and detection of DNA repair/damage can be endothelial cells injected in rats induced autoan-
used [155]. Cigarette smoke extract leads to tibody production that subsequently leads to lung
increased SA–β-gal expression in cultured type II emphysema [166]. Corroboratively, another
cells [156] or lung fibroblasts [139]. Lung fibro- study demonstrated that patients with COPD
blasts harvested from emphysematous lungs also have significantly higher serum antiendothelial
showed increased SA–β-gal expression [157]. cell antibodies than subjects without COPD
Alveolar epithelial and endothelial cells in [167].
emphysematous lungs showed increased expres- In addition, it was reported that antinuclear
sion of p21 in association with decreased telo- autoantibodies were more prevalent in patients
mere length [158]. with COPD than in healthy controls, but these
Sirtuin I (SIRT-1) is an anti-aging and anti-­ molecules were not associated with smoking sta-
inflammatory protein, essential for maintaining tus or FEV1, even though, the pathological impor-
silent chromatin via metabolic NAD(+)- tance of these observations is unclear [168, 169].
4  Pathogenesis of COPD 45

Consequences of such self-attack induce lung lungs of smokers and COPD patients. Cigarette
tissue damage and further lead to structural alter- smoke extracts derived reactive species down-
ations and creation of neoantigens [170]. regulated SIRT1 protein level in human
Autoimmunity to these modified host molecules monocyte-­macrophage cells via post-­translational
could be important in the pathogenesis of modifications, leading to increased level of pro-
emphysema. inflammatory cytokines [161].
Given their demonstrated modifiable nature,
 pigenetic Changes in COPD
E epigenetic mechanisms may open new possibili-
Development ties for therapeutic intervention.
A lot of evidence support that individual suscep-
tibility is important in the development of
COPD. Many candidate genes that could be Pathogenesis of Airway Obstruction
linked to the development of COPD have been
examined; however, the results have been quite I nfiltration of Inflammatory Cells
inconsistent, suggesting that not only genetics, in Small Airways
but also other factors, may contribute to the sus-
ceptibility of COPD. Progression of COPD is associated with the infil-
Epigenetic regulation can affect the transcrip- tration of the airway wall by innate and adaptive
tional activity of specific genes leading to altera- inflammatory immune cells and the percentage of
tion of gene expression patterns upon airways that containing these cells increased as
environmental stimulus without modifying the COPD progressed [7]. With increasing disease
DNA sequence. Typical epigenetic changes are severity, there is also increased mucus hyperse-
post-translational modifications of histones, cretion, submucosal gland hypertrophy, peribron-
DNA methylation, which modulate gene expres- chial fibrosis, and an increase in airway smooth
sion without altering the sequence of the target muscle mass. Although, it is likely that infiltrat-
genes. ing inflammatory cells contribute to the structural
There is emerging evidence supporting a role abnormalities in airway, evidence linking these
of epigenetics in the regulation of inflammatory processes remains scarce.
genes in COPD.
HDACs are key enzymes in the control of pro-
inflammatory cytokines. Oxidants reduce expres- Mucus Hypersecretion
sion levels/activity of these deacetylases.
Therefore, reduced HDAC2 expression and activ- The accumulation of mucus exudates in the air-
ity in the lung and alveolar macrophages corre- way lumen is characteristic in advanced COPD
lated with the disease severity and the intensity of [7]. Cigarette smoke that inhibits the cystic fibro-
the inflammation in COPD [26]. Overexpression sis transmembrane conductance regulator
or knockdown of HDAC2 in alveolar macro- (CFTR) function provides a mechanistic link
phages from COPD patients affected not only the between smoking and abnormal mucous secre-
inflammatory response but also the corticosteroid tion by airway epithelial cells [173].
responsiveness [27]. The increased expression of the mucin MUC5B
SIRT1 is a histone and protein deacetylase, in the bronchiolar lumen and the mucin MUC5AC
which deacetylates histones (H3 and H4) and in the bronchiolar epithelium were showed in the
non-histone proteins including transcription fac- lungs of COPD patients [174]. Activation of a
tors, co-activators, and other signaling molecules ligand-dependent EGFR signaling mediated by
(e.g., FOXO, p53, and RelA/p65) [171]. It is rec- neutrophil elastase or MMP-9 is also responsible
ognized that SIRT1 modulates NF-kB-dependent for the mucin synthesis [175–177]. Mucus-
transcription [159] and inhibits the activity of secreting goblet cells are increased in the small
NF-kB [172]. SIRT1 expression is reduced in the airways of patients with COPD [178]. Goblet cell
46 J.-H. Lee

hyperplasia may be caused by the activation of the Airway smooth muscle cells also have poten-
epidermal growth factor receptor (EGFR), which tial to contribute to the inflammatory and remod-
may be upregulated by oxidants in cigarette smoke eling processes in the small airways [191]. Recent
and by cytokines, such as TNF-α, IL-8, or IL-13 study reported that oxidative stress induces mito-
[179, 180]. Neutrophils may directly cause degran- chondrial dysfunction in airway smooth muscle
ulation of goblet cells through the release of neu- cells isolated from COPD patients, which might
trophil elastase and cathepsin G [181]. contribute to inflammation and increased airway
smooth muscle mass in COPD [192].

Airway Remodeling
Unanswered Question
Airway wall remodeling is present in the small
airways of patients with COPD, consisting of tis- Although major progress has been made in
sue repair and epithelial metaplasia that contrib- understanding COPD, many questions about
ute to airway wall thickening and airflow COPD remain unanswered.
obstruction. 1. Why do not all smokers develop COPD?
TGF-β is proposed as an important mediator 2. Why does it take decades to develop COPD?
of airway remodeling in COPD [182]. Increased 3. How the underlying inflammatory process is
TGF-β is found in the small airway epithelium of linked to pathophysiology and disease
COPD patients, and its levels correlate with the progression?
severity of obstruction [183, 184]. Adenoviral 4. Why inflammation and disease progression

transfer of TGF-β1 to murine lung shows persist even after smoking cessation?
increased fibroblast accumulation around air- 5. Why is COPD heterogeneous?
ways, and airway epithelial transgenic expression The issues will provide a valuable basis for
of TGF-β causes peribronchiolar fibrosis [185, future research into this devastating disease.
186]. TGF-β also increases the expression of
smooth muscle contractile proteins, such as
α-smooth muscle actin and calponin, in airway References
smooth muscle and fibroblasts, which increases
airway smooth muscle mass [187, 188]. 1. Brusselle GG, Joos GF, Bracke KR. New insights
TGF-β is ubiquitously expressed in three iso- into the immunology of chronic obstructive pulmo-
forms in all mammalian cell types and tissues but nary disease. Lancet. 2011;378(9795):1015–26.
2. Barnes PJ. Cellular and molecular mechanisms of
is found almost entirely in a latent form [189, chronic obstructive pulmonary disease. Clin Chest
190]. Thus, a major point of regulation of TGF-β Med. 2014;35(1):71–86.
function is through its activation. 3. Saetta M, Di Stefano A, Turato G, Facchini FM,
Squamous metaplasia actively contributes to Corbino L, Mapp CE, et al. CD8+ T-lymphocytes
in peripheral airways of smokers with chronic
airway wall thickening through alteration of obstructive pulmonary disease. Am J Respir Crit
mesenchymal interactions in COPD. Care Med. 1998;157(3 Pt 1):822–6. doi:10.1164/
IL-1β released by squamous metaplastic epithe- ajrccm.157.3.9709027.
lium activates TGF-β in human airway fibro- 4. Urbanowicz RA, Lamb JR, Todd I, Corne
JM, Fairclough LC. Enhanced effector func-
blasts [15]. tion of cytotoxic cells in the induced sputum
Smooth muscle mass seen in small airways of COPD patients. Respir Res. 2010;11:76.
of COPD patients is inversely correlated with doi:10.1186/1465-9921-11-76.
the lung function [3]. However, it is unknown 5. Sullivan AK, Simonian PL, Falta MT, Mitchell JD,
Cosgrove GP, Brown KK, et al. Oligoclonal CD4+ T
whether increased smooth muscle mass is cells in the lungs of patients with severe emphysema.
caused by an increased cell number, an Am J Respir Crit Care Med. 2005;172(5):590–6.
increased cell size, or both [191]. doi:10.1164/rccm.200410-1332OC.
4  Pathogenesis of COPD 47

6. van der Strate BW, Postma DS, Brandsma CA, way epithelial barrier function and cell-cell con-
Melgert BN, Luinge MA, Geerlings M, et al. tact recovery. Eur Respir J. 2012;39(2):419–28.
Cigarette smoke-induced emphysema: a role for the B doi:10.1183/09031936.00193810.
cell? Am J Respir Crit Care Med. 2006;173(7):751– 19. Russell RE, Culpitt SV, DeMatos C, Donnelly L,
8. doi:10.1164/rccm.200504-594OC. Smith M, Wiggins J, et al. Release and activity of
7. Hogg JC, Chu F, Utokaparch S, Woods R, Elliott matrix metalloproteinase-9 and tissue inhibitor of
WM, Buzatu L, et al. The nature of small-airway metalloproteinase-1 by alveolar macrophages from
obstruction in chronic obstructive pulmonary patients with chronic obstructive pulmonary dis-
disease. N Engl J Med. 2004;350(26):2645–53. ease. Am J Respir Cell Mol Biol. 2002;26(5):602–9.
doi:10.1056/NEJMoa032158. doi:10.1165/ajrcmb.26.5.4685.
8. Tetley TD. Inflammatory cells and chronic obstruc- 20. Meshi B, Vitalis TZ, Ionescu D, Elliott WM, Liu
tive pulmonary disease. Curr Drug Targets Inflamm C, Wang XD, et al. Emphysematous lung destruc-
Allergy. 2005;4(6):607–18. tion by cigarette smoke. The effects of latent adeno-
9. Gadgil A, Zhu X, Sciurba FC, Duncan SR. Altered viral infection on the lung inflammatory response.
T-cell phenotypes in chronic obstructive pulmonary Am J Respir Cell Mol Biol. 2002;26(1):52–7.
disease. Proc Am Thorac Soc. 2006;3(6):487–8. doi:10.1165/ajrcmb.26.1.4253.
doi:10.1513/pats.200603-064MS. 21. Capelli A, Di Stefano A, Gnemmi I, Balbo P, Cerutti
10. Domagala-Kulawik J, Maskey-Warzechowska M, CG, Balbi B, et al. Increased MCP-1 and MIP-1beta
Kraszewska I, Chazan R. The cellular composi- in bronchoalveolar lavage fluid of chronic bronchit-
tion and macrophage phenotype in induced sputum ics. Eur Respir J. 1999;14(1):160–5.
in smokers and ex-smokers with COPD. Chest. 22. Pappas K, Papaioannou AI, Kostikas K, Tzanakis
2003;123(4):1054–9. N. The role of macrophages in obstructive air-
11. Shaykhiev R, Crystal RG. Innate immunity ways disease: chronic obstructive pulmonary dis-
and chronic obstructive pulmonary disease: a ease and asthma. Cytokine. 2013;64(3):613–25.
mini-review. Gerontology. 2013;59(6):481–9. doi:10.1016/j.cyto.2013.09.010.
doi:10.1159/000354173. 23. Grumelli S, Corry DB, Song LZ, Song L, Green L,
12. Shaykhiev R, Otaki F, Bonsu P, Dang DT, Teater M, Huh J, et al. An immune basis for lung parenchy-
Strulovici-Barel Y, et al. Cigarette smoking repro- mal destruction in chronic obstructive pulmonary
grams apical junctional complex molecular architec- disease and emphysema. PLoS Med. 2004;1(1):e8.
ture in the human airway epithelium in vivo. Cell doi:10.1371/journal.pmed.0010008.
Mol Life Sci. 2011;68(5):877–92. doi:10.1007/ 24. Lim S, Roche N, Oliver BG, Mattos W, Barnes
s00018-010-0500-x. PJ, Chung KF. Balance of matrix metalloprote-
13. Thorley AJ, Tetley TD. Pulmonary epithelium, ase-­9 and tissue inhibitor of metalloprotease-1
cigarette smoke, and chronic obstructive pulmo- from alveolar macrophages in cigarette smokers.
nary disease. Int J Chron Obstruct Pulmon Dis. Regulation by interleukin-10. Am J Respir Crit
2007;2(4):409–28. Care Med. 2000;162(4 Pt 1):1355–60. doi:10.1164/
14. Auerbach O, Forman JB, Gere JB, Kassouny ajrccm.162.4.9910097.
DY, Muehsam GE, Petrick TG, et al. Changes 25. Barnes PJ. Role of HDAC2 in the pathophysiol-
in the bronchial epithelium in relation to smok- ogy of COPD. Annu Rev Physiol. 2009;71:451–64.
ing and cancer of the lung; a report of progress. N doi:10.1146/annurev.physiol.010908.163257.
Engl J Med. 1957;256(3):97–104. doi:10.1056/ 26. Ito K, Ito M, Elliott WM, Cosio B, Caramori G,
nejm195701172560301. Kon OM, et al. Decreased histone deacetylase activ-
15. Araya J, Cambier S, Markovics JA, Wolters P, ity in chronic obstructive pulmonary disease. N
Jablons D, Hill A, et al. Squamous metaplasia ampli- Engl J Med. 2005;352(19):1967–76. doi:10.1056/
fies pathologic epithelial-mesenchymal interactions NEJMoa041892.
in COPD patients. J Clin Invest. 2007;117(11):3551– 27. Ito K, Yamamura S, Essilfie-Quaye S, Cosio B, Ito
62. doi:10.1172/jci32526. M, Barnes PJ, et al. Histone deacetylase 2-mediated
16. Aarbiou J, van Schadewijk A, Stolk J, Sont JK, de deacetylation of the glucocorticoid receptor enables
Boer WI, Rabe KF, et al. Human neutrophil defen- NF-kappaB suppression. J Exp Med. 2006;203(1):7–
sins and secretory leukocyte proteinase inhibitor in 13. doi:10.1084/jem.20050466.
squamous metaplastic epithelium of bronchial air- 28. Hodge S, Hodge G, Scicchitano R, Reynolds
ways. Inflamm Res. 2004;53(6):230–8. doi:10.1007/ PN, Holmes M. Alveolar macrophages from
s00011-003-1240-x. subjects with chronic obstructive pulmonary
17. Kennedy SM, Elwood RK, Wiggs BJ, Pare PD, disease are ­deficient in their ability to phagocy-
Hogg JC. Increased airway mucosal permeability of tose apoptotic airway epithelial cells. Immunol
smokers. Relationship to airway reactivity. Am Rev Cell Biol. 2003;81(4):289–96. doi:10.1046/j.
Respir Dis. 1984;129(1):143–8. 1440-1711.2003.t01-1-01170.x.
18. Heijink IH, Brandenburg SM, Postma DS, van 29. Taylor AE, Finney-Hayward TK, Quint JK,
Oosterhout AJ. Cigarette smoke impairs air- Thomas CM, Tudhope SJ, Wedzicha JA, et al.
48 J.-H. Lee

Defective macrophage phagocytosis of bacte- expression of maturation molecules increases with

ria in COPD. Eur Respir J. 2010;35(5):1039–47. worsening chronic obstructive pulmonary disease.
doi:10.1183/09031936.00036709. Am J Respir Crit Care Med. 2009;180(12):1179–88.
30. Mullerova H, Chigbo C, Hagan GW, Woodhead doi:10.1164/rccm.200904-0552OC.
MA, Miravitlles M, Davis KJ, et al. The natu- 42. Majo J, Ghezzo H, Cosio MG. Lymphocyte
ral history of community-acquired pneumonia in population and apoptosis in the lungs of smok-
COPD patients: a population database analysis. ers and their relation to emphysema. Eur Respir
Respir Med. 2012;106(8):1124–33. doi:10.1016/j. J. 2001;17(5):946–53.
rmed.2012.04.008. 43. Di Stefano A, Caramori G, Gnemmi I, Contoli M,
31. Keatings VM, Collins PD, Scott DM, Barnes Vicari C, Capelli A, et al. T helper type 17-related
PJ. Differences in interleukin-8 and tumor necrosis cytokine expression is increased in the bronchial
factor-alpha in induced sputum from patients with mucosa of stable chronic obstructive pulmonary dis-
chronic obstructive pulmonary disease or asthma. ease patients. Clin Exp Immunol. 2009;157(2):316–
Am J Respir Crit Care Med. 1996;153(2):530–4. 24. doi:10.1111/j.1365-2249.2009.03965.x.
doi:10.1164/ajrccm.153.2.8564092. 44. Imaoka H, Hoshino T, Takei S, Kinoshita T,
32. Sparrow D, Glynn RJ, Cohen M, Weiss ST. The rela- Okamoto M, Kawayama T, et al. Interleukin-18
tionship of the peripheral leukocyte count and ciga- production and pulmonary function in
rette smoking to pulmonary function among adult COPD. Eur Respir J. 2008;31(2):287–97.
men. Chest. 1984;86(3):383–6. doi:10.1183/09031936.00019207.
33. Stanescu D, Sanna A, Veriter C, Kostianev S, 45. Miossec P, Korn T, Kuchroo VK. Interleukin-17
Calcagni PG, Fabbri LM, et al. Airways obstruction, and type 17 helper T cells. N Engl J Med.
chronic expectoration, and rapid decline of FEV1 in 2009;361(9):888–98. doi:10.1056/NEJMra0707449.
smokers are associated with increased levels of spu- 46. Pridgeon C, Bugeon L, Donnelly L, Straschil U,
tum neutrophils. Thorax. 1996;51(3):267–71. Tudhope SJ, Fenwick P, et al. Regulation of IL-17 in
34. Terashima T, Wiggs B, English D, Hogg JC, van Eeden chronic inflammation in the human lung. Clin
SF. Phagocytosis of small carbon particles (PM10) Sci (Lond). 2011;120(12):515–24. doi:10.1042/
by alveolar macrophages stimulates the release of cs20100417.
polymorphonuclear leukocytes from bone marrow. 47. Tang Q, Bluestone JA. The Foxp3+ regulatory T
Am J Respir Crit Care Med. 1997;155(4):1441–7. cell: a jack of all trades, master of regulation. Nat
doi:10.1164/ajrccm.155.4.9105091. Immunol. 2008;9(3):239–44. doi:10.1038/ni1572.
35. MacNee W, Wiggs B, Belzberg AS, Hogg JC. The 48. Lee SH, Goswami S, Grudo A, Song LZ, Bandi V,
effect of cigarette smoking on neutrophil kinetics in Goodnight-White S, et al. Antielastin autoimmunity
human lungs. N Engl J Med. 1989;321(14):924–8. in tobacco smoking-induced emphysema. Nat Med.
doi:10.1056/nejm198910053211402. 2007;13(5):567–9. doi:10.1038/nm1583.
36. Traves SL, Culpitt SV, Russell RE, Barnes PJ, 49. Barcelo B, Pons J, Ferrer JM, Sauleda J, Fuster
Donnelly LE. Increased levels of the chemokines A, Agusti AG. Phenotypic characterisation of
GROalpha and MCP-1 in sputum samples from T-lymphocytes in COPD: abnormal CD4+CD25+
patients with COPD. Thorax. 2002;57(7):590–5. regulatory T-lymphocyte response to tobacco
37. Sapey E, Stockley JA, Greenwood H, Ahmad A, smoking. Eur Respir J. 2008;31(3):555–62.
Bayley D, Lord JM, et al. Behavioral and structural doi:10.1183/09031936.00010407.
differences in migrating peripheral neutrophils from 50. Andersson CK, Mori M, Bjermer L, Lofdahl CG,
patients with chronic obstructive pulmonary disease. Erjefalt JS. Alterations in lung mast cell populations
Am J Respir Crit Care Med. 2011;183(9):1176–86. in patients with chronic obstructive pulmonary dis-
doi:10.1164/rccm.201008-1285OC. ease. Am J Respir Crit Care Med. 2010;181(3):206–
38. Stockley JA, Walton GM, Lord JM, Sapey 17. doi:10.1164/rccm.200906-0932OC.
E. Aberrant neutrophil functions in stable chronic 51. Turato G, Zuin R, Saetta M. Pathogenesis and
obstructive pulmonary disease: the neutrophil as an pathology of COPD. Respiration. 2001;68(2):117–
immunotherapeutic target. Int Immunopharmacol. 28. doi:10.1159/000050478.
2013;17(4):1211–7. doi:10.1016/j.intimp.2013.05.035. 52. Brightling CE, Monteiro W, Ward R, Parker D,
39. Soler P, Moreau A, Basset F, Hance AJ. Cigarette Morgan MD, Wardlaw AJ, et al. Sputum eosino-
smoking-induced changes in the number and differen- philia and short-term response to prednisolone in
tiated state of pulmonary dendritic cells/Langerhans chronic obstructive pulmonary disease: a randomised
cells. Am Rev Respir Dis. 1989;139(5):1112–7. controlled trial. Lancet. 2000;356(9240):1480–5.
doi:10.1164/ajrccm/139.5.1112. doi:10.1016/s0140-6736(00)02872-5.
40. Van Pottelberge GR, Bracke KR, Demedts IK, 53. Saetta M, Di Stefano A, Maestrelli P, Turato G,
De Rijck K, Reinartz SM, van Drunen CM, et al. Ruggieri MP, Roggeri A, et al. Airway eosinophilia
Selective accumulation of langerhans-type dendritic in chronic bronchitis during exacerbations. Am
cells in small airways of patients with COPD. Respir J Respir Crit Care Med. 1994;150(6 Pt 1):1646–52.
Res. 2010;11:35. doi:10.1186/1465-9921-11-35. doi:10.1164/ajrccm.150.6.7952628.
41. Freeman CM, Martinez FJ, Han MK, Ames TM, 54. Zhu J, Qiu YS, Majumdar S, Gamble E, Matin D,
Chensue SW, Todt JC, et al. Lung dendritic cell Turato G, et al. Exacerbations of bronchitis: bronchial
4  Pathogenesis of COPD 49

eosinophilia and gene expression for interleukin-4, 69. Stockley RA. Alpha1-antitrypsin review. Clin
interleukin-5, and eosinophil chemoattractants. Chest Med. 2014;35(1):39–50. doi:10.1016/j.
Am J Respir Crit Care Med. 2001;164(1):109–16. ccm.2013.10.001.
doi:10.1164/ajrccm.164.1.2007050. 70. Dickens JA, Lomas DA. Why has it been so difficult
55. Papi A, Romagnoli M, Baraldo S, Braccioni F, to prove the efficacy of alpha-1-antitrypsin replace-
Guzzinati I, Saetta M, et al. Partial reversibility of air- ment therapy? Insights from the study of disease
flow limitation and increased exhaled NO and sputum pathogenesis. Drug Des Devel Ther. 2011;5:391–
eosinophilia in chronic obstructive pulmonary disease. 405. doi:10.2147/dddt.s14018.
Am J Respir Crit Care Med. 2000;162(5):1773–7. 71. MacNee W. Pathogenesis of chronic obstruc-
doi:10.1164/ajrccm.162.5.9910112. tive pulmonary disease. Proc Am Thorac Soc.
56. Vivier E, Raulet DH, Moretta A, Caligiuri MA, 2005;2(4):258–66. doi:10.1513/pats.200504-­045SR;
Zitvogel L, Lanier LL, et al. Innate or adaptive discussion 90–1.
immunity? The example of natural killer cells. 72. Kidokoro Y, Kravis TC, Moser KM, Taylor JC,
Science (New York, NY). 2011;331(6013):44–9. Crawford IP. Relationship of leukocyte elastase con-
doi:10.1126/science.1198687. centration to severity of emphysema in homozygous
57. Cosio MG, Saetta M, Agusti A. Immunologic alpha1-antitrypsin-deficient persons. Am Rev Respir
aspects of chronic obstructive pulmonary disease. Dis. 1977;115(5):793–803.
N Engl J Med. 2009;360(23):2445–54. doi:10.1056/ 73. O’Reilly P, Jackson PL, Noerager B, Parker S,
NEJMra0804752. Dransfield M, Gaggar A, et al. N-alpha-PGP
58. Degli-Esposti MA, Smyth MJ. Close encounters of and PGP, potential biomarkers and therapeu-
different kinds: dendritic cells and NK cells take tic targets for COPD. Respir Res. 2009;10:38.
centre stage. Nat Rev Immunol. 2005;5(2):112–24. doi:10.1186/1465-9921-10-38.
doi:10.1038/nri1549. 74. Rodriguez-Cintron W, Guntupalli K, Fraire
59. Barnes PJ. The cytokine network in chronic obstruc- AE. Bronchiectasis and homozygous (P1ZZ) alpha
tive pulmonary disease. Am J Respir Cell Mol Biol. 1-antitrypsin deficiency in a young man. Thorax.
2009;41(6):631–8. doi:10.1165/rcmb.2009-0220TR. 1995;50(4):424–5.
60. Bagdonas E, Raudoniute J, Bruzauskaite I, Aldonyte 75. Parr DG, Stoel BC, Stolk J, Stockley RA. Pattern
R. Novel aspects of pathogenesis and regeneration of emphysema distribution in alpha1-antitrypsin
mechanisms in COPD. Int J Chron Obstruct Pulmon deficiency influences lung function impairment.
Dis. 2015;10:995–1013. doi:10.2147/copd.s82518. Am J Respir Crit Care Med. 2004;170(11):1172–8.
61. Chung KF, Adcock IM. Multifaceted mechanisms doi:10.1164/rccm.200406-761OC.
in COPD: inflammation, immunity, and tissue repair 76. Gadek JE, Fells GA, Crystal RG. Cigarette smok-
and destruction. Eur Respir J. 2008;31(6):1334–56. ing induces functional antiprotease deficiency in the
doi:10.1183/09031936.00018908. lower respiratory tract of humans. Science (New
62. Caramori G, Di Stefano A, Casolari P, Kirkham PA, York, NY). 1979;206(4424):1315–6.
Padovani A, Chung KF, et al. Chemokines and che- 77. Carp H, Miller F, Hoidal JR, Janoff A. Potential
mokine receptors blockers as new drugs for the treat- mechanism of emphysema: alpha 1-proteinase
ment of chronic obstructive pulmonary disease. Curr inhibitor recovered from lungs of cigarette smok-
Med Chem. 2013;20(35):4317–49. ers contains oxidized methionine and has decreased
63. Barnes PJ. Mediators of chronic obstructive pulmo- elastase inhibitory capacity. Proc Natl Acad Sci U S
nary disease. Pharmacol Rev. 2004;56(4):515–48. A. 1982;79(6):2041–5.
doi:10.1124/pr.56.4.2. 78. Stone PJ, Calore JD, McGowan SE, Bernardo J,
64. Caramori G, Casolari P, Adcock I. Role of tran- Snider GL, Franzblau C. Functional alpha 1-­protease
scription factors in the pathogenesis of asthma and inhibitor in the lower respiratory tract of cigarette
COPD. Cell Commun Adhes. 2013;20(1–2):21–40. smokers is not decreased. Science (New York, NY).
doi:10.3109/15419061.2013.775257. 1983;221(4616):1187–9.
65. Laurell CB, Eriksson S. The electrophoretic alpha1-­ 79. Boudier C, Pelletier A, Pauli G, Bieth JG. The
globulin pattern of serum in alpha1-antitrypsin defi- functional activity of alpha 1-proteinase inhibi-
ciency. 1963. COPD. 2013;10(Suppl 1):3–8. doi:10. tor in bronchoalveolar lavage fluids from healthy
3109/15412555.2013.771956. human smokers and non-smokers. Clin Chim Acta.
66. Janoff A, Sloan B, Weinbaum G, Damiano V, 1983;132(3):309–15.
Sandhaus RA, Elias J, et al. Experimental emphy- 80. Abboud RT, Fera T, Richter A, Tabona MZ, Johal
sema induced with purified human neutrophil elas- S. Acute effect of smoking on the functional activ-
tase: tissue localization of the instilled protease. Am ity of alpha1-protease inhibitor in bronchoalveolar
Rev Respir Dis. 1977;115(3):461–78. lavage fluid. Am Rev Respir Dis. 1985;131(1):79–85.
67. Senior RM, Tegner H, Kuhn C, Ohlsson K, Starcher 81. Niewoehner DE, Kleinerman J, Rice DB. Pathologic
BC, Pierce JA. The induction of pulmonary emphy- changes in the peripheral airways of young ciga-
sema with human leukocyte elastase. Am Rev Respir rette smokers. N Engl J Med. 1974;291(15):755–8.
Dis. 1977;116(3):469–75. doi:10.1056/nejm197410102911503.
68. Churg A, Wright JL. Proteases and emphysema. 82. Campbell EJ, White RR, Senior RM, Rodriguez
Curr Opin Pulm Med. 2005;11(2):153–9. RJ, Kuhn C. Receptor-mediated binding and inter-
50 J.-H. Lee

nalization of leukocyte elastase by alveolar macro- macrophage precursors to diseased sites in pul-
phages in vitro. J Clin Invest. 1979;64(3):824–33. monary emphysema. Science (New York, NY).
doi:10.1172/jci109530. 1981;212(4497):925–7.
83. Betsuyaku T, Yoshioka A, Nishimura M, Miyamoto 96. Shapiro SD, Kobayashi DK, Welgus HG.
K, Kondo T, Kawakami Y. Neutrophil elastase asso- Identification of TIMP-2 in human alveolar mac-
ciated with alveolar macrophages from older volun- rophages. Regulation of biosynthesis is opposite
teers. Am J Respir Crit Care Med. 1995;151(2 Pt to that of metalloproteinases and TIMP-1. J Biol
1):436–42. doi:10.1164/ajrccm.151.2.7842203. Chem. 1992;267(20):13890–4.
84. Reilly JJ Jr, Mason RW, Chen P, Joseph LJ, Sukhatme 97. Pons AR, Sauleda J, Noguera A, Pons J, Barcelo B,
VP, Yee R, et al. Synthesis and processing of cathep- Fuster A, et al. Decreased macrophage release of
sin L, an elastase, by human alveolar macrophages. TGF-beta and TIMP-1 in chronic obstructive pul-
Biochem J. 1989;257(2):493–8. monary disease. Eur Respir J. 2005;26(1):60–6. doi:
85. Reilly JJ Jr, Chen P, Sailor LZ, Wilcox D, Mason 10.1183/09031936.05.00045504.
RW, Chapman HA Jr. Cigarette smoking induces an 98. Leco KJ, Waterhouse P, Sanchez OH, Gowing KL,
elastolytic cysteine proteinase in macrophages dis- Poole AR, Wakeham A, et al. Spontaneous air space
tinct from cathepsin L. Am J Physiol. 1991;261(2 Pt enlargement in the lungs of mice lacking tissue
1):L41–8. inhibitor of metalloproteinases-3 (TIMP-3). J Clin
86. Shi GP, Munger JS, Meara JP, Rich DH, Chapman Invest. 2001;108(6):817–29. doi:10.1172/jci12067.
HA. Molecular cloning and expression of human 99. Sies H. Oxidative stress: oxidants and antioxidants.
alveolar macrophage cathepsin S, an elastinolytic cys- Exp Physiol. 1997;82(2):291–5.
teine protease. J Biol Chem. 1992;267(11):7258–62. 100. Domej W, Oettl K, Renner W. Oxidative stress and
87. Shapiro SD, Kobayashi DK, Ley TJ. Cloning and free radicals in COPD—implications and relevance
characterization of a unique elastolytic metallopro- for treatment. Int J Chron Obstruct Pulmon Dis.
teinase produced by human alveolar macrophages. 2014;9:1207–24. doi:10.2147/copd.s51226.
J Biol Chem. 1993;268(32):23824–9. 101. MacNee W. Oxidants and COPD. Curr Drug Targets
88. Foronjy RF, Okada Y, Cole R, D’Armiento Inflamm Allergy. 2005;4(6):627–41.
J. Progressive adult-onset emphysema in transgenic 102. Mak JC. Pathogenesis of COPD. Part II. Oxidative-­
mice expressing human MMP-1 in the lung. Am antioxidative imbalance. Int J Tuberc Lung Dis.
J Physiol Lung Cell Mol Physiol. 2003;284(5):L727– 2008;12(4):368–74.
37. doi:10.1152/ajplung.00349.2002. 103. Eiserich JP, van der Vliet A, Handelman GJ,
89. D’Armiento J, Dalal SS, Okada Y, Berg RA, Chada Halliwell B, Cross CE. Dietary antioxidants and
K. Collagenase expression in the lungs of trans- cigarette smoke-induced biomolecular damage:
genic mice causes pulmonary emphysema. Cell. a complex interaction. Am J Clin Nutr. 1995;62(6
1992;71(6):955–61. Suppl):1490s–500s.
90. Shiomi T, Okada Y, Foronjy R, Schiltz J, Jaenish 104. Cosio BG, Tsaprouni L, Ito K, Jazrawi E, Adcock
R, Krane S, et al. Emphysematous changes are IM, Barnes PJ. Theophylline restores histone
caused by degradation of type III collagen in trans- deacetylase activity and steroid responses in COPD
genic mice expressing MMP-1. Exp Lung Res. macrophages. J Exp Med. 2004;200(5):689–95.
2003;29(1):1–15. doi:10.1084/jem.20040416.
91. Muley T, Wiebel M, Schulz V, Ebert W. Elastinolytic 105. Cavarra E, Lucattelli M, Gambelli F, Bartalesi B,
activity of alveolar macrophages in smoking-­ Fineschi S, Szarka A, et al. Human SLPI inactiva-
associated pulmonary emphysema. Clin Investig. tion after cigarette smoke exposure in a new in vivo
1994;72(4):269–76. model of pulmonary oxidative stress. Am J Physiol
92. Ohnishi K, Takagi M, Kurokawa Y, Satomi Lung Cell Mol Physiol. 2001;281(2):L412–7.
S, Konttinen YT. Matrix metalloproteinase-­ 106. Shapiro SD. Proteolysis in the lung. Eur Respir
mediated extracellular matrix protein degrada- J Suppl. 2003;44:30s–2s.
tion in human pulmonary emphysema. Lab Invest. 107. Sussan TE, Rangasamy T, Blake DJ, Malhotra D,
1998;78(9):1077–87. El-Haddad H, Bedja D, et al. Targeting Nrf2 with
93. Imai K, Hiramatsu A, Fukushima D, Pierschbacher the triterpenoid CDDO-imidazolide attenuates
MD, Okada Y. Degradation of decorin by matrix cigarette smoke-induced emphysema and cardiac
metalloproteinases: identification of the cleavage dysfunction in mice. Proc Natl Acad Sci U S A.
sites, kinetic analyses and transforming growth 2009;106(1):250–5. doi:10.1073/pnas.0804333106.
factor-beta1 release. Biochem J. 1997;322(Pt 108. Yokohori N, Aoshiba K, Nagai A. Increased levels
3):809–14. of cell death and proliferation in alveolar wall cells
94. Hautamaki RD, Kobayashi DK, Senior RM, Shapiro in patients with pulmonary emphysema. Chest.
SD. Requirement for macrophage elastase for ciga- 2004;125(2):626–32.
rette smoke-induced emphysema in mice. Science 109. Segura-Valdez L, Pardo A, Gaxiola M, Uhal BD,
(New York, NY). 1997;277(5334):2002–4. Becerril C, Selman M. Upregulation of gelatinases A
95. Hunninghake GW, Davidson JM, Rennard S, Szapiel and B, collagenases 1 and 2, and increased parenchy-
S, Gadek JE, Crystal RG. Elastin fragments attract mal cell death in COPD. Chest. 2000;117(3):684–94.
4  Pathogenesis of COPD 51

110. Tuder RM. Aging and cigarette smoke: fueling the thelial cells. Cardiovasc Res. 2011;92(1):141–8.
fire. Am J Respir Crit Care Med. 2006;174(5):490– doi:10.1093/cvr/cvr165.
1. doi:10.1164/rccm.200607-924ED. 122. Hwang JW, Chung S, Sundar IK, Yao H,
111. Vlahos R, Bozinovski S. Glutathione peroxidase-1 Arunachalam G, McBurney MW, et al. Cigarette
as a novel therapeutic target for COPD. Redox smoke-induced autophagy is regulated by SIRT1-­
Rep. 2013;18(4):142–9. doi:10.1179/13510002 PARP-­ 1-dependent mechanism: implication in
13y.0000000053. pathogenesis of COPD. Arch Biochem Biophys.
112. Rangasamy T, Cho CY, Thimmulappa RK, Zhen 2010;500(2):203–9. doi:10.1016/
L, Srisuma SS, Kensler TW, et al. Genetic abla- 123. Monick MM, Powers LS, Walters K, Lovan N,
tion of Nrf2 enhances susceptibility to cigarette Zhang M, Gerke A, et al. Identification of an
smoke-induced emphysema in mice. J Clin Invest. autophagy defect in smokers’ alveolar macrophages.
2004;114(9):1248–59. doi:10.1172/jci21146. J Immunol. 2010;185(9):5425–35. doi:10.4049/
113. Aoshiba K, Yokohori N, Nagai A. Alveolar jimmunol.1001603.
wall apoptosis causes lung destruction and 124. Kim HP, Wang X, Chen ZH, Lee SJ, Huang MH,
emphysematous changes. Am J Respir Cell Wang Y, et al. Autophagic proteins regulate cigarette
Mol Biol. 2003;28(5):555–62. doi:10.1165/ smoke-induced apoptosis: protective role of heme
rcmb.2002-0090OC. oxygenase-1. Autophagy. 2008;4(7):887–95.
114. Petrache I, Natarajan V, Zhen L, Medler TR, Richter AT, 125. Kumar PA, Hu Y, Yamamoto Y, Hoe NB, Wei TS,
Cho C, et al. Ceramide upregulation causes pulmonary Mu D, et al. Distal airway stem cells yield alveoli
cell apoptosis and emphysema-like disease in mice. Nat in vitro and during lung regeneration following
Med. 2005;11(5):491–8. doi:10.1038/nm1238. H1N1 influenza infection. Cell. 2011;147(3):525–
115. Petrache I, Medler TR, Richter AT, Kamocki K, 38. doi:10.1016/j.cell.2011.10.001.
Chukwueke U, Zhen L, et al. Superoxide dismutase 126. Kauffman SL. Cell proliferation in the mammalian
protects against apoptosis and alveolar enlarge- lung. Int Rev Exp Pathol. 1980;22:131–91.
ment induced by ceramide. Am J Physiol Lung Cell 127. Butler JP, Loring SH, Patz S, Tsuda A, Yablonskiy
Mol Physiol. 2008;295(1):L44–53. doi:10.1152/ DA, Mentzer SJ. Evidence for adult lung growth
ajplung.00448.2007. in humans. N Engl J Med. 2012;367(3):244–7.
116. Tuder RM, Yoshida T, Fijalkowka I, Biswal S, doi:10.1056/NEJMoa1203983.
Petrache I. Role of lung maintenance program in 128. Wansleeben C, Barkauskas CE, Rock JR, Hogan
the heterogeneity of lung destruction in emphysema. BL. Stem cells of the adult lung: their development
Proc Am Thorac Soc. 2006;3(8):673–9. doi:10.1513/ and role in homeostasis, regeneration, and disease.
pats.200605-124SF. Wiley Interdiscip Rev Dev Biol. 2013;2(1):131–48.
117. Tang K, Rossiter HB, Wagner PD, Breen EC. Lung-­ doi:10.1002/wdev.58.
targeted VEGF inactivation leads to an emphysema 129. Takahashi Y, Izumi Y, Kohno M, Kimura T,
phenotype in mice. J Appl Physiol. 2004;97(4):1559– Kawamura M, Okada Y, et al. Thyroid transcrip-
66. doi:10.1152/japplphysiol.00221.2004; discus- tion factor-1 influences the early phase of compen-
sion 49. satory lung growth in adult mice. Am J Respir Crit
118. Kasahara Y, Tuder RM, Cool CD, Lynch DA, Care Med. 2010;181(12):1397–406. doi:10.1164/
Flores SC, Voelkel NF. Endothelial cell death rccm.200908-1265OC.
and decreased expression of vascular endothelial 130. Sakurai MK, Lee S, Arsenault DA, Nose V, Wilson
growth factor and vascular endothelial growth fac- JM, Heymach JV, et al. Vascular endothelial growth
tor receptor 2 in emphysema. Am J Respir Crit factor accelerates compensatory lung growth after
Care Med. 2001;163(3 Pt 1):737–44. doi:10.1164/ unilateral pneumonectomy. Am J Physiol Lung Cell
ajrccm.163.3.2002117. Mol Physiol. 2007;292(3):L742–7. doi:10.1152/
119. Marwick JA, Stevenson CS, Giddings J, MacNee W, ajplung.00064.2006.
Butler K, Rahman I, et al. Cigarette smoke disrupts 131. Zhang Q, Bellotto DJ, Ravikumar P, Moe OW,
VEGF165-VEGFR-2 receptor signaling complex Hogg RT, Hogg DC, et al. Postpneumonectomy
in rat lungs and patients with COPD: morphologi- lung expansion elicits hypoxia-inducible factor-­
cal impact of VEGFR-2 inhibition. Am J Physiol 1alpha signaling. Am J Physiol Lung Cell Mol
Lung Cell Mol Physiol. 2006;290(5):L897–908. Physiol. 2007;293(2):L497–504. doi:10.1152/
doi:10.1152/ajplung.00116.2005. ajplung.00393.2006.
120. Chen ZH, Kim HP, Sciurba FC, Lee SJ, Feghali-­ 132. Kaza AK, Kron IL, Leuwerke SM, Tribble CG,
Bostwick C, Stolz DB, et al. Egr-1 regulates autoph- Laubach VE. Keratinocyte growth factor enhances
agy in cigarette smoke-induced chronic obstructive post-pneumonectomy lung growth by alveolar
pulmonary disease. PLoS One. 2008;3(10):e3316. proliferation. Circulation. 2002;106(12 Suppl
doi:10.1371/journal.pone.0003316. 1):I120–4.
121. Csordas A, Kreutmayer S, Ploner C, Braun PR, 133. Barkauskas CE, Cronce MJ, Rackley CR, Bowie
Karlas A, Backovic A, et al. Cigarette smoke extract EJ, Keene DR, Stripp BR, et al. Type 2 alveo-
induces prolonged endoplasmic reticulum stress and lar cells are stem cells in adult lung. J Clin Invest.
autophagic cell death in human umbilical vein endo- 2013;123(7):3025–36. doi:10.1172/jci68782.
52 J.-H. Lee

134. Li Y, Gu C, Xu W, Yan J, Xia Y, Ma Y, et al. disease (COPD). Biochem Soc Trans. 2009;37(Pt
Therapeutic effects of amniotic fluid-derived mes- 4):819–23. doi:10.1042/bst0370819.
enchymal stromal cells on lung injury in rats with 148. Kuro-o M, Matsumura Y, Aizawa H, Kawaguchi H,
emphysema. Respir Res. 2014;15:120. doi:10.1186/ Suga T, Utsugi T, et al. Mutation of the mouse klotho
s12931-014-0120-3. gene leads to a syndrome resembling ageing. Nature.
135. Huh JW, Kim SY, Lee JH, Lee JS, Van Ta Q, Kim 1997;390(6655):45–51. doi:10.1038/36285.
M, et al. Bone marrow cells repair cigarette smoke-­ 149. Suga T, Kurabayashi M, Sando Y, Ohyama Y, Maeno
induced emphysema in rats. Am J Physiol Lung Cell T, Maeno Y, et al. Disruption of the klotho gene
Mol Physiol. 2011;301(3):L255–66. doi:10.1152/ causes pulmonary emphysema in mice. Defect in
ajplung.00253.2010. maintenance of pulmonary integrity during postnatal
136. Dunsmore SE, Rannels DE. Extracellular matrix life. Am J Respir Cell Mol Biol. 2000;22(1):26–33.
biology in the lung. Am J Physiol. 1996;270(1 Pt doi:10.1165/ajrcmb.22.1.3554.
1):L3–27. 150. Fujita T, Shirasawa T, Uchida K, Maruyama
137. Nakamura Y, Romberger DJ, Tate L, Ertl RF, N. Gene regulation of senescence marker protein-30
Kawamoto M, Adachi Y, et al. Cigarette smoke (SMP30): coordinated up-regulation with tissue
inhibits lung fibroblast proliferation and chemotaxis. maturation and gradual down-regulation with aging.
Am J Respir Crit Care Med. 1995;151(5):1497–503. Mech Ageing Dev. 1996;87(3):219–29.
doi:10.1164/ajrccm.151.5.7735606. 151. Maruyama N, Ishigami A, Kuramoto M, Handa S,
138. Carnevali S, Petruzzelli S, Longoni B, Vanacore Kubo S, Imasawa T, et al. Senescence marker pro-
R, Barale R, Cipollini M, et al. Cigarette smoke tein-­30 knockout mouse as an aging model. Ann
extract induces oxidative stress and apoptosis in N Y Acad Sci. 2004;1019:383–7. doi:10.1196/
human lung fibroblasts. Am J Physiol Lung Cell annals.1297.068.
Mol Physiol. 2003;284(6):L955–63. doi:10.1152/ 152. Fulop T, Larbi A, Kotb R, de Angelis F, Pawelec
ajplung.00466.2001. G. Aging, immunity, and cancer. Discov Med.
139. Nyunoya T, Monick MM, Klingelhutz A, Yarovinsky 2011;11(61):537–50.
TO, Cagley JR, Hunninghake GW. Cigarette 153. von Zglinicki T. Oxidative stress shortens telomeres.
smoke induces cellular senescence. Am J Respir Trends Biochem Sci. 2002;27(7):339–44.
Cell Mol Biol. 2006;35(6):681–8. doi:10.1165/ 154. Savale L, Chaouat A, Bastuji-Garin S, Marcos
rcmb.2006-0169OC. E, Boyer L, Maitre B, et al. Shortened telomeres
140. Carnevali S, Nakamura Y, Mio T, Liu X, Takigawa in circulating leukocytes of patients with chronic
K, Romberger DJ, et al. Cigarette smoke extract obstructive pulmonary disease. Am J Respir Crit
inhibits fibroblast-mediated collagen gel contrac- Care Med. 2009;179(7):566–71. ­ doi:10.1164/
tion. Am J Physiol. 1998;274(4 Pt 1):L591–8. rccm.200809-1398OC.
141. Holz O, Zuhlke I, Jaksztat E, Muller KC, Welker L, 155. Jeyapalan JC, Sedivy JM. Cellular senescence and
Nakashima M, et al. Lung fibroblasts from patients organismal aging. Mech Ageing Dev. 2008;129(7–
with emphysema show a reduced proliferation rate 8):467–74. doi:10.1016/j.mad.2008.04.001.
in culture. Eur Respir J. 2004;24(4):575–9. doi:10.1 156. Tsuji T, Aoshiba K, Nagai A. Cigarette smoke
183/09031936.04.00143703. induces senescence in alveolar epithelial cells.
142. Togo S, Holz O, Liu X, Sugiura H, Kamio K, Am J Respir Cell Mol Biol. 2004;31(6):643–9.
Wang X, et al. Lung fibroblast repair functions in doi:10.1165/rcmb.2003-0290OC.
patients with chronic obstructive pulmonary disease 157. Muller KC, Welker L, Paasch K, Feindt B,
are altered by multiple mechanisms. Am J Respir Erpenbeck VJ, Hohlfeld JM, et al. Lung fibro-
Crit Care Med. 2008;178(3):248–60. doi:10.1164/ blasts from patients with emphysema show mark-
rccm.200706-929OC. ers of senescence in vitro. Respir Res. 2006;7:32.
143. von Zglinicki T, Burkle A, Kirkwood TB. Stress, doi:10.1186/1465-9921-7-32.
DNA damage and ageing—an integrative approach. 158. Tsuji T, Aoshiba K, Nagai A. Alveolar cell senes-
Exp Gerontol. 2001;36(7):1049–62. cence in patients with pulmonary emphysema.
144. Lou Z, Chen J. Cellular senescence and DNA repair. Am J Respir Crit Care Med. 2006;174(8):886–93.
Exp Cell Res. 2006;312(14):2641–6. doi:10.1016/j. doi:10.1164/rccm.200509-1374OC.
yexcr.2006.06.009. 159. Yeung F, Hoberg JE, Ramsey CS, Keller MD, Jones
145. Fraga MF, Agrelo R, Esteller M. Cross-talk between DR, Frye RA, et al. Modulation of NF-kappaB-­
aging and cancer: the epigenetic language. Ann dependent transcription and cell survival by the
N Y Acad Sci. 2007;1100:60–74. doi:10.1196/ SIRT1 deacetylase. EMBO J. 2004;23(12):2369–80.
annals.1395.005. doi:10.1038/sj.emboj.7600244.
146. Friguet B. Oxidized protein degradation and
160. Nakamaru Y, Vuppusetty C, Wada H, Milne JC, Ito
repair in ageing and oxidative stress. FEBS M, Rossios C, et al. A protein deacetylase SIRT1 is
Lett. 2006;580(12):2910–6. doi:10.1016/j.febslet. a negative regulator of metalloproteinase-9. FASEB
2006.03.028. J. 2009;23(9):2810–9. doi:10.1096/fj.08-125468.
147. MacNee W. Accelerated lung aging: a novel patho- 161. Rajendrasozhan S, Yang SR, Kinnula VL, Rahman
genic mechanism of chronic obstructive pulmonary I. SIRT1, an antiinflammatory and antiaging protein,
4  Pathogenesis of COPD 53

is decreased in lungs of patients with chronic COPD. Antioxid Redox Signal. 2008;10(4):799–

obstructive pulmonary disease. Am J Respir Crit 811. doi:10.1089/ars.2007.1938.
Care Med. 2008;177(8):861–70. doi:10.1164/ 173. Cantin AM, Hanrahan JW, Bilodeau G, Ellis L,
rccm.200708-1269OC. Dupuis A, Liao J, et al. Cystic fibrosis transmem-
162. Kirkham PA, Caramori G, Casolari P, Papi AA, brane conductance regulator function is sup-
Edwards M, Shamji B, et al. Oxidative stress-induced pressed in cigarette smokers. Am J Respir Crit
antibodies to carbonyl-modified protein correlate Care Med. 2006;173(10):1139–44. doi:10.1164/
with severity of chronic obstructive pulmonary dis- rccm.200508-1330OC.
ease. Am J Respir Crit Care Med. 2011;184(7):796– 174. Caramori G, Di Gregorio C, Carlstedt I, Casolari P,
802. doi:10.1164/rccm.201010-1605OC. Guzzinati I, Adcock IM, et al. Mucin expression in
163. Nunez B, Sauleda J, Anto JM, Julia MR, Orozco M, peripheral airways of patients with chronic obstructive
Monso E, et al. Anti-tissue antibodies are related to pulmonary disease. Histopathology. 2004;45(5):477–
lung function in chronic obstructive pulmonary dis- 84. doi:10.1111/j.1365-2559.2004.01952.x.
ease. Am J Respir Crit Care Med. 2011;183(8):1025– 175. Kohri K, Ueki IF, Nadel JA. Neutrophil elastase
31. doi:10.1164/rccm.201001-0029OC. induces mucin production by ligand-dependent
164. Feghali-Bostwick CA, Gadgil AS, Otterbein epidermal growth factor receptor activation. Am
LE, Pilewski JM, Stoner MW, Csizmadia E, J Physiol Lung Cell Mol Physiol. 2002;283(3):L531–
et al. Autoantibodies in patients with chronic 40. doi:10.1152/ajplung.00455.2001.
obstructive pulmonary disease. Am J Respir Crit 176. Takeyama K, Jung B, Shim JJ, Burgel PR, Dao-Pick
Care Med. 2008;177(2):156–63. doi:10.1164/ T, Ueki IF, et al. Activation of epidermal growth
rccm.200701-014OC. factor receptors is responsible for mucin synthesis
165. Rinaldi M, Lehouck A, Heulens N, Lavend’homme induced by cigarette smoke. Am J Physiol Lung Cell
R, Carlier V, Saint-Remy JM, et al. Antielastin B-cell Mol Physiol. 2001;280(1):L165–72.
and T-cell immunity in patients with chronic obstruc- 177. Deshmukh HS, Case LM, Wesselkamper SC,
tive pulmonary disease. Thorax. 2012;67(8):694– Borchers MT, Martin LD, Shertzer HG, et al.
700. doi:10.1136/thoraxjnl-2011-200690. Metalloproteinases mediate mucin 5AC expres-
166. Taraseviciene-Stewart L, Scerbavicius R, Choe KH, sion by epidermal growth factor receptor activation.
Moore M, Sullivan A, Nicolls MR, et al. An animal Am J Respir Crit Care Med. 2005;171(4):305–14.
model of autoimmune emphysema. Am J Respir doi:10.1164/rccm.200408-1003OC.
Crit Care Med. 2005;171(7):734–42. doi:10.1164/ 178. Saetta M, Turato G, Baraldo S, Zanin A, Braccioni
rccm.200409-1275OC. F, Mapp CE, et al. Goblet cell hyperplasia and
167. Karayama M, Inui N, Suda T, Nakamura Y, Nakamura epithelial inflammation in peripheral airways of
H, Chida K. Antiendothelial cell antibodies in smokers with both symptoms of chronic bronchitis
patients with COPD. Chest. 2010;138(6):1303–8. and chronic airflow limitation. Am J Respir Crit
doi:10.1378/chest.10-0863. Care Med. 2000;161(3 Pt 1):1016–21. doi:10.1164/
168. Greene CM, Low TB, O’Neill SJ, McElvaney ajrccm.161.3.9907080.
NG. Anti-proline-glycine-proline or antielas- 179. Shim JJ, Dabbagh K, Ueki IF, Dao-Pick T, Burgel
tin autoantibodies are not evident in chronic PR, Takeyama K, et al. IL-13 induces mucin produc-
inflammatory lung disease. Am J Respir Crit tion by stimulating epidermal growth factor recep-
Care Med. 2010;181(1):31–5. doi:10.1164/ tors and by activating neutrophils. Am J Physiol
rccm.200904-0545OC. Lung Cell Mol Physiol. 2001;280(1):L134–40.
169. Wood AM, de Pablo P, Buckley CD, Ahmad A, 180. Takeyama K, Dabbagh K, Jeong Shim J, Dao-Pick
Stockley RA. Smoke exposure as a determinant T, Ueki IF, Nadel JA. Oxidative stress causes mucin
of autoantibody titre in alpha(1)-antitrypsin defi- synthesis via transactivation of epidermal growth
ciency and COPD. Eur Respir J. 2011;37(1):32–8. factor receptor: role of neutrophils. J Immunol.
doi:10.1183/09031936.00033710. 2000;164(3):1546–52.
170. Enelow RI, Mohammed AZ, Stoler MH, Liu AN, 181. Sommerhoff CP, Nadel JA, Basbaum CB, Caughey
Young JS, Lou YH, et al. Structural and func- GH. Neutrophil elastase and cathepsin G stimulate
tional consequences of alveolar cell recognition secretion from cultured bovine airway gland serous
by CD8(+) T lymphocytes in experimental lung cells. J Clin Invest. 1990;85(3):682–9. doi:10.1172/
disease. J Clin Invest. 1998;102(9):1653–61. jci114492.
doi:10.1172/jci4174. 182. Wang RD, Wright JL, Churg A. Transforming
171. Rajendrasozhan S, Yao H, Rahman I. Current per- growth factor-beta1 drives airway remodeling in cig-
spectives on role of chromatin modifications and arette smoke-exposed tracheal explants. Am J Respir
deacetylases in lung inflammation in COPD. COPD. Cell Mol Biol. 2005;33(4):387–93. doi:10.1165/
2009;6(4):291–7. rcmb.2005-0203OC.
172. Rajendrasozhan S, Yang SR, Edirisinghe I, Yao H, 183. Takizawa H, Tanaka M, Takami K, Ohtoshi T, Ito K,
Adenuga D, Rahman I. Deacetylases and NF-kappaB Satoh M, et al. Increased expression of transform-
in redox regulation of cigarette smoke-induced ing growth factor-beta1 in small airway epithelium
lung inflammation: epigenetics in pathogenesis of from tobacco smokers and patients with chronic
54 J.-H. Lee

o­ bstructive pulmonary disease (COPD). Am J Respir 188. Cohen P, Rajah R, Rosenbloom J, Herrick
Crit Care Med. 2001;163(6):1476–83. doi:10.1164/ DJ. IGFBP-3 mediates TGF-beta1-induced
ajrccm.163.6.9908135. cell growth in human airway smooth muscle
184. de Boer WI, van Schadewijk A, Sont JK, Sharma HS, cells. Am J Physiol Lung Cell Mol Physiol.
Stolk J, Hiemstra PS, et al. Transforming growth fac- 2000;278(3):L545–51.
tor beta1 and recruitment of macrophages and mast 189. Annes JP, Munger JS, Rifkin DB. Making sense of
cells in airways in chronic obstructive pulmonary dis- latent TGFbeta activation. J Cell Sci. 2003;116(Pt
ease. Am J Respir Crit Care Med. 1998;158(6):1951– 2):217–24.
7. doi:10.1164/ajrccm.158.6.9803053. 190. Rifkin DB. Latent transforming growth factor-beta
185. Lee CG, Cho SJ, Kang MJ, Chapoval SP, Lee (TGF-beta) binding proteins: orchestrators of TGF-­
PJ, Noble PW, et al. Early growth response gene beta availability. J Biol Chem. 2005;280(9):7409–
1-­ mediated apoptosis is essential for transform- 12. doi:10.1074/jbc.R400029200.
ing growth factor beta1-induced pulmonary fibro- 191. Chung KF. The role of airway smooth muscle in the
sis. J Exp Med. 2004;200(3):377–89. doi:10.1084/ pathogenesis of airway wall remodeling in chronic
jem.20040104. obstructive pulmonary disease. Proc Am Thorac
186. Sime PJ, Xing Z, Graham FL, Csaky KG, Gauldie Soc. 2005;2(4):347–54. doi:10.1513/pats.200504-­
J. Adenovector-mediated gene transfer of active 028SR; discussion 71–2.
transforming growth factor-beta1 induces pro- 192. Wiegman CH, Michaeloudes C, Haji G, Narang
longed severe fibrosis in rat lung. J Clin Invest. P, Clarke CJ, Russell KE, et al. Oxidative stress-­
1997;100(4):768–76. doi:10.1172/jci119590. induced mitochondrial dysfunction drives inflam-
187. Black PN, Young PG, Skinner SJ. Response of air- mation and airway smooth muscle remodeling in
way smooth muscle cells to TGF-beta 1: effects on patients with chronic obstructive pulmonary dis-
growth and synthesis of glycosaminoglycans. Am ease. J Allergy Clin Immunol. 2015;136(3):769–80.
J Physiol. 1996;271(6 Pt 1):L910–7. doi:10.1016/j.jaci.2015.01.046.
Part II
Pathophysiology of COPD
Eun Kyung Kim

Chronic obstructive pulmonary disease (COPD) Persistent reduction in forced expiratory flow
is a progressive inflammatory disease of the lung rates is the most typical finding in COPD. The
that involves complex interaction of cells and airflow limitation principally results from an
mediators. And it involves destruction of various increase in the resistance of the small conducting
tissue and repair mechanisms leading to struc- airways and a decrease in pulmonary elastic
tural changes that result in progressive airflow recoil due to emphysematous lung destruction.
limitation with incomplete reversibility [1]. Emphysema and small airway pathology are both
The pathological changes in patients with present in most patients with COPD; however,
COPD occur in the following compartments of they do not appear to be mechanistically related
the lungs: the central, large airways; the periph- to each other, and their relative contributions to
eral, small airways; the lung parenchyma; and the obstruction vary from one person to another [1].
pulmonary vasculature. Changes in large airways Increases in the residual volume and the residual
cause cough and sputum. The relative contribu- volume/total lung capacity ratio (RV/TLC), non-­
tion to the obstruction of the airways was made uniform distribution of ventilation, and
by the pathological changes in the small airways ventilation-­perfusion mismatching also occur.
and alveoli [2]. The pathologic abnormalities of The pathophysiology of COPD is complicated
COPD are associated with lung inflammation and largely undiscovered. This is complicated by
resulting from an imbalance of proteinases and the fact that there is heterogeneity of the disease,
antiproteinases, and oxidative stress induced by with some patients showing a predominant
noxious particles and gases in susceptible indi- emphysema pattern, whereas in others small air-
viduals [2]. The physiologic changes of COPD way disease predominates, although many
can be described with mucus hypersecretion, cili- patients have a mixed pattern. This chapter pro-
ary dysfunction, airflow limitation, pulmonary vides a general overview of the pathophysiology
hyperinflation, gas exchange abnormalities, pul- of COPD.
monary hypertension, and cor pulmonale.

 ucus Hypersecretion and Ciliary


E.K. Kim Innate immune response to inhaled toxic particles

Department of Internal Medicine, CHA Bundang and gases in cigarette smoke results in inflamma-
Medical Center, CHA University School of Medicine, tion in the epithelium of the central large airways
Seongnam, South Korea
e-mail:, and in the mucus-producing glands leading to

© Springer-Verlag Berlin Heidelberg 2017 57

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_5
58 E.K. Kim

cough and sputum production that define chronic small airway patency by destruction of alveolar
bronchitis [3]. Chronic mucus hypersecretion attachments, also reduces lung elastic recoil pres-
may be a reflection of the inflammatory response sure which leads to a reduced driving pressure for
in the submucosal glands, which is associated expiratory flow through narrowed and poorly
with increased mucus production, reduced muco- supported airways in which airflow resistance is
ciliary clearance [4, 5]. Inflammatory cells release significantly increased [12].
proteases that are potent secretagogues for mucus Airflow limitation, also known as airflow
[6]. Oxidants derived from cigarette smoke and obstruction, particularly during expiration is car-
released from inflammatory leukocytes may also dinal to COPD diagnosis. It is the result of the
be involved in overproduction of mucin by induc- balance between the elastic recoil of the lungs
tion of the MUC5AC gene [7]. Bronchi also promoting flow and the resistance of the airways
undergo squamous metaplasia, disrupting muco- limiting flow. Expiratory flow limitation arises
ciliary clearance and predisposing to carcinogen- because of the combined effects of airway nar-
esis. The contribution of mucus hypersecretion to rowing (caused by mucosal edema, mucus plug-
the airflow limitation in COPD is still uncertain. It ging, airway remodeling, and peribronchial
appears that it contributes little in the early stages fibrosis); reduced lung elastic recoil (reduced
of COPD because mucus production in smokers driving pressure for expiratory flow); and dis-
with normal lung function does not appear to pre- rupted alveolar attachments, which predispose to
dict later development of COPD [8]. However, dynamic airway collapse [13–15].
chronic mucus hypersecretion may contribute in In normal lungs, as well as in lungs affected
the later stages of the disease because of an by COPD, maximal expiratory flow diminishes
increased risk of exacerbations that may acceler- as the lungs empty because the lung parenchyma
ate the loss of FEV1. provides progressively less elastic recoil and
because the cross-sectional area of the airways
falls, raising the resistance to airflow. Due to
Airflow Obstruction reduced airway caliber and increased airway
resistance, there is reduced airflow during expira-
The pathological consequences of the COPD tion which prolongs the removal of air from the
induce series of physiological changes. Elastin lungs [16]. The volume of air remaining in the
proteolysis results in reduction in elastic recoil lung at the end of spontaneous expiration is
pressures in the lungs. It results in significant air- increased in COPD compared with healthy per-
way narrowing with reduction in air flow in the son [12]. During spontaneous breathing at rest in
small airways and air-trapping in the lungs. patients with COPD, which have airflow limita-
Fibrotic remodeling of the airways results in tion, end expiratory lung volume is also “dynami-
fixed airway narrowing causing increased airway cally” determined. In flow-limited patients, the
resistance which is not fully reversible even with mechanical time constant for lung emptying is
bronchodilators. A major site of airway obstruc- increased in most alveolar units, but the expira-
tion in COPD is the smaller conducting airways: tory time available is often insufficient. As a
membranous and respiratory bronchioles (<2 mm result, air-trapping occurs [12, 17].
in diameter) [9]. Inflammation and peribronchial The physiological feature of COPD is usually
fibrosis contribute to the fixed airway obstruction determined by spirometry, which detected with
in the small airways in COPD. Extensive alveolar forced expiratory maneuvers using spirometric
and bronchiolar epithelial cells and pulmonary lung volume ratio of volume of air removed in
capillary apoptosis result in histological feature first second (FEV1) and total volume of air
such as emphysema and physiological feature exhaled during the entire spirometric maneuver
such as decreased surface area of alveoli for gas [forced vital capacity (FVC)] during forceful
exchange and ventilation-perfusion (VA/Q) mis- expiration after maximal inhalation. Patients with
match [10, 11]. Emphysema does not remain airflow obstruction related to COPD have a
5  Pathophysiology of COPD 59

chronically reduced ratio of FEV1/FVC (FEV1/ [12, 22, 23]. This pushes the equi-pressure point
FVC <0.7). In contrast to asthma, the reduced further away from the alveoli. It is believed that
FEV1 in COPD seldom shows large responses to the symptomatic improvement produced by
inhaled bronchodilators although improvements bronchodilators is majorly by reduction in lung
up to 15% are common [18]. hyperinflation rather than bronchodilation [23].
Fixed airway obstruction depicts intensity of Hyperinflation of the thorax during tidal
small airway inflammation (mucosal edema, breathing may be beneficial, it preserves maxi-
fibrotic remodeling of the airway, and mucus mum expiratory airflow, because as lung volume
impaction) and possibly increased cholinergic increases, elastic recoil pressure increases, and
airway smooth muscle tone [5, 13, 19–21]. airways enlarge so that airway resistance
Contraction of airway smooth muscle, accumula- decreases.
tion of inflammatory cells, mucus, and plasma Despite compensating for airway obstruction,
exudate in airways might explain a part of the hyperinflation can push the diaphragm into a flat-
mechanism of airflow limitation and can be tened position with a number of adverse effects.
improved by the treatment. The zone of apposition between the diaphragm
and the abdominal wall is decreased. So, positive
abdominal pressure during inspiration is not
Hyperinflation applied as effectively to the chest wall, hindering
rib cage movement and impairing inspiration.
Lung hyperinflation or air-trapping is one of the Additionally, as the muscle fibers of the flattened
hallmarks of COPD pathophysiology. Increased diaphragm are shorter than those of a more nor-
residual volume, total lung capacity, functional mally curved diaphragm, they are less capable of
residual capacity (FRC), and increased RV/TLC generating inspiratory pressures than normal.
are common in COPD. Furthermore, the flattened diaphragm (with
Hyperinflation is primary cause of dyspnea, increased radius of curvature, r) must generate
poor quality of life, and advertent disease prog- greater tension (t) to develop the transpulmonary
nosis associated with COPD [12]. Damage to pressure (p) required to produce tidal breathing
elastin and narrowing of airways are major causes (This follows from Laplace’s law; p = 2 t/r). Also,
for air-trapping in COPD. The barrel-shaped because the thoracic cage is distended beyond its
chest in COPD is attributed to hyperinflation of normal resting volume, during tidal breathing the
the lungs. inspiratory muscles must do work to overcome
The inward lung elastic recoil pressures are the resistance of the thoracic cage to further infla-
the primary forces which drive air out of the tion instead of gaining the normal assistance
small airways and alveoli during tidal expiration from the chest wall recoiling outward toward its
for which it has to counteract outward recoil resting volume [12, 24, 25].
pressures generated by thoracic wall. During end The hyperinflation is dynamic in nature and
tidal expiration, both these forces equally balance is present in all stages of the disease, which is
each other causing relative airflow stasis and usually manifested in hyperventilation states
relaxed lung state. The volume of air trapped in which are outcome of expiratory flow limita-
the lungs at this equi-pressure stage is physiolog- tion, anxiety, and ventilation perfusion mis-
ically termed as FRC [12, 22]. matches related to COPD exacerbations or any
In COPD, weakening of elastic tissue gener- activity which increases oxygen demand. In
ates inadequate inward lung elastic recoil pres- hyperventilation during each breath, inhalation
sures to cause movement of air out of the lungs. commences before full exhalation is achieved,
Therefore to counteract the outward recoil pres- this results in air-­trapping which enhances with
sures of thoracic cage, the reduced elastic recoil each breath. With every breath FRC increases
of the lungs generates equi-pressure states with and inspiratory capacity reduces. When the
larger FRC resulting in state of hyperinflation rising FRC encroaches inspiratory reserve
60 E.K. Kim

volumes further increase in inspiratory muscle COPD, those with predominant emphysema have
activation produce little or no additional venti- high VA/Q areas within the lungs, whereas those
lation [26]. with predominant chronic bronchitis have low
Also, FRC no longer occurs at the passive VA/Q regions as a result of small airways distor-
point of equilibrium between chest wall and lung tion and mucus plugging [29, 30].
recoil, but occurs at a positive end-expiratory In mild-to-moderate COPD, Barbera and col-
pressure (PEEP) before exhalation has achieved leagues [31] and Rodriguez-Roisin and col-
the relaxation volume [26]. The increasing leagues [29] have demonstrated that significant
PEEP during dynamic hyperinflation VA/Q mismatching and loss of protective hypoxic
places diaphragm in a mechanical disadvantaged vasoconstriction can occur while breathing at
position shortens its operating length and alters rest. Thus, the resting alveolar-to-arterial oxygen
the mechanical linkage between its various parts. tension gradient was abnormally widened
This progressively diminishes the tension and (>15 mmHg) in most of a small sample of patients
resulting trans-diaphragmatic pressure generated with milder COPD who also had predominantly
by diaphragmatic contraction [12]. These altered low regional VA/Q ratios measured by multiple
actions of diaphragm can potentially cause dis- inert gas elimination techniques [31].
ability and impending respiratory failure during The attendant abnormalities in arterial blood
COPD exacerbations. gases, if sustained, stimulate integrated compen-
satory adaptations over time. Thus, activation of
neurohumoral, renal, and hemodynamic homeo-
Abnormality of Gas Exchange static mechanisms, together with modulation of
the central respiratory controller, combine to pre-
COPD is the most common chronic respiratory serve critical arterial oxygenation and acid-base
disease state associated with chronic and/or acute status.
respiratory insufficiency. The partial pressure of Preservation of arterial oxygenation during
oxygen in arterial blood (PaO2) usually remains exercise suggests that compensatory increases in
near normal until the FEV1 is decreased to ~50% ventilation in the setting of a normal increase in
of predicted, and even much lower FEV1 values cardiac output ensure improved overall VA/Q
can be associated with a normal PaO2, at least at relations during exercise in mild-to-moderate
rest. An elevation of arterial level of carbon diox- COPD [31]. Ultimately, in advanced COPD, the
ide (PaCO2) is not expected until the FEV1 is compensations may fail and reduced alveolar
<25% of predicted and even then may not occur ventilation at a given CO2 production leads to
[1, 27]. Hypoxemia occurs only during exercise CO2 retention.
in the early stage of the disease. However, it
appears to be at rest as the disease progress.
Uneven distribution of both alveolar ventila- Pulmonary Arterial Hypertension
tion and pulmonary blood flow, namely, VA/Q
mismatch, remains the most important cause of Pulmonary arterial hypertension (PAH) is a
arterial hypoxemia, with or without hypercapnia, known independent prognostic marker of
in both stable and exacerbated COPD [28]. COPD. Structural changes in COPD initiate
Non-uniform ventilation and VA/Q mismatch- instability in pulmonary hemodynamics. Most
ing are characteristic of COPD, reflecting the het- moderate-to-severe COPD patients develop some
erogeneous nature of the disease process within degree of mild PAH (25–35 mmHg) over the
the airways and lung parenchyma. Physiologic course of time, and rarely severe PAH
studies are consistent with multiple parenchymal (>45 mmHg) [32]; however, it may be rare in
compartments having different rates of ventila- mild-moderate COPD.
tion due to regional differences in compliance Pathophysiological consequences of COPD
and airway resistance. Among patients with such as hypoxia, emphysema, hyperinflation,
5  Pathophysiology of COPD 61

hypoxia-induced polycythemia, inflammation of which is associated with greater consumption of

lung and systemic inflammation can induce pul- oxygen by the respiratory muscles.
monary capillary muscularization, plexiform Abnormality of ventilation during acute exac-
lesions, intimal-wall thickenings, endothelial erbation of COPD was the result of greater nar-
dysfunctions, and apoptosis which can poten- rowing of the airways by inflammation of the
tially generate enhanced pulmonary vascular airway, bronchospasm, or hypersecretion of
resistance which predisposes PAH [32]. Increased mucus [20]. Pulmonary vasoconstriction due to
intrathoracic pressures in emphysema induce hypoxemia changed distribution of perfusion.
positive pressures on right ventricle which results During exacerbations, the oxygen demands of the
in increased pulmonary artery wedge pressure. respiratory muscles increase considerably due to
Left ventricular diastolic dysfunction associated greater airway resistance and to the additional
with cardiovascular morbidities in COPD can efforts of inspiratory muscles to overcome
also cause back pressure changes in pulmonary dynamic hyperinflation [31].
vasculature and right ventricles. Minute ventilation did not decrease during
Chronic severe pulmonary hypertension exacerbation. Instead, a slight increase was
increases right ventricular afterload and leads to shown when compared with stable conditions
the clinical syndrome (cor pulmonale) of right [28]. This finding reinforces the concept that the
heart failure with systemic congestion and the development of hypercapnia during exacerba-
inability to adapt right ventricular output to tions of COPD is essentially due to increased
peripheral vascular demands. Pulmonary hyper- VA/Q mismatch rather than hypoventilation.
tension severe enough to cause cor pulmonale Another thing should be considered in exac-
and right ventricular failure due to COPD typi- erbation of COPD. Airflow limitation during
cally occurs in individuals who have marked expiration is a pathophysiological hallmark of
decreases in FEV1 (<25% of predicted) and COPD. In patients with exacerbated COPD, the
chronic hypoxemia (PaO2 < 55 mmHg); however, time available for lung emptying (expiratory
recent evidence suggests that some patients will time) during spontaneous breathing is often
develop significant pulmonary hypertension insufficient to allow end expiratory lung volume
independent of COPD severity [33, 34]. A recent to decline to its natural relaxation volume, which
prospective study with a large number of patients leads to lung overinflation [36]. Thus, in those
with COPD who underwent right heart catheter- patients, end expiratory lung volume becomes
ization during exercise found abnormal eleva- dynamically rather than statically determined.
tions in pulmonary artery pressures as a function Dynamic hyperinflation refers to acute and vari-
of cardiac output, even in those without resting able increase in end expiratory lung volume
pulmonary hypertension [35]. above its baseline value. Dynamic hyperinflation
occurs during exercise in COPD patients as
inspired tidal volume increases and expiratory
 athophysiology of COPD
P time decreases, and is associated with severe
Exacerbation mechanical constraints on ventilation and per-
ceived respiratory discomfort. In this setting, the
Acute exacerbation is one of the most common reduction in inspiratory capacity (IC), which
complications in the evolution of COPD. These reflects the increase in end expiratory lung vol-
episodes are characterized by worsening of pul- ume, correlates strongly with the perception of
monary gas exchange that results in severe inspiratory difficulty. During COPD exacerba-
hypoxemia with or without hypercapnia irrespec- tions, airway resistance is abruptly increased
tive of the precipitating factor [28]. Altered gas (due to bronchospasm, narrowing of the airways
exchange during exacerbation of COPD results by inflammation of the airway, or hypersecretion
from the increased VA/Q mismatch together with of mucus) and this worsens airflow limitation
a reduced oxygen content of mixed venous blood, during expiration. Therefore, the time constant
62 E.K. Kim

for lung emptying is prolonged and end expira- 10. Tuder RM, Petrache I, Elias JA, Voelkel NF, Henson
PM. Apoptosis and emphysema: the missing link.
tory lung volume is dynamically increased.
Am J Respir Cell Mol Biol. 2003;28(5):551–4.
Furthermore, during an exacerbation, patients doi:10.1165/rcmb.F269.
tend to adopt a rapid shallow breathing pattern 11. Kasahara Y, Tuder RM, Cool CD, Lynch DA, Flores
which further limits the time available for lung SC, Voelkel NF. Endothelial cell death and decreased
expression of vascular endothelial growth factor
emptying, thus promoting dynamic hyperinfla-
and vascular endothelial growth factor receptor 2 in
tion in a vicious cycle. In fact, any acute increase emphysema. Am J Respir Crit Care Med. 2001;163(3
in ventilation can be associated with dynamic Pt 1):737–44. doi:10.1164/ajrccm.163.3.2002117.
hyperinflation in patients with COPD. There is 12. O’Donnell D, Laveneziana P. Physiology and conse-
quences of lung hyperinflation in COPD. Eur Respir
evidence that acute dynamic hyperinflation may
Rev. 2006;15(100):61–7.
be life threatening and requiring mechanical 13. McDonough JE, Yuan R, Suzuki M, Seyednejad N,
ventilation during severe exacerbations of Elliott WM, Sanchez PG, et al. Small-airway obstruc-
asthma or COPD [37]. tion and emphysema in chronic obstructive pulmo-
nary disease. N Engl J Med. 2011;365(17):1567–75.
14. Black LF, Hyatt RE, Stubbs SE. Mechanism of expi-
References ratory airflow limitation in chronic obstructive pulmo-
nary disease associated with 1-antitrypsin deficiency.
1. Pauwels RA, Buist AS, Calverley PM, Jenkins CR, Am Rev Respir Dis. 1972;105(6):891–9.
Hurd SS. Global strategy for the diagnosis, manage- 15. Leaver DG, Tatterfield AE, Pride NB. Contributions
ment, and prevention of chronic obstructive pulmonary of loss of lung recoil and of enhanced airways collaps-
disease. NHLBI/WHO global initiative for chronic ibility to the airflow obstruction of chronic bronchitis
obstructive lung disease (GOLD) workshop summary. and emphysema. J Clin Invest. 1973;52(9):2117–28.
Am J Respir Crit Care Med. 2001;163(5):1256–76. doi:10.1172/jci107396.
doi:10.1164/ajrccm.163.5.2101039. 16. Fry DL, Hyatt RE. Pulmonary mechanics: a unified
2. Hogg JC. Lung structure and function in COPD. Int analysis of the relationship between pressure, vol-
J Tuberc Lung Dis. 2008;12(5):467–79. ume and gasflow in the lungs of normal and diseased
3. Hogg JC. Pathophysiology of airflow limitation in human subjects. Am J Med. 1960;29(4):672–89.
chronic obstructive pulmonary disease. Lancet (London, 17. Vinegar A, Sinnett EE, Leith DE. Dynamic mecha-
England). 2004;364(9435):709–21. doi:10.1016/ nisms determine functional residual capacity in mice,
s0140–6736(04)16900-6. Mus musculus. J Appl Physiol Respir Environ Exerc
4. Fletcher C, Peto R. The natural history of chronic air- Physiol. 1979;46(5):867–71.
flow obstruction. Br Med J. 1977;1(6077):1645–8. 18. Mannino DM, Buist AS. Global burden of COPD:
5. Mullen JB, Wright JL, Wiggs BR, Pare PD, Hogg risk factors, prevalence, and future trends. Lancet.
JC. Reassessment of inflammation of airways 2007;370(9589):765–73.
in chronic bronchitis. Br Med J (Clin Res Ed). 19. Cosio M, Ghezzo H, Hogg JC, Corbin R, Loveland
1985;291(6504):1235–9. M, Dosman J, et al. The relations between struc-
6. Sommerhoff CP, Nadel JA, Basbaum CB, Caughey tural changes in small airways and pulmonary-­
GH. Neutrophil elastase and cathepsin G stimulate function tests. N Engl J Med. 1978;298(23):1277–81.
secretion from cultured bovine airway gland serous doi:10.1056/nejm197806082982303.
cells. J Clin Invest. 1990;85(3):682–9. doi:10.1172/ 20. Kuwano K, Bosken CH, Pare PD, Bai TR, Wiggs
jci114492. BR, Hogg JC. Small airways dimensions in asthma
7. Shao MX, Nakanaga T, Nadel JA. Cigarette smoke and in chronic obstructive pulmonary disease. Am
induces MUC5AC mucin overproduction via tumor Rev Respir Dis. 1993;148(5):1220–5. doi:10.1164/
necrosis factor-alpha-converting enzyme in human ajrccm/148.5.1220.
airway epithelial (NCI-H292) cells. Am J Physiol 21. Matsuba K, Wright JL, Wiggs BR, Pare PD, Hogg
Lung Cell Mol Physiol. 2004;287(2):L420–7. JC. The changes in airways structure associated with
doi:10.1152/ajplung.00019.2004. reduced forced expiratory volume in one second. Eur
8. Vestbo J, Lange P. Can GOLD stage 0 provide infor- Respir J. 1989;2(9):834–9.
mation of prognostic value in chronic obstructive 22. Brashier BB, Kodgule R. Risk factors and patho-

pulmonary disease? Am J Respir Crit Care Med. physiology of chronic obstructive pulmonary
2002;166(3):329–32. doi:10.1164/rccm.2112048. disease (COPD). J Assoc Physicians India.
9. Hogg JC, Macklem PT, Thurlbeck WM. Site and 2012;60(Suppl):17–21.
nature of airway obstruction in chronic obstructive 23. Ferguson GT. Why does the lung hyperinflate? Proc
lung disease. N Engl J Med. 1968;278(25):1355–60. Am Thorac Soc. 2006;3(2):176–9. ­ doi:10.1513/
doi:10.1056/nejm196806202782501. pats.200508-094DO.
5  Pathophysiology of COPD 63

24. Minh V, Dolan GF, Konopka RF, Moser KM. Effect ties and ventilation-perfusion relationships in mild
of hyperinflation on inspiratory function of the dia- chronic obstructive pulmonary disease. Am J Respir
phragm. J Appl Physiol. 1976;40(1):67–73. Crit Care Med. 1994;149(2 Pt 1):423–9. doi:10.1164/
25. Decramer M. Hyperinflation and respiratory muscle ajrccm.149.2.8306040.
interaction. Eur Respir J. 1997;10(4):934–41. 32. Chaouat A, Naeije R, Weitzenblum E. Pulmonary

Calverley PMA. Dynamic hyperinflation. Proc hypertension in COPD. Eur Respir J. 2008;32(5):1371–
Am Thorac Soc. 2006;3(3):239–44. doi:10.1513/ 85. doi:10.1183/09031936.00015608.
pats.200508-084SF. 33. Wright JL, Levy RD, Churg A. Pulmonary hyperten-
27. Lane DJ, Howell JB, Giblin B. Relation between air- sion in chronic obstructive pulmonary disease: current
ways obstruction and CO2 tension in chronic obstruc- theories of pathogenesis and their implications for
tive airways disease. Br Med J. 1968;3(5620):707–9. treatment. Thorax. 2005;60(7):605–9. doi:10.1136/
28. Barbera JA, Roca J, Ferrer A, Felez MA, Diaz O, Roger thx.2005.042994.
N, et al. Mechanisms of worsening gas exchange dur- 34. Sabit R, Bolton CE, Fraser AG, Edwards JM, Edwards
ing acute exacerbations of chronic obstructive pulmo- PH, Ionescu AA, et al. Sub-clinical left and right ven-
nary disease. Eur Respir J. 1997;10(6):1285–91. tricular dysfunction in patients with COPD. Respir Med.
29. Rodriguez-Roisin R, Drakulovic M, Rodriguez DA, Roca 2010;104(8):1171–8. doi:10.1016/j.rmed.2010.01.020.
J, Barbera JA, Wagner PD. Ventilation-­perfusion imbal- 35. Hilde JM, Skjorten I, Hansteen V, Melsom MN, Hisdal J,
ance and chronic obstructive pulmonary disease staging Humerfelt S, et al. Haemodynamic responses to exercise
severity. J Appl Physiol (1985). 2009;106(6):1902–8. in patients with COPD. Eur Respir J. 2013;41(5):1031–
doi:10.1152/japplphysiol.00085.2009. 41. doi:10.1183/09031936.00085612.
30. Wagner PD, Dantzker DR, Dueck R, Clausen JL,
36. O’Donnell DE, Parker CM. COPD exacerbations.

West JB. Ventilation-perfusion inequality in chronic 3: pathophysiology. Thorax. 2006;61(4):354–61.
obstructive pulmonary disease. J Clin Invest. doi:10.1136/thx.2005.041830.
1977;59(2):203–16. doi:10.1172/jci108630. 37.
Calverley PM. Respiratory failure in chronic
31. Barbera JA, Riverola A, Roca J, Ramirez J, Wagner obstructive pulmonary disease. Eur Respir J Suppl.
PD, Ros D, et al. Pulmonary vascular abnormali- 2003;47:26s–30s.
Diagnosis and Assessment
of COPD 6
Yong Bum Park

Introduction Diagnosis

A clinical diagnosis of COPD should be consid- A clinical diagnosis of COPD should be consid-
ered in patients over the age of 40 who present ered in patients over the age of 40 who present with
with dyspnea, chronic cough or sputum produc- dyspnea, chronic cough or sputum production, and
tion, and a history exposure to risk factors for a history exposure to risk factors for the disease.
the disease. Spirometry is fundamental to mak- Spirometry is fundamental to making a diagnosis
ing a diagnosis of COPD, and the presence of a of COPD, and the presence of a post-­bronchodilator
post-­bronchodilator FEV1/FVC < 0.70 confirms FEV1/FVC < 0.70 confirms the presence of persis-
the presence of persistent airflow limitation. tent airflow limitation. The spirometric criteria for
COPD assessment is to determine the severity airflow limitation remains a post-bronchodilator
of disease, including the severity of airflow lim- fixed ratio of FEV1/FVC < 0.70.
itation, the impact on the patient’s health status, The use of the fixed FEV1/FVC ratio to define
and the future risk. COPD assessment is impor- airflow limitation is not perfect. Using this cutoff
tant for therapy and prognosis. COPD often may lead to an overdiagnosis of COPD in the
coexists with comorbidities that may have sig- elderly [1], and an underdiagnosis in young adults
nificant impact on mortality and prognosis. At [2]. Because of these imperfections in the FEV1/
the initial assessment and follow-up visits FVC ratio, there have been several alternate meth-
should include a discussion of symptoms, par- ods proposed to diagnose obstruction. One pro-
ticularly any new or worsening symptoms and posal is to use the lower limit of normal (LLN) for
physical examination. the cutoff of FEV1/FVC ratio. The LLN takes into
consideration the age, height, and gender for each
individual. This minimizes the age-related changes
in the FEV1/FVC ratio and may reduce the mis-
classification of airway obstruction [2]. However,
Y.B. Park it also may be more difficult for primary care clini-
Division of Pulmonary, Allergy and Critical Care
Medicine, Hallym University Medical Center,
cians to perform and interpret. In addition, using
Seoul, South Korea this cutoff may not improve clinical care.
Department of Internal Medicine, Hallym University
It may be difficult for some patients to per-
Kangdong Sacred Heart Hospital, Seoul, South Korea form a forced exhalation maneuver for several
Lung Research Institute of Hallym University
reasons, including poor mental status and cough-
College of Medicine, Seoul, South Korea ing. Another option is the FEV1/FEV6 ratio. The
e-mail: FEV1/FEV6 ratio has been shown to be an

© Springer-Verlag Berlin Heidelberg 2017 65

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_6
66 Y.B. Park

acceptable surrogate for the FEV1/FVC ratio [3]. The presence of purulent sputum reflects an
It is also thought to be an easier test for patients increase in inflammatory conditions, and its devel-
to perform than the FVC maneuver. opment may identify the onset of bacterial exacer-
Unfortunately, a considerable number of sub- bation [7]. Wheezing and chest tightness are
jects had not been diagnosed [4, 5]. Frequently the nonspecific symptoms that may vary between
disease will not be diagnosed until it is quite pro- days. An absence of wheezing and chest tightness
gressed. Targeted screening of symptomatic does not exclude a diagnosis of COPD, nor does
patients with risk factors for COPD results in better the presence of these symptoms confirm a diagno-
diagnosis rates and more appropriate therapy [6]. sis of asthma. Fatigue, weight loss, and anorexia
are common problems in patients with severe and
very severe COPD. Ankle swelling can be a sign
Symptoms of cor pulmonale. Symptoms of depression and/or
anxiety occur and are associated with increased
The characteristic symptoms of COPD are chronic risk of exacerbations and poorer health status.
and progressive dyspnea, cough, and sputum pro-
duction that can be variable from day to day and
from individual to individual. Chronic cough and Medical History
sputum production may precede the development
of airflow limitation by many years. Conversely, A detailed medical history in a patient with
significant airflow limitation may develop without COPD should assess:
chronic cough and sputum production. In the
early stages, COPD may produce minimal or no Exposure to risk Smoking and occupational and
factors environmental exposures including
symptoms and as the disease progresses the symp- biomass fuel
toms in individual patients vary. Symptoms Past medical Asthma, atopy, bronchial
should be assessed as objectively as possible by history hyperresponsiveness; respiratory
using questionnaires such as the mMRC, CCQ, infection in childhood
and CAT. In general, dyspnea is the symptom Other respiratory infections
including tuberculosis
which causes them most concern. A person may
Family history COPD, asthma, alpha-1 antitrypsin
seek medical facility either because of chronic of COPD deficiency
symptoms or because of a first exacerbation. Symptom Cough, sputum, dyspnea
Dyspnea is major cause of disability and anxi- Progressive, chronic
ety associated with the disease. Dyspnea is typi- Typically develops in adult life
cally present only with exertion until late in the More frequent or prolonged “winter
course of the disease. Cough may be the present- colds”
ing symptom and may be very troublesome and Recent medical History of exacerbations and
can compromise quality of life significantly. history previous hospitalization for
respiratory disorder
However, cough is frequently neglected by the
Presence of Cardiovascular disease,
patient as an expected result of smoking or envi- comorbidities osteoporosis, musculoskeletal
ronmental exposure. Initially, the cough may be disorder, malignancy
intermittent, but later is present every day. In some Impact of Limitation of activity, missed work
cases, significant airflow limitation may develop disease on and economic impact, effect on
patient’s life family routines, feeling on
without the presence of a cough. Sputum is insidi-
depression and anxiety, well-being
ous in the majority of patients. Regular production and sexual activity
of sputum for 3 or more months in 2 consecutive Support for the Family, society, and daycare
years (in the absence of any other conditions that patient
may explain it) is the epidemiological definition of Possibilities for Especially smoking cessation
chronic bronchitis. Patients producing large vol- reducing risk
umes of sputum may have underlying bronchiectasis.
6  Diagnosis and Assessment of COPD 67

Physical Examination whether airflow limitation is due to emphysema

or airway disease.
Since the typical physical findings in COPD usu-
ally do not appear until the disease has become Spirometry
severe, the absence of abnormal findings does not Spirometry is the most important test to diagnose
rule out the possibility of COPD. When COPD COPD. Peak expiratory flow measurement may
becomes severe, patients present more apparent significantly underestimate the severity of the air-
physical signs. These include the barrel shaped flow limitation. Spirometry should measure the
chest, purse-lipped breathing, and debility. volume of air forcibly exhaled from the point of
Pulmonary hypertension may develop in patients maximal inspiration (forced vital capacity, FVC)
with COPD and may be worse during exercise. It and the volume of air exhaled during the first sec-
may be suspected with a loud pulmonary compo- ond of this maneuver (forced expiratory volume
nent of the second heart sound and heart sounds in one second, FEV1) and the ratio of these two
may be displaced to the midline due to measurements (FEV1/FVC) should be calculated.
hyperinflation. A reduction in FEV1/FVC ratio is diagnostic of
airway obstruction. An FEV1/FVC ratio of <0.70
after bronchodilator is typically considered diag-
I ntroduction to Pulmonary Function nostic of COPD. The FEV1/FVC ratio can estab-
Testing lish a diagnosis of obstruction but is not useful to
monitor disease progression and severity of
Pulmonary function testing has three basic com- COPD. Along with spirometry, a flow-volume
ponents: (1) spirometry, (2) lung volumes, and loop is also typically generated. A flow-volume
(3) diffusing capacity of the lung for carbon mon- loop plots flow on the y-axis and volume on the
oxide (DLCO). Each of these components can be x-axis. A normal flow-volume loop has a charac-
affected by COPD. The most important test is teristic shape (Fig. 6.1a). In obstructive lung dis-
spirometry. Measurement of lung volume and ease such as COPD, the expiratory limb takes on
diffusion capacity may be helpful, particularly a coved shape (Fig. 6.1b).

a b
Obstructive Disease
Expiration Expiration
EF Peak 75%VC EF Peak
6 Emphysema
6 EF
5 50%VC Bronchitis
4 4 0.5 sec
3 EF
25%VC 1.0 sec
2 0.5 sec 2 3.0 sec
Flow 1/sec

Flow 1/sec

1.0 sec
0 0
6 75%VC 25%VC


7 IF 50%VC 6
0 1 2 3 4 5 0 1 2 3 4 5
Volume [liters] Volume [liters]

Fig. 6.1  Flow-volume curve for healthy person (a) and patients with airflow obstruction (b)
68 Y.B. Park

Lung Volumes DLCO, the patient inhales a single breath contain-

In addition to spirometry, lung volumes can also ing a minute amount of CO and holds it for 10 s.
be measured. Lung volumes consist of total lung The breath is then exhaled and the exhaled breath
capacity (TLC), residual volume (RV), and func- is analyzed for CO. The change in the concentra-
tional residual capacity (FRC). The TLC is the tion of the CO is then multiplied by the single
volume of air contained in the lung after a full breath TLC to calculate the DLCO. Some patients
inhalation. The RV is the volume of air left in the with severe COPD may have difficulty perform-
lung after a full exhalation. The FRC is the vol- ing the breath hold required to measure DLCO.
ume of gas left in the lungs after a tidal breath. The DLCO is decreased in proportion to the
The most commonly used method is the body severity of emphysema of the destruction of the
plethysmography. Emphysema destroys lung tis- alveoli and the loss of capillary bed. It is also
sue, leading to loss of elastic recoil. Loss of elas- reduced in other diseases that destroy the alveolar
tic recoil allows the lungs to be stretched to capillary bed.
abnormally large volumes, resulting in an
increased TLC. RV and FRC can also be increased
in COPD when disease progression destroys the Assessment of the Disease
elastic tethers. This leads to premature closing of
the airways, which causes abnormal amounts of COPD is heterogeneous, so no single measure
air to be trapped in the lung. In some patients, can give an adequate assessment of the severity
there is also inflammation of the small airways, of the disease in an individual patient. However,
which causes narrowing, further contributing to severity assessment is important because it has
air trapping and the increase in RV. significances for therapy and relates to prognosis.
Air trapping can lead to hyperinflation. There is COPD assessment is to determine the severity of
both a static and dynamic component of hyperinfla- disease, including the severity of airflow limita-
tion. Static hyperinflation refers to the baseline level tion, the impact on the patient’s health status, and
of air trapping seen at rest. This is due to the loss of the future risk (such as exacerbation, hospital
elastic recoil properties of the lung and fixed airway admission, or death).
obstruction. Dynamic hyperinflation occurs during Global initiative for chronic obstructive lung
exercise or times of rapid respiratory rate. In these disease (GOLD) [8] documents that COPD
situations, the patient is unable to finish exhaling assessment must consider the following aspects
before the next breath starts. With each breath, the of the disease separately to achieve these goals.
patient becomes progressively more hyperinflated. • Current levels of patient’s symptoms.
This causes inefficient respiratory muscles function • Severity of spirometric abnormality.
and increases the work of breathing. • Exacerbation risk.
• Presence of comorbidity.
 iffusing Capacity of the Lung
for Carbon Monoxide
The diffusing capacity of the lung for carbon The Assessment of Symptoms
monoxide (DLCO) is a measure of how easily
carbon monoxide (CO) molecules transfer from A simple measurement of dyspnea such as the
the alveolar gas to the hemoglobin of the red cells Modified British Medical Research Council
in the pulmonary circulation. To measure the (mMRC) Questionnaire was considered adequate
6  Diagnosis and Assessment of COPD 69

for assessment of symptoms, as the mMRC respiratory symptoms that is beyond normal day-
relates well to other measures of health status [9] to-day variations and leads to a change in medica-
and predicts future mortality risk [10]. However, tion [11–13]. The rate of exacerbations varies
it is now recognized that COPD has broad range between patients. Exacerbations of COPD are
of effects on health status. For this reason, a com- important, since they have a serious negative
prehensive symptom assessment is recommended impact on health status, and are associated with
rather than just a measure of dyspnea. accelerated lung function decline and increased
The most comprehensive disease-specific mortality. Although exacerbations are considered
health-related quality of life or health status ques- to become more frequent as the severity of the
tionnaires such as CRQ and SGRQ are too com- underlying COPD increases, the most reliable pre-
plex to use in routine practice, but two shorter dictor of exacerbation appears to be a history of
comprehensive measures (COPD Assessment Test, exacerbations.
CAT and COPD Control Questionnaire, CCQ) Classification of exacerbations was assessed
have been developed and are suitable. To find more according to Anthonisen et al. [14]. A type I exac-
about symptomatic assessment tool, refer to the erbation is defined by the presence of three major
next section (Symptomatic assessment of COPD). symptoms (worsening dyspnea with increased
sputum volume and purulence), type II by the
presence of two major symptoms and type III
Spirometric Assessment exacerbations as the presence of one major symp-
tom present on the worst day of an exacerbation.
The table shows the classification of airflow limi- Definitions of exacerbations vary between clini-
tation severity in COPD. cal studies that evaluate drugs to prevent exacer-
bations. Recent clinical trials define moderate
Classification of severity of airflow limitation in COPD exacerbations as those that require treatment with
(post-bronchodilator FEV1 in patients with FEV1/
FVC < 0.70) systemic corticosteroids or antibiotics (or both)
GOLD 1: Mild FEV1 ≥ 80% predicted and severe exacerbations as those that require
GOLD 2: 50% ≤ FEV1 < 80% predicted hospital admission.
Moderate Exacerbations of chronic pulmonary disease
GOLD 3: Severe 30% ≤ FEV1 < 50% predicted tool (EXACT) was a single, standardized instru-
GOLD 4: Very FEV1 < 30% predicted ment of collecting patient-reported outcome
severe (PRO) data, which helps to capture not only the
frequency of exacerbations but also the severity
Spirometry should be performed after the administra- and the time-course [15, 16].
tion of a short-acting inhaled bronchodilator. However,
there is only a weak correlation between FEV1, symp-
toms, and patient’s health-­related quality of life. Additional Investigations

 ximetry and Arterial Blood Gas

Assessment of Exacerbation Risk Measurement
Pulse oximetry should be used to assess all sta-
An exacerbation of COPD is defined as an acute ble patients with FEV1 < 35% predicted or with
event characterized by a worsening of the patient’s clinical signs suggestive of respiratory failure or
70 Y.B. Park

right heart failure. If peripheral saturation is Exercise Testing

<92%, arterial blood gases should be assessed. Exercise tests are useful for evaluating exercise tol-
Arterial blood gases show mild or moderate erance, identifying factors that limit exercise, and
hypoxemia without hypercapnia in the early assessing the effectiveness of pulmonary rehabilita-
stage of COPD. In the later stages of the dis- tion. Exercise disability is a predictor of prognosis
ease, hypoxemia tends to become more severe and marker of health status. Both the paced shuttle
and may be accompanied by hypercapnia with walk tests and the 6-min walk test can be used.
increased serum bicarbonate levels. Blood gas Cardiopulmonary exercise test using cycle or tread-
abnormalities worsen during acute exacerba- mill can identify coexisting or alternative conditions
tions and may also worsen during exercise and such as cardiac diseases or neuromuscular diseases.
Composite Scores
Several variables including low FEV1, a low
 lpha-1 Antitrypsin Deficiency
A functional exercise capacity, a high degree of
Screening dyspnea, low body-mass index, and presence of
Mutation in alpha-1 protease inhibitor represents hypoxemia or hypercapnia are associated with an
the only proven genetic abnormality that predis- increased risk for mortality. A relatively simple
poses to COPD. The typical patient tends to pres- approach to identifying disease severity using a
ent at a younger age (<45 years) with lower lobe combination of most of the above variables has
emphysema. A serum concentration of alpha-1 been proposed. All these approaches need valida-
antitrypsin below 15–20% of the normal value is tion across a wide range of disease severities and
highly suggestive of homozygous alpha-1 anti- in different clinical settings to confirm that they
trypsin deficiency. are suitable for routine clinical use.

Summary table of multidimensional index of COPD:

PFT Dyspnea BMI Smoking Exacerbation QoL Co-morbidity Age










BODE: the body-mass index (B), the degree of airflow ADO: age (A), dyspnea (D), and the degree airflow
obstruction (O) and dyspnea (D), and exercise capacity obstruction (O)
(E), measured by the 6-min walk test CPI: COPD prognostic index
mBODE: modified BODE SAFE: SGRQ score, air-flow limitation and exercise
eBODE: exacerbation + BODE tolerance
BODEx: the body-mass index (B), the degree of airflow DOSE: dyspnea (D), airflow obstruction (O), smoking status
obstruction (O) and dyspnea (D), and severe exacerbation (Ex) (S), and exacerbation frequency (E)
uBODE: updated BODE
6  Diagnosis and Assessment of COPD 71

The Classification of COPD • Patient Group A—Low risk, Less symptoms

• Patient Group B—Low risk, More symptoms
The GOLD consensus report proposed a new • Patient Group C—High risk, Less symptoms
classification for COPD in 2011 to more compre- • Patient Group D—High risk, More symptoms
hensively assess disease severity [17]. This new In GOLD Report 2017, it has been revised and
classification system combined the symptoms in ABCD groups are derived from patients’ symp-
addition to COPD exacerbation history and air- toms and their history of exacerbation. Spirometry
flow limitation measured by FEV1. still has its role for diagnosis and prognostication
but out of classification criteria.

≥ 2 or
≥ 1 leading to hospital
(C) (D) admission

(Exacerbation History)
1 (not leading to
hospital admission)
(A) (B)

CAT < 10 CAT ≥ 10


mMRC 0-1 mMRC ≥2


In Korea, a revised COPD guideline was cutoff value for the high risk of exacerbation dif-
released in 2012 and updated in 2014 by the ferent from that of GOLD report. The Korean
Korean Academy of Tuberculosis and Respiratory COPD guideline stratified first on the basis of risk
diseases [18]. The new Korean COPD guideline of exacerbation with either FEV1 (<60% or
also emphasized the combined assessment for ≥60%) or exacerbation history (0–1 vs. ≥2) or
COPD patients. However, there were some dif- history of hospitalization due to exacerbation (0
ferences between the Korean guideline and vs. ≥1) in the previous year resulting in low risk
GOLD consensus report. The Korean guideline or high risk group (group “da”). The low risk
classified the COPD patients into three groups. group then stratified with mMRC or CAT like in
The Korean guideline combined GOLD C, D GOLD report resulting in low symptom (group
group into one group and used the spirometry “ga”) or high symptom (group “na”) group.
72 Y.B. Park

≥ 2 or
≥ 1 leading to hospital

(FEV1 % predicted value)

Group Da admission

(Exacerbation History)
< 60%


1 (not leading to
hospital admission)
≥ 60% Group Ga Group Na

CAT < 10 CAT ≥ 10
mMRC 0-1 mMRC ≥2

Korean COPD classification system

In Spanish COPD guideline, it was proposed proposed phenotypes are: (A) infrequent exacer-
that four different phenotypes of prognostic and bators with either chronic bronchitis or emphy-
therapeutic relevance characterized by the com- sema; (B) overlap COPD-asthma; (C) frequent
bination of the classical types of emphysema, exacerbators with emphysema predominant; and
chronic bronchitis, exacerbators, and patients (D) frequent exacerbators with chronic bronchitis
with overlap COPD-asthma be defined [19]. The predominant (Fig. 6.1).

Exacerbator Exacerbator
Phenotype Phenotype
≥ 2 exacerbations/year with chronic
with emphysema

Non-exacerbator Non-exacerbator Phenotype

< 2 exacerbations/year

Emphysema Chronic bronchitis

Phenotype Phenotype

Clinical phenotypes proposed by GesEPOC

Differential Diagnosis normal range excludes a diagnosis of COPD. In

some patients with chronic asthma, a clear distinc-
The most difficult clinical problem is to distin- tion from COPD is not possible using current
guish asthma from COPD. Spirometry and clinical imaging and physiological examination, and it is
characteristics are the useful method. Improvement assumed that asthma and COPD coexist in these
in lung function and clinical course is the charac- patients. Other potential diagnoses are usually
teristic feature of asthma. Improvement into the easier to distinguish from COPD.
6  Diagnosis and Assessment of COPD 73

COPD and its differential diagnoses include a discussion of symptoms, particularly

COPD Onset in mid-life any new or worsening symptoms and physical
Symptoms slowly progressive examination.
History of smoking or exposure
to occupational and
environmental exposures
including biomass fuel
 onitor Disease Progression
Asthma Onset early in life (often
and Development of Complications
Symptoms vary widely from Some patients with COPD deteriorate faster than
day to day others and it is important to identify these indi-
Symptoms worse at night/early viduals who may need specialist referral and
morning investigation. Decline in lung function is best
Allergy, rhinitis, and/or eczema tracked by spirometry performed at least once a
also present
year or more frequently if indicated. Questionnaires
Family history of asthma
such as the CAT can be performed every 2 or
Congestive heart Chest radiography shows
failure enlarged heart and pulmonary 3 months; trends and changes are more valuable
edema than single measurements.
Pulmonary function test At each visit, ask about changes in symptoms
indicated restrictive ventilatory since the last visit, including cough and sputum,
breathlessness, fatigue, activity limitation, and
Bronchiectasis Large volumes of purulent
sleep disturbances.
Commonly associated with
At each visit, identify current smoking status
bacterial infection and offer smoking cessation advice. Encouraging
Chest radiography shows patients with COPD to stop smoking is one of the
bronchial dilatation and most important components of their management.
bronchial wall thickening
Tuberculosis Onset all ages
Chest radiography shows lung  onitor Pharmacotherapy and Other
infiltration and/or cavity
formation Medical Treatment
Microbiological confirmation
High prevalence of tuberculosis Adherence to the maintenance treatment, inhaler
technique, effectiveness of the current regimen,
and side effects of treatment should be

Routine follow-up is essential in COPD. Monitor Exacerbation History

However, there are no data to guide decisions on
how frequently patients should be reviewed but The assessment of an exacerbation is based on
clearly this will vary according to individual the patient’s medical history and clinical signs of
circumstances and the severity of the patient’s severity and some laboratory test, if available.
disease. Symptoms and objective measures of Assessthe frequency, severity, and possible
airflow limitation should be monitored to causes of any exacerbations. Unscheduled visits,
determine when to modify therapy and to iden- use of emergency care, and hospitalization are
tify complications that may develop. At the ini- important. Severity of exacerbation can be
tial assessment and follow-up visits should estimated by the increased need for bronchodilator
74 Y.B. Park

or corticosteroids and by the need for antibiotic diagnosis and management of COPD. Eur Respir
J. 2006;28(5):945–52.
treatment. Hospitalizations should be docu-
7. Stockley RA, O’Brien C, Pye A, Hill SL. Relationship
mented, including duration of stay, and any use of sputum color to nature and outpatient manage-
of intensive care unit or mechanical ventilator ment of acute exacerbations of COPD. Chest.
support. 2000;117(6):1638–45.
8. Global Initiative for Chronic Obstructive Lung
Disease. Global strategy for the diagnosis, manage-
ment and Prevent of chronic obstructive pulmonary
Monitor Comorbidities disease, updated 2015.
9. Bestall JC, Paul EA, Garrod R, Garnham R, Jones PW,
Wedzicha JA. Usefulness of the Medical Research
COPD often coexists with other disease that may
Council (MRC) dyspnoea scale as a measure of dis-
have a significant impact on prognosis. ability in patients with chronic obstructive pulmonary
Comorbidities are common at any severity of disease. Thorax. 1999;54(7):581–6.
COPD. The focus should be on identification and 10. Nishimura K, Izumi T, Tsukino M, Oga T. Dyspnea
is a better predictor of 5-year survival than air-
management of these individual problems in line
way obstruction in patients with COPD. Chest.
with local treatment guidance. 2002;121(5):1434–40.
11. Rodriguez-Roisin R. Toward a consensus definition
for COPD exacerbations. Chest. 2000;117(5 Suppl
References 12. Burge S, Wedzicha JA. COPD exacerbations: definitions
and classifications. Eur Respir J Suppl. 2003;41:46s–53s.
1. Hardie JA, Buist AS, Vollmer WM, Ellingsen I, Bakke 13. Celli BR, Barnes PJ. Exacerbations of chronic obstruc-
PS, Morkve O. Risk of over-diagnosis of COPD in tive pulmonary disease. Eur Respir J. 2007;29(6):
asymptomatic elderly never-smokers. Eur Respir 1224–38.
J. 2002;20(5):1117–22. 14. Anthonisen NR, Manfreda J, Warren CP, Hershfield
2. Swanney MP, Ruppel G, Enright PL, Pedersen ES, Harding GK, Nelson NA. Antibiotic therapy in
OF, Crapo RO, Miller MR, et al. Using the lower exacerbations of chronic obstructive pulmonary dis-
limit of normal for the FEV1/FVC ratio reduces ease. Ann Intern Med. 1987;106:196–204.
the misclassification of airway obstruction. Thorax. 15. Leidy NK, Wilcox TK, Jones PW, et al. Development
2008;63(12):1046–51. of the EXAcerbations of chronic obstructive pulmo-
3. Jing JY, Huang TC, Cui W, Xu F, Shen HH. Should nary disease tool (EXACT): a patient-reported out-
FEV1/FEV6 replace FEV1/FVC ratio to detect come (PRO) measure. Value Health. 2010;13:965–75.
airway obstruction? A metaanalysis. Chest. 16. Jones PW, Chen WH, Wilcox TK, et al. Characterizing
2009;135(4):991–8. and quantifying the symptomatic features of COPD
4. Yoo KH, Kim YS, Sheen SS, Park JH, Hwang YI, exacerbations. Chest. 2011;139:1388–94.
Kim SH, Yoon HI, Lim SC, Park JY, Park SJ, Seo 17. Global Initiative for Chronic Obstructive Lung

KH, Kim KU, Oh YM, Lee NY, Kim JS, Oh KW, Disease. Global strategy for the diagnosis, manage-
Kim YT, Park IW, Lee SD, Kim SK, Kim YK, Han ment and Prevent of chronic obstructive pulmonary
SK. Prevalence of chronic obstructive pulmonary disease, revised 2011.
disease in Korea: the fourth Korean National Health 18. Korean Academy of Tuberculosis and Respiratory

and nutrition examination survey, 2008. Respirology. Diseases. COPD guideline revised 2014. http://www.
5. Mannino DM, Gagnon RC, Petty TL, Lydick 19. Miravitlles M, Soler-Cataluña JJ, Calle M, Molina J,
E. Obstructive lung disease and low lung function in Almagro P, Quintano JA, Riesco JA, Trigueros JA,
adults in the United States: data from the National Piñera P, Simón A, Rodríguez-Hermosa JL, Marco E,
Health and nutrition examination survey, 1988-1994. López D, Coll R, Coll-Fernández R, Lobo MÁ, Díez J,
Arch Intern Med. 2000;160(11):1683–9. Soriano JB, Ancochea J. Spanish guideline for COPD
6. Walker PP, Mitchell P, Diamantea F, Warburton CJ, (GesEPOC). Arch Bronconeumol. 2014;50(Suppl
Davies L. Effect of primary-care spirometry on the 1):1–16. doi:10.1016/S0300-2896(14)70070-5.
Symptomatic Assessment of COPD
Paul W. Jones

Introduction Another statement that is still sometimes

heard is: ‘I can measure lung function, so I don’t
This chapter is mainly about formal assessment need to measure symptoms’. That argument is
and quantification of symptoms, rather the ele- now difficult to sustain because there is a large
ments of eliciting a clinical history, although the body of evidence to show that the FEV1 is only
latter can be informed by the scientific analysis weakly correlated with other clinically important
needed to create an instrument that can measure outcomes such as breathlessness, exercise toler-
symptoms. Whilst measurement has become a ance, exacerbations, and health status. This is
fundamental part of medicine, it has come only illustrated very well by data from the ECLIPSE
slowly into that most basic clinical assessment— study [1]. Whilst there is a statistically significant
assessing symptoms and their impact on the correlation between FEV1 and these other out-
patient’s daily life and sense of well-being. The comes at a population level, these associations
latter statement may trigger the question—‘Why are much too weak to be of value when assessing
should I measure symptoms, since I am already individual patients (Fig. 7.1). For example, a
trained to assess a patient from their history and patient with moderate airflow limitation (e.g.
form a clinical judgement?’ There is a general set 70% predicted) may have very good exercise
of answers to this question that apply to any clini- capacity, as measured by their 6-min walking dis-
cal measurement: formal should remove observer tance (6MWD), and very good health status,
bias; numbers can be recorded more easily than measured using the St George’s Respiratory
descriptions; formal assessment with a common Questionnaire (SGRQ), but equally they could
language can aid communication between health- have very poor 6MWD and SGRQ score.
care professional (and between healthcare profes- It goes without saying that a measurement is
sional and patients); perhaps most importantly, it only as good as the reliability of the instrument
allows a reliable record to be made of changes used and the manner in which it is used.
over time, since formal numerical assessments Symptomatic assessment in pulmonary medicine
should be independent of whoever is making them. was first used in clinical studies, using well-­
validated instruments that were too complex for
routine use, but these have provided a lot of
insight into the health status of patients with
P.W. Jones, Ph.D., F.R.C.P., F.E.R.S. COPD and factors that influence it, together with
Institute for Infection and Immunity,
its response to treatment and progression over
St George’s, University of London,
Cranmer Terrace, London SW170RE, UK

© Springer-Verlag Berlin Heidelberg 2017 75

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_7
76 P.W. Jones

a Rho=–0.36 b Rho=–0.34

6MWD (Metres)
mMRC Score

2 600

1 400

0 200

0 20 40 60 80
Post-dose FEV1 (% Predicted) 20 40 60 80
Post-dose FEV1 (% Predicted)

c Rho=–0.21 d Rho=–0.38
6 80
Number of exacerbations

SGRQ-C total score

3 40

0 0
20 40 60 80
0 20 40 60 80
Post-dose FEV1 (% Predicted)
Post-dose FEV1 (% Predicted)

Fig. 7.1  Correlation between post-bronchodilator FEV1 and: (a) mMRC score; (b) 6-min walking distance; (c) number
of exacerbations and (d) SGRQ. From [1]

time. Newer health status instruments are simpler from 1–5 to 0–4. For that reason, it is important
to use and can be used reliably in routine clinical to be clear which version is being used, the
practice. modified version being abbreviated to mMRC.
Throughout the rest of this section, the generic
name MRC will be used. The original MRC
Breathlessness scale is shown in Fig. 7.2, and it will be seen
that it might be better described as a measure
It is appropriate to begin with breathlessness, of disability associated with breathlessness. It
since this is a core symptom in COPD that leads is not a direct measure of breathlessness such
to impaired exercise capacity and poor quality of as the visual analogue scale (VAS) and Borg
life; it was also the first symptom to be measured scale.
in a standardized way, through the Medical The MRC scale was initially developed as an
Research Council (MRC) Dyspnoea Scale that epidemiological tool and was only fully validated
was developed 60 years ago. Confusingly when as a measure of COPD severity many years later
modified by the American Thoracic Society for [2]. Studies showing that the MRC was a better
use in the USA, the scaling range was changed predictor of mortality than the FEV1 [3], led to its
7  Symptomatic Assessment of COPD 77

MRC dyspnoea scale developed for clinical trials. Now a daily diary
Grade 1: Breathless on strenuous exercise card has been created, which is now called the
Evaluating Respiratory Symptoms in COPD
Grade 2: Short of breath when hurrying (E-RS), although when first developed it was
or walking up a slight hill
known as the EXACT-RS [7]. It has 11 items that
Grade 3: Walk slower than others provide a total score and three domains:
or stop when walking at own pace on level ground Breathlessness (five items); Cough and Sputum
(three items) and Chest Symptoms (three items).
Grade 4: Stop every 100 m
or after a few minutes
The latter includes factors such as chest tight-
ness. The E-RS has been validated [8] and full
Grade 5: Too breathless to leave the house
details are available from the website although
or breathless on washing/dressing
users will have to register to get full details
Fig. 7.2  MRC dyspnoea scale (­
descriptions/). Whilst daily diaries are not practi-
cal for routine use, the finding that there are three
incorporation into the BODE prognostic instru- scientifically determined domains within stable
ment [4]. The chief disadvantage of the MRC COPD symptoms provides useful confirmation
scale is that the intervals between the different of established clinical experience. The E-RS has
grades are very large and, at an individual patient recently been shown to be responsive to pharma-
level, few patients will improve with treatment by cological treatment [9].
one grade and fortunately few patients will dete-
riorate by one grade over months to years of
follow-up. Exacerbations
The relative insensitivity of the MRC led to
the development of the Baseline and Transition It is appropriate to include exacerbations in a
Dyspnea Indices (BDI and TDI), like the MRC chapter devoted to assessment of symptoms
scale they are indirect measures of breathlessness because exacerbations are symptomatic events.
since they link dyspnoea to activity [5]. Furthermore, the patient’s history of exacerba-
Interestingly, the simple direct measures of tions appears to be the single best predictor of
breathlessness, such as the VAS and Borg scales, future risk of exacerbations [10], so clinical
have not found application in routine pulmonary assessment and standardization are important.
medicine, outside the exercise laboratory, unlike Like respiratory symptoms, exacerbations
scales for pain which commonly use that type of have only recently been subject to rigorous scien-
methodology. A more recent 12-item dyspnoea tific analysis [11] this supported the development
scale (the Dyspnoea-12) was developed with of the EXACT diary card [12] http://www.exact-
modern psychometric methods using descrip- The
tions used by patients to describe their breath- E-RS symptom diary, described above, was
lessness [6]. It has been validated, but has not yet developed from the 14-item EXACT, so it is per-
found its way into routine clinical practice, haps not surprising that it has the same three
although it is short enough to be practical in that domains: breathlessness, chest symptoms, and
setting. cough and sputum. Until the development of the
EXACT, the consensus definition of an exacerba-
tion was: ‘a sustained worsening of the patient’s
Respiratory Symptoms condition, from the stable state and beyond nor-
mal variation, that is acute in onset’ [13]; later
No standardized measures of respiratory symp- ‘and leads to a change in treatment’ was added
toms have been developed for use in routine prac- [14]. It will be noted that the ‘worsening’ and
tice; in fact, until recently they hadn’t even been ‘day-to-day variation’ are not defined and neither
78 P.W. Jones

is ‘sustained’ although the common definition in Health Status—Complex

clinical trials requires symptoms to have wors- Questionnaires
ened for 2–3 days. The criterion about change in
symptoms necessitating a change in treatment led COPD is a very complex disease and health sta-
to the creation of a severity grading: mild (use of tus questionnaires attempt to quantify its overall
extra routine medication such short-acting bron- impact on a patient’s daily life and well-being.
chodilator), moderate (use of systemic cortico- Health status scores provide a marker of a per-
steroid and/or antibiotics), severe (moderate plus son’s health-related quality of life (HRQoL), but
hospital admission or emergency room atten- they should not be called HRQoL scores. The
dance). Though very widely used, these severity distinction between health status and HRQoL is
grades have never been validated. In contrast, the important. HRQoL is a clinical outcome that is
EXACT has strict criteria for determining that unique to each person, since the factors that
an acute change in symptom score is an determine his or her quality of life will differ,
exacerbation. depending on their personality and circum-
Although it is a research tool, the development stances. For the same reason, illness will affect
of the EXACT has been important for routine HRQoL in a way that is unique to each person. It
practice for two main reasons. First, it has empha- is not possible to make standardized measure-
sized that there are distinct symptomatic compo- ments that allow comparisons between patients
nents of an exacerbation, not just an increase in and treatments because the components of the
cough with sputum production. Furthermore, the scale will differ between patients. In contrast, a
contribution of each component to the patients health status questionnaire is standardized
worsening may vary between exacerbations and marker of impaired HRQoL that is made up of
patients; indeed, an exacerbation may meet the items that should be common (at least poten-
EXACT criteria for an exacerbation without the tially) to every patient; this means that they assess
patient having a worsening of cough and sputum. each patient as if he/she was a ‘typical’ patient—
A second and perhaps even more important con- in the same way that % predicted FEV1 treats
tribution of the EXACT has been to show that every patient as being typical of someone of the
patients experience frequent unreported exacer- same age, height, sex and race.
bations that appear to have similar short-term The first health status measure to be developed
(6 months) consequences as those that are was the Chronic Respiratory Questionnaire
reported [15]. The reasons for the lack of report- (CRQ) [16]. It is widely used although largely in
ing are ill-understood. Unreported exacerbations the context of rehabilitation. It is too complex for
may occur in patients who are a little less severe routine clinical use although it is often used to
than those patients who do report them, but a monitor and evaluate rehabilitation programmes.
more striking observation is that there are big dif- It has four domains and through these it drew
ferences in reporting rates between countries that attention to two important aspects of COPD that
may not be related to differences in severity. A had been largely under-recognized; these are
possible reason for under-reporting may be that fatigue and mastery, the latter being the patient’s
patients do not understand that worsening of sense of being in control of their condition. One
chest symptoms with a viral upper respiratory of the challenges with questionnaires designed to
tract infection is not a feature of a cold, but is in measure overall health status is the production of
fact a worsening of their chest condition. This a valid total score. Interestingly, the CRQ’s devel-
suggests that physicians should enquire specifi- oper never approved the calculation of a total
cally for symptoms of exacerbations (not just score, but researchers often calculate one,
cough and sputum), and a simple severity guide although its validity has never been established.
would be to ask whether the worsening of chest The SGRQ was developed and validated in
symptoms and breathlessness was sufficient to both asthma and COPD [17], but its use has
disturb the patient’s daily routine. largely been confined to COPD, although recently
7  Symptomatic Assessment of COPD 79

it has been used successfully to assess the similarly large improvements seen with lung vol-
response to treatment in patients with severe ume reduction for emphysema [22], but it should
asthma [18]. To address the problem of creating a be appreciated that both of these treatments are
valid total score, each item has specific weight, invasive and are only used for patients at the
which means that its score is best calculated severe end of the disease-spectrum. In the much
using a computer. A revised and slightly shorter larger population of less severe COPD patients,
40-item set was created and termed the SGRQ for the benefits with inhaled therapy is smaller,
COPD (SGRQ-C) [19]. Manuals and over 60 lan- although modern once-daily long-acting bron-
guage versions are available for download (www. chodilators appear to be more effective than older A user-friendly scoring twice-daily agents and show average improve-
system for both versions is now available for a ments that are at the level of the MCID [20]. In
number of platforms ( contrast, pulmonary rehabilitation, which is a
treatment that is appropriate to all COPD patients,
produces an average improvement in SGRQ
Health Status—Insights score that is approximately 1.5 times the MCID
from Research Studies [23]. The benefits from these complementary
treatment modalities are likely to be additive, so
The SGRQ is too complex for routine use, but effective bronchodilation followed by rehabilita-
much has been learnt from its application to tion should, on average, lead to a patient achiev-
research studies that is applicable to routine care. ing a good improvement in their HRQoL.
There is a large body of data from clinical trials One of the interesting observations that has
about a range of different treatments. For exam- come from pharmacological treatment trials in
ple, a recent network meta-analysis has shown COPD has come from the placebo arms of the tri-
quite large differences between different bron- als. These typically shows an improvement in
chodilators in SGRQ scores (and breathlessness SGRQ score that is around half the MCID. This
measured by the TDI) [20]. A significant contri- ‘clinical trial effect’ is not understood. It is not
bution has been the demonstration that SGRQ immediate in onset; the patients slowly improve,
scores may detect a worthwhile patient benefit in typically for up to 6 months, and the improvement
the absence of large improvements in objective is maintained for 1 year (Fig. 7.3). The mecha-
physical measurements. For example, following nisms are not understood. One factor may be a
placement of endobronchial coils for severe Hawthorne effect, in which the patient, their phy-
emphysema, the improvement in SGRQ was sician or both of them, change their behaviour
twice the minimum clinically important differ- when a patient enters a trial. This could include
ence (MCID), despite there being only small better compliance with concomitant therapy, life-
improvements in FEV1 and 6MWD [21]. The style changes and perhaps earlier intervention if
ability of the SGRQ to quantify health status there were any worsening. There may be some-
improvements from the patient’s perspective has thing in the doctor–patient relationship that also
now been recognized by the Food and Drugs improves the patient’s sense of well-­being; per-
Administration (FDA) in the USA which has haps giving the patient greater confidence in man-
qualified it for use as a co-primary outcome in aging their condition, lessening the effect that it
new drug registration trials. has on their lives. One randomized trial of antide-
At this point, it should be appreciated that pressant therapy has shed a little light on this phe-
health status questionnaires are designed to be nomenon since it showed a bigger improvement in
‘treatment agnostic’—i.e. a change in score patients who were seen regularly in the trial com-
should indicate the same health status benefit, pared to those who were seen only at the begin-
regardless of the therapeutic mechanism of ning and the end [24]. Whatever the mechanism,
action. The large improvement seen in the study these observations suggest that patients do receive
of endobronchial coils, just described, reflects benefit just from regular medical review.
80 P.W. Jones

a b

Adjusted mean change SGRQ

3 0
Adjusted mean change SGRQ

SYMPTOM score (points)

2 –1 Placebo
TOTAL score (points)

1 –2 Salmeterol
0 –3 Fluticasone proprionate
–4 Salmeterol/FP
-3 –7
-4 –8
-5 –9

0 24 48 72 96 120 156 0 24 48 72 96 120 156

Time (weeks) Time (weeks)
c d

Adjusted mean change SGRQ

Adjusted mean change SGRQ

4 2

IMPACT score (points)

ACTIVITY score (points)

1 0
–1 –2
–2 –3
–3 –4
–4 –5

0 24 48 72 96 120 156 0 24 48 72 96 120 156

Time (weeks) Time (weeks)

Fig. 7.3  Changes in SGRQ score over 156 weeks in the TORCH trial. (a) Total score; (b) Symptoms score; (c) Activity
score; (d) Impacts score. Data from [26]

Apart from quantifying therapeutic benefit, trial effect described earlier). The second study
perhaps the most significant contribution of showed different patterns of worsening in patients
health status measurement has been to provide a related to their change in FEV1 over time. Patients
better understanding of changes in the patients’ who were fast FEV1 decliners showed the fastest
health over time. This has been seen and mea- deterioration in SGRQ, exceeding the MCID
sured with both disease-specific questionnaires after 4 years; whereas slow decliners showed lit-
and with generic physical and mental health, as tle worsening [29]. Most interestingly, those
measured with the SF-36 [25]. At least two dis- patients whose FEV1 was maintained actually
ease factors are known to be associated with showed an improvement in SGRQ that exceed
worsening health status: decline in FEV1 [26] and the MCID by 4 years. In the latter study and in
frequency of exacerbations [27]. Patients do not the TORCH clinical trial ([26] and Fig. 7.3),
deteriorate at the same rate and two cohort stud- there was a different pattern of disease progres-
ies from Japan provide useful insights into the sion across the domains of the SGRQ; the symp-
relationship between worsening FEV1 and wors- toms domain score showed a sustained
ening health status. In one, there was no deterio- improvement regardless of change in FEV1 (in
ration in health status over the first 18 months, the cohort study) or treatment group (TORCH
but thereafter the SGRQ score deteriorated pro- trial), whereas there was a worsening in the
gressively, whilst the FEV1 showed little consis- Activity and Impacts scores across treatment
tent directional change [28]. The initial period of groups in TORCH and, if the FEV1 deteriorated,
health status stability may be due, not only to the in the cohort study. In the latter study, the Activity
impact of treatment when the patient joined the component did not improve, even if FEV1 did not
clinic, but also due to the effect of keeping them worsen, in fact there was a general trend for
under review in the clinic (analogous to the clinical Activity to worsen. These findings all point to
7  Symptomatic Assessment of COPD 81

heterogeneity of the pattern of health status wors- with the established measures such as the CRQ
ening, both in the degree and character of that [35], SGRQ [36] and CCQ [37]. It has also been
change. Finally, deterioration in SGRQ score proven to be a good predictor of exacerbations in
may be a predictor of worse health in the future patients who were already at higher risk of exac-
since an analysis of the ECLIPSE study suggests erbations because of an exacerbation in the pre-
that patients who deteriorate by more than the ceding year.
MCID have a greater risk of hospitalization and There are many publications about the CAT
death 2 years later [30]. that further tested its validity and responsiveness
There is evidence that pharmacological treat- to treatment, and these are detailed in two recent
ment can have an impact on worsening in health systematic reviews [38, 39]. Data summarized in
status. The ISOLDE study showed that inhaled these reviews suggest that it appears to be as
corticosteroid (ICS) alone could alter the rate of responsive to rehabilitation as the CRQ and
worsening of SGRQ and general health status SGRQ, with a change in score, compared to usual
scores [25]. More recently, the TORCH study care, that is ≈1.5 times the MCID (note: the indi-
(long-acting beta2-agonist (LAMA) plus ICS) vidual patient MCID for the CAT = 2 units [40]).
and UPLIFT (long-acting anti-muscarinic In terms of pharmacological treatment, the CAT
(LAMA)) showed that maintenance treatment has been shown to be as sensitive as the SGRQ in
can produce improvements in health status that response to ICS/LAMA and LAMA/LABA treat-
may be maintained for 3 years or more [26, 31]. ment [41]; the mean improvement with both
The true size of the effect may have been under-­ treatments was 2.2–2.4 units, i.e. well above the
estimated in these trials because of differential MCID.
drop out of the sickest patients who had the worst
health status, an effect that was greatest in the
placebo arm [32]. Symptomatic Assessment
in Routine Practice

Health Status—Shorter There are perhaps three broad areas in COPD

Questionnaires management where formal assessment of symp-
toms has a place: baseline assessment, evaluating
Whilst a great deal has been learnt from health the response to treatment and monitoring.
status measurement in clinical studies, the com- The simplest assessment of a COPD patient is
plexity of the questionnaires prevented them to ask about the severity of their disease; how-
from being used in routine clinical practice, but ever, there is good evidence that this may lead to
there are now two that are suitable for use in that an under-estimated of the true impact of the dis-
setting. The first to be developed was the Clinical ease. For example, 36% of patients who said that
COPD Questionnaire (CCQ) [33]. It has ten their disease was only mild or moderate also indi-
items, exists in daily and weekly versions, and is cated that they were MRC Grade 5, i.e. they were
available in 60 languages. A website gives more breathless when washing and dressing or too
details ( The 8-item COPD breathless to leave the house [42]. In 2011, the
Assessment Test (CAT) (Fig. 7.4) was developed Global Initiative for Chronic Obstructive
to facilitate communication between doctor and Pulmonary Disease (GOLD) revision of its
patient, but with good measurement properties so assessment scheme ( recom-
that it would be suitable for use in clinical trials mended, for the first time, a process based on
[34]. The website contains a user guide and symptoms and risk of exacerbations. It suggested
allows online completion and scoring in 64 lan- use of either the mMRC or the CAT, recommend-
guages ( Mean CAT ing the maintenance treatment should be started
scores appear to be very consistent across countries with a mMRC score ≥ 2, or CAT ≥ 10 for the
(Fig. 7.5). The CAT has shown good correlations CAT. Unfortunately, it made an error in equating
82 P.W. Jones

Fig. 7.4 COPD
Assessment Test (CAT).
Available from www.

these thresholds because subsequently it has been The mMRC can be used if the CAT is not avail-
shown that mMRC ≥ 2 equates to CAT > 20 [43], able, but with the caveat that the available evi-
which is a ‘severe impact’ according to the CAT dence suggests that the equivalent threshold is
handbook ( In fact, mMRC 0, not >1. Confusingly, mMRC Grade 0
CAT ≥ 10 corresponds approximately to mMRC does not mean the absence of symptoms (which
Grade 0 (breathless on strenuous exercise). More would be zero on the original MRC scale), it is
recently, GOLD has emphasized the importance breathlessness with strenuous exercise. In this,
of the CAT since it measures much more than just we see an example in which some degree of
breathlessness, in fact only 25% of the items in breathlessness is seen as being ‘normal’ in
the CAT are concerned with activity limitation. COPD.
7  Symptomatic Assessment of COPD 83

Fig. 7.5  CAT scores 30

from clinics in different
countries. Data from the 25
following references:
Country 1 [45]; Country 20
2 [36]; Country 3 [46];
Country 4 [47] CAT 15






Ne aly





Arabic Countries3



G Hong Kong2

South Korea
Reference 1

The assessment of the response to treatment A quick assessment of response to treatment

requires special consideration. The limits of day-­
• Have you noticed a difference?
day repeatability of most measurements in
• What have you noticed, for example*
COPD (including the FEV1) are wider than the
• Are you less breathless?
size of the MCID, so pairs of measurements • Can you do more?
before and after treatment may give a misleading • Can you do things more easily or quickly?
indication of the size of change in an individual • Can you sleep better?
patient. However, there is evidence that patients’ • Tell me one thing that you have noticed is better
global ratings of treatment efficacy may, on aver-
Fig. 7.6 A simple method of evaluating response to
age, provide a reliable measure of the overall
symptomatic treatment in COPD patients
treatment effect. In a 16-week study of salme-
terol in COPD, a 4-unit change in SGRQ (the
MCID for the SGRQ) was associated with a is that the patient should be able to specify one
global estimate of ‘effective’ treatment and an activity or symptomatic aspect of their disease
8-unit change was ‘very effective’ [44]. In con- that has noticeably improved. If they can’t do
trast, a response that was ‘satisfactory’ was asso- that, it is reasonable to assume that they haven’t
ciated with a change that was less than two units. had a worthwhile benefit.
Thus, the patient’s global estimate of a treatment Perhaps the most useful application of health
effect has some scientific validity! Questionnaires status questionnaires is in routine monitoring.
are not the right way to assess an individual’s Since the CAT and the SGRQ are closely corre-
response to treatment, however, because each lated, a picture of the different patterns of change
patient may respond to treatment and notice ben- in CAT score may be created from a number of
efit in a different way. For example, a woman sources discussed earlier. These are illustrated in
may once again be able to play nine rounds of Fig. 7.7. A combination of effective bronchodila-
golf, whilst a man may find that he can clean his tion and rehabilitation should produce a total
apartment without having to stop for breath. A improvement of at least 3–4 units, which should
traditional clinical history-taking approach is be maintained within a 2-unit range. On average,
required to evaluate benefit. The method used by the worsening should be ≤1 unit per year. A
this author illustrated in Fig. 7.6. It should be cause should be sought if there is any worsening
emphasized that the most important component beyond that range. For example: less adherence
84 P.W. Jones

Fig. 7.7  A schematic of 24

changes in CAT score
that may be seen in an 22 Inhaler technique
individual with
COPD. The green 20
Two unreported
shaded area shows the exacerbations
range of usual between CAT 18
visit variations that score
might be expected. If a 16
worsening outside that
range is seen, enquiries 14
should be made to
12 Bronchodilator
identify a cause Baseline +
assessment rehabilitation
1 4 8 20 32 48 52 60 72 90 102


to treatment, failure of inhaler technique, fre- 3. Nishimura K, Izumi T, Tsukino M, Oga T. Dyspnea
is a better predictor of 5-year survival than
quent exacerbations (reported or unreported) and
airway obstruction in patients with COPD. Chest.
rapid decline of lung function. Monitoring with 2002;121(5):1434–40.
the CAT is much easier than spirometry which is 4. Celli BR, Cote CG, Marin JM, Casanova C, Montes
more time consuming and may be sensitive to a de Oca M, Mendez RA, et al. The body-mass index,
airflow obstruction, dyspnea and exercise capac-
wider range of disease factors.
ity index in chronic obstructive pulmonary disease.
NEJM. 2004;350:1005–12.
5. Mahler DA, Weinberg DH, Wells CK, Feinstein
Summary AR. Measurements of dyspnea. Contents, interob-
server correlates of two new clinical indices. Chest.
Formal symptom assessment is now an estab- 6. Yorke J, Moosavi SH, Shuldham C, Jones
lished place in COPD management. The intro- PW. Quantification of dyspnoea using descriptors:
duction of simple health status instruments such development and initial testing of the Dyspnoea-12.
Thorax. 2010;65(1):21–6.
as the CAT has made it possible to make reli-
7. Leidy NK, Sexton CC, Jones PW, Notte SM, Monz
able, routine and valid assessments of the over- BU, Nelsen L, et al. Measuring respiratory symptoms
all impact of COPD on a patient’s health. They in clinical trials of COPD: reliability and validity of a
do not replace clinical assessment or lung func- daily diary. Thorax. 2014;69(5):443–9.
8. Leidy NK, Murray LT, Monz BU, Nelsen L, Goldman
tion, but add to the information available to the
M, Jones PW, et al. Measuring respiratory symptoms
clinician in deciding management of their of COPD: performance of the EXACT- Respiratory
patient. Symptoms Tool (E-RS) in three clinical trials. Respir
Res. 2014;15(1):124.
9. Jones PW, Leidy NK, Hareendran AR, Lamarca,
Chuecos F, Garcia Gil E. The effect of aclidin-
References ium bromide on daily respiratory symptoms of
COPD, measured using the Evaluating Respiratory
1. Agusti A, Calverley PMA, Celli B, Coxson HO, Symptoms in COPD (E-RS: COPD) diary: pooled
Edwards L, Lomas D, et al. Characterisation of COPD analysis of two 6-month phase III studies. Respir
heterogeneity in the ECLIPSE cohort. Respir Res. Res. 2016;17(1):61.
2010;11(1):122. 10. Hurst JR, Vestbo J, Anzueto A, Locantore N,

2. Bestall JC, Paul EA, Garrod R, Garnham R, Jones PW, Müllerova H, Tal-Singer R, et al. Susceptibility to
Wedzicha JA. Usefulness of the Medical Research exacerbation in chronic obstructive pulmonary dis-
Council (MRC) dyspnoea scale as a measure of dis- ease. N Engl J Med. 2010;363(12):1128–38.
ability in patients with chronic obstructive pulmonary 11. Jones PW, Chen W-H, Wilcox TK, Sethi S, Leidy
disease. Thorax. 1999;54(7):581–6. NK, Group E-PS. Characterizing and quantifying the
7  Symptomatic Assessment of COPD 85

symptomatic features of COPD exacerbations. Chest. 27.

Spencer S, Calverley PMA, Burge PS, Jones
2011;139(6):1388–94. PW. Impact of preventing exacerbations on deterio-
12. Leidy NK, Wilcox TK, Jones PW, Roberts L, Powers ration of health status in COPD. Eur Respir J. 2004;
JH, Sethi S, et al. Standardizing measurement of 23(5):698–702.
chronic obstructive pulmonary disease exacerbations. 28. Oga T, Nishimura K, Tsukino M, Sato S, Hajiro T,
Am J Respir Crit Care Med. 2011;183(3):323–9. Mishima M. Longitudinal deteriorations in patient
13. Rodriguez-Roisin R. Toward a consensus definition reported outcomes in patients with COPD. Respir
for COPD exacerbations. Chest. 2000;117(5 Suppl Med. 2007;101(1):146–53.
2):398S–401S. 29. Nagai K, Nishimura M, Makita H, Suzuki M,

14. Burge S, Wedzicha JA. COPD exacerbations: defini- Shimizu K, Konno S, et al. Differential changes in
tions and classifications. Eur Respir J. 2003;21(41 quality of life components over 5 years in chronic
suppl):46s–53s. obstructive pulmonary disease patients. COPD.
15. Jones PW, Lamarca R, Chuecos F, Singh D, Agustí 2015;10(1):745–57.
A, Bateman ED, et al. Characterisation and impact of 30. Wilke S, Jones PW, Mullerova H, Vestbo J, Tal-Singer
reported and unreported exacerbations: results from R, Franssen FME, et al. One-year change in health
ATTAIN. Eur Respir J. 2014;44(5):1156–65. status and subsequent outcomes in COPD. Thorax.
16. Guyatt GH, Berman LB, Townsend M, Pugsley
SO, Chambers LW. A measure of quality of life 31. Tashkin DP, Celli B, Senn S, Burkhart D, Kesten S,
for clinical trials in chronic lung disease. Thorax. Menjoge SS, et al. A 4-year trial of tiotropium in
1987;42:773–8. chronic obstructive pulmonary disease. N Engl J Med.
17. Jones PW, Quirk FH, Baveystock CM, Littlejohns 2008;359(15):1543–54.
P. A self-complete measure for chronic airflow limita- 32. Vestbo J, Anderson JA, Calverley PMA, Celli B,

tion—the St George’s respiratory questionnaire. Am Ferguson GT, Jenkins C, et al. Bias due to with-
Rev Respir Dis. 1992;145:1321–7. drawal in long-term randomised trials in COPD:
18. Ortega HG, Liu MC, Pavord ID, Brusselle GG,
evidence from the TORCH study. Clin Respir
FitzGerald JM, Chetta A, et al. Mepolizumab treat- J. 2011;5(1):44–9.
ment in patients with severe eosinophilic asthma. N 33. van der Molen T, Willemse BWM, Schokker S, ten
Engl J Med. 2014;371(13):1198–207. Hacken NHT, Postma DS, Juniper EF. Development,
19. Meguro M, Barley EA, Spencer S, Jones PW.
validity and responsiveness of the clinical COPD ques-
Development and validation of an improved COPD- tionnaire. Health Qual Life Outcomes. 2003;1(1):13.
specific version of the St George’s respiratory ques- 34. Jones PW, Harding G, Berry P, Wiklund I, Chen

tionnaire. Chest. 2007;132:456–63. WH, Kline LN. Development and first validation
20. Cope S, Donohue JF, Jansen JP, Kraemer M, Capkun-­ of the COPD assessment test. Eur Respir J. 2009;
Niggli G, Baldwin M, et al. Comparative efficacy of 34(3):648–54.
long-acting bronchodilators for COPD—a network 35. Jones PW, Harding G, Wiklund I, Berry P, Tabberer
meta-analysis. Respir Res. 2013;14(1):100. M, Yu R, et al. Tests of the responsiveness of the
21. Sciurba FC, Criner GJ, Strange C, et al. Effect of endo- COPD assessment test following acute exacerbation
bronchial coils vs usual care on exercise tolerance in and pulmonary rehabilitation COPD assessment test
patients with severe emphysema: the renew random- responsiveness. Chest J. 2012;142(1):134–40.
ized clinical trial. JAMA. 2016;315(20):2178–89. 36. Kwon N-H, Amin M, Hui DS, Jung K-S, Lim SY,
22. National Emphysema Treatment Trial Research
Ta HD, et al. Validity of the chronic obstructive pul-
Group. Cost effectiveness of lung-volume-reduction monary disease assessment test (CAT) translated
surgery for patients with severe emphysema. NEJM. into local languages for Asian patients. Chest. 2013;
2003;348:2092–102. 143:703–10.
23. Bernard M, Dympna C, Declan D, Kathy M, Edel M, 37. Ställberg B, Nokela M, Ehrs P-O, Hjemdal P, Jonsson
Yves L. Pulmonary rehabilitation for chronic obstruc- EW. Validation of the clinical COPD questionnaire
tive pulmonary disease. Cochrane Database Syst Rev. (CCQ) in primary care. Health Qual Life Outcomes.
2015;2:CD003793. 2009;7(1):1–9.
24. McCarney R, Warner J, Iliffe S, van Haselen R, Griffin 38. Gupta N, Pinto LM, Morogan A, Bourbeau J. The
M, Fisher P. The Hawthorne effect: a randomised, con- COPD assessment test: a systematic review. Eur
trolled trial. BMC Med Res Methodol. 2007;7(1):30. Respir J. 2014;44(4):873–84.
25. Spencer S, Calverley PMA, Burge PS, Jones
39. Karloh M, Mayer AF, Maurici R, Pizzichini MMM,
PW. Health status deterioration in patients with Jones PW, Pizzichini E. The COPD assessment test:
chronic obstructive pulmonary disease. Am J Respir what do we know so far?: a systematic review and
Crit Care Med. 2001;163:122–8. meta-analysis about clinical outcomes prediction
26. Jones P, Anderson J, Calverley P, Celli B, Ferguson G, and classification of patients into gold stages. Chest.
Jenkins C, et al. Health status in the TORCH study of 2016;149(2):413–25.
COPD: treatment efficacy and other determinants of 40. Kon SSC, Canavan JL, Jones SE, Nolan CM,

change. Respir Res. 2011;12(1):71. Clark AL, Dickson MJ, et al. Minimum clinically
86 P.W. Jones

important difference for the COPD assessment 44. Jones PW, Bosh TK. Changes in quality of life in
test: a prospective analysis. Lancet Respir Med. COPD patients treated with salmeterol. Am J Respir
2014;2(3):195–203. Crit Care Med. 1997;155:1283–9.
41. Singh D, Worsley S, Zhu C-Q, Hardaker L, Church 45. Jones PW, Brusselle G, Dal Negro RW, Ferrer M,
A. Umeclidinium/vilanterol versus fluticasone propi- Kardos P, Levy ML, et al. Properties of the COPD
onate/salmeterol in COPD: a randomised trial. BMC assessment test in a cross-sectional European study.
Pulm Med. 2015;15:91. Eur Respir J. 2011;38(1):29–35.
42. Rennard S, DeCramer M, Calverley PMA, Pride NB, 46. Jones PW, Shahrour N, Nejjari C, Lahlou A, Doble A,
Soriano JB, Vermeire PA, et al. Impact of COPD in Rashid N, et al. Psychometric evaluation of the COPD
North America and Europe in 2000: subjects’ perspec- assessment test: data from the BREATHE study in the
tive of confronting COPD international survey. Eur Middle East and North Africa region. Respir Med.
Respir J. 2002;20:799–805. 2012;106(Suppl 2):S86–99.
Jones PW, Adamek L, Nadeau G, Banik 47. Tsuda T, Suematsu R, Kamohara K, Kurose M,

N. Comparisons of health status scores with MRC Arakawa I, Tomioka R, et al. Development of the
grades in COPD: implications for the GOLD 2011 Japanese version of the COPD assessment test. Respir
classification. Eur Respir J. 2013;42(3):647–54. Invest. 2012;50(2):34–9.
Imaging of COPD
Sang Min Lee, Song Soo Kim, Hye Jeon Hwang,
and Joon Beom Seo

Computed Tomography (CT) Z-axis and two-dimensional detectors are devel-

oped, acquisition of a number of slices per rota-
CT Basic Physics tion is possible. Recently developed multi-slice
and multi-detector CT allows cone beam CT and
Computed tomography (CT) of the thorax is a volumetric imaging. After CT scan, the X-ray
very quick and the most advanced imaging tech- attenuation, called tissue density, is expressed as
nique. CT scanner is designed to use a form of a Hounsfield Units (HU). The scale of HU values
gantry which allows rotation of the X-ray tube range from −1000 HU (attenuation value of air)
and the detector around the patient during breath to 3000 HU, and 0 HU corresponds to the attenu-
hold. With the development of CT technology, ation of water (Fig. 8.1). Generally, normal lung
single breath-hold scanning of the thorax can be has an attenuation value around −850 HU on
achieved, and image reconstruction of sagittal inspiratory CT because normal lung attenuation
and coronal planes is also feasible with minimal reflects the mixed attenuation of intrapulmonary
loss of spatial resolution. During the decades, the air and lung parenchyma.
fan beam of a CT scanner is broadened along the
 adiation Dose of COPD CT
The radiation dose during CT scan is presented as
the gray or mGy unit, which is proportional to the
amount of energy that the irradiated body part is
S.M. Lee, M.D. • J.B. Seo, M.D., Ph.D. (*) expected to absorb. The Sievert (Sv) unit is used
Division of Cardiothoracic Radiology, Department of in the report of the effective dose. Regarding
Radiology, Asan Medical Center, University of Ulsan chest CT scan for evaluation of chronic obstruc-
College of Medicine, Seoul, South Korea
tive pulmonary disease (COPD), acceptable low
e-mail:;, dose screening and standard dose of CTs can be
performed at effective dose of approximately 2
S.S. Kim, M.D.
Department of Radiology, Chungnam National and 7 mSv, respectively [1].
University Hospital, Chungnam National University
School of Medicine, Daejeon, South Korea CT Protocol
Optimization of CT protocol and quality control
H.J. Hwang, M.D. of image acquisition are critical for assessment of
Department of Radiology, Hallym University College
COPD [2, 3]. Ideally, single CT protocol using
of Medicine, Hallym University Sacred Heart
Hospital, Seoul, South Korea dedicated single CT scanner and software system
e-mail: based on exactly the same parameters of image

© Springer-Verlag Berlin Heidelberg 2017 87

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_8
88 S.M. Lee et al.



–1000–950 –400 –100 –60 0 40 80 400 3000
Soft tissue

Fig. 8.1  Hounsfield unit on CT images. The Hounsfield played according to target organs adjusting window width
unit (HU) is a quantitative value for describing attenuation and level. Window width describes the range of HU on CT
on CT images. Zero HU and −1000 HU are defined as image and window level is the median HU of window
attenuation of distilled water and air at standard pressure width. In the thorax, lung, mediastinum, and bone win-
and temperature. CT images can be appropriately dis- dow setting are usually used

acquisition and reconstruction should be used. mended. Expiratory CT is gradually regarded as

Nevertheless, it is impossible in most cases, par- an important part of COPD evaluation for the
ticularly for multicenter trials. There are several presence and distribution of air trapping on visual
important issues to be considered for the optimi- and quantitative assessment [10]. Expiratory CT
zation of CT protocol which include kilovoltage scan may be obtained at relatively low dose after
setting, tube currents, respiration level, image inspiratory CT acquisition. However, different
thickness, reconstruction kernel, and so on noise level between inspiratory and expiratory
(Table  8.1). Generally speaking, the volumetric CT is potential problem for the quantitative anal-
CT acquisition obtained at maximal inspiration ysis. Expiratory scan can be obtained at the end
with standardized breathing instruction is essen- of a tidal expiration, which corresponds to func-
tial for accurate COPD assessment using tional residual capacity, or after full expiration,
CT. Thin-­ section CT reconstructions (even or which is in residual volume status.
less than 1 mm in thickness) are required for
proper characterization and quantification of
emphysema and airway change in COPD. It has  iagnosis of Morphologic Change
been known that CT estimates of emphysema in COPD
severity increase as section thickness decreases
and that higher frequency (edge-enhancing,  mphysema: Definition of CT Finding,
sharper) image reconstruction kernel results in Subtype, Pathologic Correlation
higher CT measurements of emphysema than Pulmonary emphysema is defined as an abnormal
lower frequency resolution (smoothing) kernel permanent enlargement of the airspaces distal to
[4–6]. Low radiation dose CT scan allows the the terminal bronchioles, which results in lower
visual evaluation of emphysema, however, which CT attenuation values than those of normal
trade off relatively high image noise, resulting in healthy lung areas. Although pulmonary function
overestimation of emphysema extent. Relatively test (PFT) is useful for clinical severity assess-
higher dose CT is also necessary for the accurate ment of COPD, it does not represent the type and
measurement of airway dimensions. Recently, range of heterogeneous pathophysiologic abnor-
further reduction of CT dose can be possible with malities in COPD [11, 12]. In addition, it tends to
the use of novel technique of iterative reconstruc- be relatively insensitive to both early stages and
tion [7–9]. However, it is not recommended for small changes in COPD [13]. CT scan may be
the quantitative assessment of COPD CT because ideal for the detection and characterization of
the influence of changing reconstruction method COPD in that it allows for an in vivo analysis of
has not been fully studied. Dose modulation tech- morphologic characteristics and distribution of
nique to reduce radiation dose is not recom- emphysema. Visual assessment is also useful to
8  Imaging of COPD 89

Table 8.1  General principle of CT protocol for COPD evaluation

Inspiratory CT Expiratory CT (optional)
Scan type, mode Spiral (volumetric) Spiral (volumetric)
Rotation time (s) As short as possible, usually no greater As short as possible, usually no greater
than 0.5 s than 0.5 s
Detector configuration More than 16 channels × submillimeter More than 16 channels × submillimeter
collimation collimation
Pitch Usually smaller than 1.0 Usually smaller than 1.0
kVp 120 120
mA 40 mAs (low dose) up to 200 mAs 40 mAs (low dose) up to 200 mAs
(moderate dose) (moderate dose)
Dose modulation Not recommended Not recommended
Reconstruction I for visual assessment
Algorithm Sharp or high-frequency kernel Sharp or high-frequency kernel
Thickness (mm) Submillimeter (0.5–0.75 mm) Submillimeter (0.5–0.75 mm)
Interval (mm) 0.5 0.5
DFOV (cm) To cover the whole lung To cover the whole lung
Reconstruction II for QCT
Algorithm Neutral, smooth kernel Neutral, smooth kernel
Thickness (mm) Submillimeter (0.5–0.75 mm) Submillimeter (0.5–0.75 mm)
Interval (mm) 0.5 0.5
DFOV (cm) To cover the whole lung To cover the whole lung

subtype the emphysema into centrilobular, (Fig. 8.2). Paraseptal emphysema tends to be lim-
­panlobular, and paraseptal types. Centrilobular ited in extent and occurs most commonly along
emphysema is the most common morphologic the subpleural portion of the upper lung, often
subtype of pulmonary emphysema. Pathologically, coexisting with other types of emphysema and
it begins near the center of the secondary pulmo- fibrosis. Small focal lucencies, up to 10 mm in
nary lobule (the most fundamental structural size, can be seen on CT (Fig. 8.2). Bullae or
component of the lung containing parenchyma, blebs, termed interchangeably, are focal regions
airways, lymphatics, and vasculature) in the of emphysema with no discernible wall, usually
region of the proximal respiratory bronchiole. more than 1 cm in diameter at subpleural loca-
Parenchymal destruction starts in the center of tion. In some cases, they can be very large and
secondary pulmonary lobule results in the char- may result in pneumothorax in COPD patients
acteristic apposition of normal and emphysema- (Fig. 8.3).
tous lung (area of low attenuated destruction
surrounded by normal tissue). On CT, small  irway Change: Bronchus, Trachea
round low attenuated holes are evenly distributed Bronchial wall thickening, bronchial luminal
with ill-defined borders in early stage and these irregularity, and bronchiectasis are commonly
low attenuated areas become confluent and insep- seen in patients with COPD with mixed findings
arable with paucity of pulmonary vascularity in of emphysema. Bronchial wall thickening is
late stage (Fig. 8.2). Panlobular emphysema is regarded as a sign of chronic bronchitis, easily
characterized by permanent destruction of the identified in heavy cigarette smokers.
entire acinus distal to the respiratory bronchiole, Pathophysiologically, irreversible and progres-
and its pathogenesis relates to alpha-1 antitrypsin sive histologic changes in airways show diffuse
deficiency. Parenchymal destruction involving hyperplasia of mucous glands associated with
entire acinus, which is contrast to centrilobular hypersecretion and bronchial wall thickening.
subtype, affects the lower lobes more severely Traditionally, bronchial wall is regarded to be
90 S.M. Lee et al.

a b

c d

Fig. 8.2  Emphysema subtypes on CT images. (a) Normal pleura and septal lines. (d) Panlobular emphysema.
lung parenchyma. (b) Centrilobular emphysema. Scattered Secondary pulmonary lobules area completely replaced
small focal lucencies (parenchymal destruction) in upper with emphysema with showing uniform and relatively
lobes, measuring more or less than 1 cm in diameter. (c) homogeneous lucencies across parts of the secondary pul-
Paraseptal emphysema. Small focal lucencies, up to 1 cm monary lobules
in diameter, are located in subpleural ar8ea adjacent to the

a b

Fig. 8.3  Large bullae and pneumothorax in a COPD large subpleural bullae with collapsed central lung. (b)
patients. A 71-year-old male with COPD complaint of Left pneumothorax was noted with collapsed lung and
sudden dyspnea. (a) CT image at 3 month ago showed large bullae
8  Imaging of COPD 91

thick if the diameter ratio of inner to outer lumen diagnosis of small airway disease in COPD is
of bronchus is greater than 0.8. Nevertheless, the fundamental for early diagnosis of COPD. The
diagnosis of bronchial wall thickening of COPD small airways are referred to as airway lumen
with naked eyes on CT is subjective and still lim- less than 2 mm, which cannot be visualized
ited (Fig. 8.4). Moreover, it is virtually impossi- directly using even recently developed CT scan-
ble to differentiate bronchial wall thickening of ners. Thus, the finding of air or gas trapping,
COPD from acute bronchitis or asthma. which appears to decrease lung attenuation on
Bronchiectasis can be accompanied in COPD expiratory CT, can be used as indirect sign of
patients and may represent severe airflow obstruc- small airway dysfunction in COPD because this
tion [14, 15]. Trachea and main bronchus abnor- finding is thought to be caused by early collapse
malities can be visually defined on CT in COPD of small airway on expiration. Expiratory CT is
patients. Tracheobronchomalacia, saber-sheath increasingly regarded as an essential tool for the
deformity of trachea, and outpouching of trachea evaluation of air trapping as an obstructive pat-
and main bronchus can be seen in advanced tern of small airway disease in COPD patients.
COPD (Fig. 8.5). Tracheobronchomalacia is tra- In many cases, especially in emphysema domi-
ditionally defined as a more than 50% collapse of nant COPD patients, the detection of small air-
the trachea and main bronchus at end-expiratory way dysfunction may be hampered by the
CT. Saber-sheath deformity is seen as coronal presence of emphysema because the lung den-
narrowing and sagittal widening of the intratho- sity of emphysema area on expiration CT can be
racic tracheal diameter. Tracheal outpouching is low without small airway dysfunction. In such
defined as a focal herniation of mucosa through situation, side-by-­side comparison of inspira-
the tracheal wall. tory and expiratory CT images is necessary to
detect lack of normal increase of lung attenua-
Air Trapping tion and decrease of lung volume on expiration
It has been understood that small airway airflow (Fig. 8.6).
resistance is the major site of obstruction in
patients with COPD and precede the onset of  thers: Chest Wall, Diaphragm, Heart,
emphysematous destruction in both centrilobu- Pulmonary Vessel, and Bone
lar and paraseptal emphysema phenotypes of Morphologic changes secondary to pulmonary
COPD [16]. Therefore, early detection and hyperinflation in COPD include chest wall

a b

Fig. 8.4  Airway changes in COPD. (a) Example of bron- dilatation of bronchial lumen in both lower lungs is noted.
chial wall thickening. Typical and severe thickening of The inner luminal diameter of bronchus is greater than the
wall of entire segmental bronchi in both lower lobes is diameter of the accompanying pulmonary artery. There
noted. The diameter ratio of inner to outer lumen is are also loss of normal tapering with bronchial wall
smaller than 0.8. (b) Example of bronchiectasis. Marked thickening
92 S.M. Lee et al.

a b

c d

Fig. 8.5  Tracheobronchomalacia. Tracheobronchomalacia strate severe lumen narrowing of the trachea on expiration
is defined as abnormal collapse of airway lumen on expira- (b) and CT images on inspiration (c) and expiration (d) also
tion. This abnormal finding can be seen in COPD patients. depicted severe lumen narrowing of both bronchi (d). In
CT images on inspiration (a) and expiration (b) demon- addition, severe air trapping is seen in both lower lobes

d­ eformity, diaphragmatic change, changes in the in turn, may also reduce pulmonary function or
heart, and pulmonary vasculatures. Barrel chest even cause COPD exacerbations [17].
­deformity is the well-known chest wall deformity
of COPD and depression of diaphragm indicates
the flattening of the domes of the diaphragm due Severity Assessment of COPD
to hyperinflation of the lung in COPD (Fig. 8.7).
As the progression of COPD, the heart tends to  xtent of Emphysema: Visual
be more vertically oriented due to hyperinflation Assessment
of the lung. In later stage, right ventricular and Visual assessment of COPD is relatively simple,
atrial dilatation, dilatation of central pulmonary cheap, and independent from variation of CT
arteries, and acute tapering of distal pulmonary machine or reconstruction algorithms.
vessels can be seen as a finding of pulmonary Furthermore, comprehensive visual emphysema
hypertension. Osteoporosis is also one of the sys- assessment of CT in COPD allows assessment of
temic effects associated with COPD attributed by the pattern, subtype, regional location, and degree
inactivity, COPD-related systemic inflammation, of emphysema. It also has an advantage for
the use of systemic corticosteroids, and vitamin D detecting lots of accompanying pathologic
deficiency. Bone fracture related to ­osteoporosis, changes in the parenchyma as well as in the small
8  Imaging of COPD 93

a b

c d

Fig. 8.6  Air trapping assessment using inspiration- and defined as areas showing lack of normal increase of lung
expiration CT images. Air trapping should be assessed by attenuation and decrease of lung volume on expiration
side-by-side comparison of inspiration (a, b) and expira- image (c, d). The areas of air trapping are marked in
tion (c, d) CT images. Expiratory air trapping can be arrows

and large airways. Visual assessment allows for system: 0, 1–5%, 6–25%, 26–50%, 51–75%, and
the detection of early emphysema in asymptom- greater than 75% [19]. However, main limitation
atic smokers even before the development of air- of visual assessment has been the relatively low
flow limitation, which is essential for the inter-reader agreement [20, 21]. The inter-reader
diagnosis of COPD. It is also useful to follow the agreement was moderate for the presence or
progression of emphysema over time. For the absence of emphysema and for the presence of
optimal evaluation of emphysema with CT panlobular emphysema; fair for the presence of
images in COPD patients, thin-section, high-­ centrilobular and paraseptal subtypes [22]. In an
resolution CT images should be used at recom- effort to improve the inter-reader agreement,
mended window settings (usually with a window usage of standardized reference images has been
level of −700 and window width of 1000–1500). attempted with promising results [19].
On visual assessment, emphysema is classified as
centrilobular, panlobular, and paraspetal emphy-  xtent of Emphysema: Computer-­
sema (Fig. 8.2). The extent of emphysema has Based Quantification
been assessed by using visual scoring system For the objective and reproducible assessment of
[18]. Typically, the extent of emphysema in each emphysema, computer-based quantification
lobe can be assessed by using a six-point scale method, the so-called quantitative CT (QCT), has
94 S.M. Lee et al.

a b

c d

Fig. 8.7  Morphologic change of the diaphragm in COPD with COPD shows depressed and flattened shape of the
patients. Compared with CT images in a normal individ- diaphragm. This change is due to hyperinflation of the
ual (a, b), coronal CT images (c, d) in a 74-year-old male lung in COPD

been introduced. As briefly explained in “CT method is called as “density mask” and the
Basic Physics” section, emphysema area on CT threshold of −910 HU was initially applied [24].
with relatively increased air fraction in inspira- Recent studies using thin-section, multi-detector
tion lung is shown as area of decreased CT atten- CT scanners showed that the highest correlation
uation approaching air attenuation of −1000 HU, between QCT and histology is found when the
compared with normal lung CT attenuation threshold set at −960 or −970 HU [25]. However,
around −850 HU [23]. Accordingly, if a certain the lower the thresholds, the more sensitive the
threshold value is applied, the area of emphy- image noise; therefore, the threshold of −950 HU
sema with decreased CT attenuation can be is now most commonly used. When the correc-
objectively divided from normal lung area. This tion for lung volume changes influenced by
8  Imaging of COPD 95

degree of inspiration is applied, this quantitative  orrelation Between the Extent

method for emphysema is near perfectly repro- of Emphysema and Clinical Parameters
ducible method. The term of percent of emphy- Visual, subjective assessment of the emphysema
sema (emphysema index, EI or %LAA−950) stands using contiguous 10-mm thick CT started in 1986,
for the relative area of lung less than −950 HU there were significant correlations between CT
(Figs. 8.8 and 8.9). Another method is using the visual scores and macroscopic emphysema.
nth cutoff percentile in the attenuation distribu- However, even with the development of high-­
tion curve using the CT attenuation at a certain resolution CT, visual grading assessment is not
percentile along the frequency histogram of pul- really a quantitative measure but just grading the
monary parenchymal attenuation (density value degree of emphysema according to categories of
in HU under which n% of frequency histogram) emphysema severity. Recently, using pulmonary
(Fig.  8.8). This value is called as “percentile lobe-by-lobe visual assessment, severity of
index,” and it is reported that it has an advantage emphysema correlates quite well with physiologic
on longitudinal evaluation and less sensitive to parameters (FEV1 and FEV1/FVC) and GOLD
lung volume changes [26–28]. The first percen- stage [19]. The correlation coefficient ranges
tile value is much optimal for correlation with between 0.67 (for GOLD stage) and −0.74 (for
histology; however, it is known to be sensitive to FEV1/FVC) and notably the range of correlation
an artifact from image noise and truncation coefficients are similar to the correlations between
effect. Instead, the 15th percentile threshold is extent of emphysema on QCT and each physio-
commonly used [28, 29]. The last method is to logic parameter (0.62 for GOLD stage and −0.70
assess the mean of the whole lung attenuation for FEV1/FVC). However, inter-reader agreement
(mean lung density, MLD). Regional heterogene- regarding severity of emphysema on visual assess-
ity of emphysema can be assessed quantitatively ment tends to be variable, so QCT is preferred for
to assess the regional distribution of emphysema. assessing disease severity of emphysema.
Most available QCT methods can divide each Moreover, QCT measurements have shown to
lung into upper, mid, and lower zones of equal correlate better than visual CT assessment with
height or volume, and ratios between the extent macroscopic measurement of emphysema [21].
of emphysema in upper and lower lung can be
computed. Newer methods can also permit seg-  egional Heterogeneity of Emphysema
mentation of lobes to compute lobar volumes and Severity and distribution of emphysema differs in
extent of emphysema objectively (Fig. 8.9). lung regions such as core (inner) vs. rind (outer),
upper vs. lower, even among each lobe (Fig. 8.10).
Comparison Between  Visual There have been several reports regarding
Assessment and CT Quantification regional variation of emphysema. Basal distribu-
In the assessment of emphysema in COPD tion of emphysema is associated with greater
patients, although QCT measures correlate with impairment of FEV1 but less impairment of gas
the severity of visual CT assessment, the level of exchange (PaO2) and alveolar-arterial oxygen
correlation is only moderate [22]. Especially in gradient than the apical distribution of emphy-
less severe categories of emphysema, visual sema [31]. Emphysema areas on CT are more
assessment by radiologists tends to be usually often found in the inner segment of the lung than
underestimated for the extent of emphysema in the outer segment and the extent of emphy-
when compared to QCT measures, while in those sema in inner segment of the lower lung in QCT
with more severe emphysema, the radiologists is much more clearly correlated with airflow
tend to relatively overestimate the emphysema limitation than those in outer segment [32].
extent [30]. Therefore, visual assessment and In another report applying the slope of the EI
QCT measures should be used complementarily in the upper-lower, anterior-posterior, and
and performed independently for the assessments central-­peripheral direction in both side lung,
of severity of emphysema. the heterogeneity of emphysema distribution in
96 S.M. Lee et al.

Relative frequency (%)

–1000 –950 –900 –850 –800 –750 –700 –650 –600 –550 –500 –450 –400
15th percentile Attenuation (HU)


Cumulative frequency (%)







–1000 –950 –900 –850 –800 –750 –700 –650 –600 –550 –500 –450 –400

15th percentile Attenuation (HU)

Fig. 8.8  Quantitative CT measurements of emphysema tile index method uses a certain percentile point (i.e., low-
regarding LLA%, percentile index, and mean lung density est 15th percentile) and the HU value for that percentile is
(MLD). The concept of QCT measurement is illustrated in calculated. The last index is the mean of lung density
Fig. (a). Frequency distribution curves are plotted accord- (MLD). Left side shift of frequency curve in patient with
ing to the apparent X-ray attenuation values. The thresh- emphysema is demonstrated in Fig. (b). As a result of the
old of −400 HU is to define lung area. Threshold (LAA shift, LAA % increases and percentile index decreases in
%) technique uses a predefined threshold of HU (i.e., patients with emphysema. The area of emphysema can be
–950 HU) is chosen and the percentage of lung less than overlaid on the CT images to highlight the emphysema
this value can be calculated. Contrary to this, the percen- lesion (c, d)
8  Imaging of COPD 97

c d

Fig. 8.8 (continued)

Fig. 8.9  Lobar-specific quantification of emphysema. Example of software providing automatic segmentation of lobes
and quantitative assessment of extent of emphysema at the whole lung and lobar level

anterior-posterior and upper-lower direction was cant predictor for improvement of pulmonary
independent determinants of FEV1 and FEV1/ function after LVRS [34].
FVC and the lower and posterior regional domi-
nant emphysema is associated with a decrease in  irway Change: Visual Assessment
FEV1 and FEV1/FVC [33]. Regional assessment Although visual assessment of large airways is a
using QCT helps in selecting candidates for lung subjective process, the presence of airway wall
volume reduction surgery (LVRS) and provides thickening or dilation of large airways can repre-
rationale for the mechanisms of improvement sent bronchial inflammation with remodeling, and
after LVRS (Fig. 8.11). The extent of emphysema it also contributes to the symptomatic exacerba-
of the upper-rind region of the lungs is a signifi- tion in COPD patients. Bronchial wall thickening
98 S.M. Lee et al.

a b

Fig. 8.10  Regional heterogeneity. CT images with color (17.56 and 23.27%). This regional heterogeneity can also
mapping (a, b) show different predominance of emphy- be quantified by emphysema on each lobes (c)
sema in two patients with similar emphysema index

is commonly found in heavy cigarette smokers airways from thin-section volumetric datasets,
with a sign of chronic bronchitis (Fig. 8.4). CT scan seems to be well positioned to become
Bronchiectasis is also a common finding in COPD the method of choice to noninvasively measuring
associated with severe airflow obstruction and risk airway wall dimensions of luminal diameter and
for COPD exacerbation [14, 15, 35]. wall thickness to the level of segmental and sub-
segmental airways. The simple analysis of “full-­
 irway Change: Computer-Based
A width at half maximum” algorithm is commonly
Quantification used to evaluate the airways including absolute
With the current development of available soft- measures (bronchial luminal diameter or area,
ware permitting multiplanar reconstruction of bronchial wall thickness or area, and total
8  Imaging of COPD 99

a b

Fig. 8.11  Effect of lung volume reduction procedure. (a) severely affected by emphysema. (b) After the procedure,
CT image in a 61-year-old male showed severe and RUL was near totally collapsed on CT and pulmonary
regional heterogeneity of emphysema. Endobronchial L function of this patient much improved (FEV1,
was performed to collapse the RUL which was most 0.46 L —> 0.82; mMRC, Gr4 —> 1)

b­ ronchial area) and relative measures (bronchial airway disease [39]. Moderate correlations
wall area %, WA%: 100 × (wall area)/ (−0.56 < r < −0.62) between airway wall mea-
(lumen + wall area)) (Fig. 8.12). Variable soft- surements and airflow obstruction (FEV1 and
ware algorithms to define the boundaries of air- FEV1% predicted) have been reported and stron-
way wall have been proposed and tested. Another ger correlations were noted when only small air-
commonly used measure is the square root of ways were analyzed [40]. In recent report,
wall area of a hypothetical bronchus having inter- bronchial wall thickening as well as severity of
nal perimeter of 10 mm (Pi10) [36]. emphysema measured on QCT is associated with
exacerbation frequency, independently; bron-
 orrelation Between Airway Measures
C chial predominant and emphysema predominant
and Clinical Parameters subtypes of COPD can be defined [41].
Many studies showed that patients with the great-
est WA% had the lowest FEV1 expressed as a per-  ther Large Airway Changes in COPD
cent predicted [37]. The WA% has been Among the large airway changes in COPD,
considered as the most commonly employed Saber-sheath trachea is not an uncommon finding
metric for clinical investigation, and there are in COPD (Fig. 8.5). It defines as the ratio of the
modest correlations between airway WA% and sagittal to coronal diameter is greater than 2 and
lung physiologic impairment [10, 38]. Moreover, the extra-thoracic portion of the trachea is not
central airway remodeling apparent on CT may narrowed. By comparison with normal healthy
reflect the distal histopathologic remodeling of persons, COPD patients show that this tracheal
the small airways, so the greater the central morphologic change of the elongation of the sag-
airway wall thickening, the more small the
­ ittal diameter correlated with the severity of
100 S.M. Lee et al.

Attenuation (HU)

Half maximum

Airway Airway Airway
center lumen wall

b c

Fig. 8.12  Quantification of airway wall thickening using boundaries, the airway wall area can be highlighted with
computerized method principle for full-width-at-half- color overlay (b). This method is called as full-width-at-­
maximum (FWHM) method. In the attenuation profile half-maximum method and is most common method to
along an outward flowing ray from the luminal center-­ quantify wall thickness. The airway dimensions of the
point through the airway wall, the inner and outer airway whole airway trees can be assessed using automatic air-
wall boundaries are assumed halfway to the maximum on way segmentation, centerline extraction, followed by air-
the lumen side and halfway to the minimum on the paren- way quantification (c)
chymal side, respectively. (a) After the detection of

emphysema and QCT indices, reflecting airflow indirect sign to evaluate small airway remodel-
limitation and air trapping [42]. Furthermore, ing. However, accurate discrimination between
expiratory tracheal collapse in obese COPD emphysematous area and air trapping area is dif-
patients shows greater quality of life impairment ficult and challenging, and even normal healthy
and worse exercise performance than expected person can show minimal air tapping area in both
based on functional measures [43]. basal lobes. Recently, there have been introduced
several methods to evaluate the degree of patho-
 mall Airway Disease: Visual
S logic air trapping in COPD patients using QCT.
The small airways are referred to as airway lumen  mall Airway Disease: Computer-Based
less than 2 mm, those cannot be visualized using Quantification
current CT scanners. The presence of air trapping With the usage of expiratory CT images, quanti-
on expiratory CT scan can be identified as an tative assessment of air trapping is possible.
8  Imaging of COPD 101

However, there are severe inborn limitations correspondence of lung region between inspira-
when only expiratory CT is used because it is tion and expiration CT images can be assessed
impossible to separate trapped air area from air and both images can be co-registered. By direct
remaining in emphysematous spaces. Moreover, assessment of density difference between inspi-
air trapping phenomenon can also be seen in ration and expiration CT, the density change map
healthy smokers and healthy individuals with can be generated and the areas with decreased
normal lung physiology. However, even with density change can be defined as the area of air
these drawbacks, many studies have evaluated trapping. By using this method, in addition to the
the presence of air trapping in COPD by assess- global assessment, regional assessment of air
ing the area fraction of the lung lesion having CT trapping is possible (Fig. 8.13) [49, 50].
values lower than the threshold of −856 or
−850 HU (LAAexp856 or LAAexp850) in expiration  ther Components of COPD:
[44]. With this simple method, they reported that Correlation with Clinical Parameters
high correlations were noted between LAAexp850
and FEV1/FVC and FEV1 percent predicted. As Texture-Based Emphysema Assessment
an effort to overcome this single threshold With an effort to discriminate various obstructive
method of combining air trapping and emphy- lung diseases from normal lung, more sophisti-
sema quantification into single measure, quanti- cated automatic classification system based on
fying air trapping outside the emphysematous the texture and shape features of CT images has
area is possible through the density-based quanti- been introduced. Using this method, further dif-
fication method [45, 46]. With excluding emphy- ferentiation of lung areas, for example, into nor-
sema portion of all voxels with attenuation lower mal lung, small airway disease, centrilobular
than −950 HU from inspiration and expiration emphysema and panlobular emphysema, is pos-
scans and calculating the relative volumes for sible. This quantification method showed compa-
whole lung with attenuation value less than rable correlation with the pulmonary function
−860 HU on each inspiratory CT (inspiratory test results when compared with conventional
relative volume< −860 HU) and expiratory CT (expi- density-based quantification [51, 52]. This
ratory relative volume< −860 HU), the relative vol- method can also be used for the assessment of
ume change between −860 and −950 HU can be combined pulmonary fibrosis.
calculated as follows: Relative volume change <
−860 HU (%) = expiratory relative volume < −860 HU— Pulmonary Vascular Change
inspiratory relative volume< −860 HU. Results from Pulmonary vascular change is also one of the
this method show that air trapping correlates with characteristic features of COPD and the extent of
lung physiologic parameters significantly emphysema, rather than airway obstruction, is
(r = 0.50–0.80). Other methods for measuring air responsible for pulmonary endothelial dysfunc-
trapping have been addressed as an index of air tion in COPD [53]. After volumetric CT scans of
trapping including the ratio of inspiratory to expi- the lung, pulmonary vasculature was automati-
ratory lung volume (E/I-ratioLV) and the expira- cally segmented from the parenchyma using soft-
tory to inspiratory ratio of mean lung density ware [54, 55] (Fig. 8.14). With the usage of QCT
(E/I-ratioMLD) [47, 48]. E/I-ratioMLD correlates measuring the cross sectional area (CSA) of
with clinical parameters of COPD such as BODE-­ small pulmonary vessels (sub-subsegmental
index (0.48 < r < 0.68) and E/I-ratioLV shows level, CSA less than 5 mm2), the total CSA of
almost perfect correlation with E/I-ratioMLD small pulmonary vessels in COPD shows strong
(r = 0.95, p < 0.001). All of these values have a (negative) correlation with the extent of emphy-
limitation that it can’t represent regional distribu- sema (%LAA−950), whereas weak correlation
tion of air trapping. Recently, new approach of with airflow obstruction [56]. The anatomic
air trapping assessment has been proposed [49, extent of emphysema instead of airway obstruc-
50]. By using software techniques, anatomical tion is responsible for impairment of pulmonary
102 S.M. Lee et al.

a b

c d

Fig. 8.13  Subtraction image from co-registered inspira- comparing CT attenuation between inspiration CT and
tory and expiratory images. Image maps (c, d) derived registered expiration CT, area of air trapping with little
from co-registered inspiratory (a) and expiratory (b) change in CT attenuation can be extracted and visualized
images depict changes in lung attenuation from inspira- in color overlay (c). This process is applied in the whole
tion to expiration. Using image registration technique, the lung and coronal distribution of air trapping can be visual-
expiration CT image is deformed to match with the cor- ized (d)
responding anatomical area on inspiration CT (c). By

a b

Fig. 8.14  Vascular subbranches of lung. (a) Extraction vessels in CT and (b) Vascular tree reconstruction; random
coloring; each color represents one vascular branch and mediastinal region is cropped
8  Imaging of COPD 103

vascular structure. Moreover, the percentage of mass is a better predictor of mortality than BMI
the total CSA for the lung area is significantly in patients with COPD [60, 61]. Thoracic respira-
higher in airway dominant phenotype than tory muscles are unique and crucial for alveolar
emphysema dominant phenotype. ventilation and weakness of respiratory muscle
results in dyspnea and respiratory failure associ-
Osteoporosis ated with mortality in COPD patients. It is
Osteoporosis should be considered as one of the reported that intercostal mass and intercostal
important pathology of COPD, because it may attenuation measured by QCT are significantly
cause vertebral compression fractures, which can correlated with FEV1 and extent of emphysema
also deteriorate FEV1 and decrease in vital capac- of QCT measurement [62]. A decrease in tho-
ity. Actually, the loss of vertebral bone mineral racic muscle mass with increasing intercostal fat
density on CT is closely related to the severity of is associated with worsening of COPD (Fig. 8.16).
emphysema showing many risk factors of low Hyperinflation in COPD makes diaphragm to be
BMI, decreased activity, systemic inflammation, flatter and shorter. As the progression of COPD,
and use of corticosteroids [17, 57, 58] (Fig. 8.15). breathing becomes gradually more dependent on
There has been reported that the decrease in tho- the thoracic intercostal muscles than diaphragm
racic vertebral bone mineral density is greater in (Fig. 8.7).
patients with a history of exacerbations than in
those without a history of exacerbations. Indeed, Atherosclerosis
osteoporosis progression should be checked in In COPD patients, reduced FEV1 has known to
COPD patients, especially in those with a history be an increased risk factor for cardiovascular dis-
of frequent exacerbations [59]. eases and mortality, independent of smoking [63,
64]. In other words, systemic inflammation in
Chest Wall and Diaphragm COPD patients may accelerate the rates of car-
COPD is also characterized by progressive diovascular disease, and this degree or status of
impairment of respiratory function and dysfunc- atherosclerosis may be associated with impaired
tion of respiratory muscle. There have been lung function and emphysema in COPD patients.
reports that the depletion of peripheral muscle There has been an attempt to demonstrate the

a b

iVol : 0
mean : 206.7 iVol : 0
max : 360 mean : 93.0
min : 31 max : 248
stddev : 58.9 min : -97
area : 126.2mm^2 stddev : 57.2
areaOver0 : 126.2mm^2 area : 136.7mm^2
areaOver0 : 129.5mm^2

Fig. 8.15  Thoracic bone density on CT image. Vertebral racic vertebra. (b) CT image of a 79-year-old male with
bone mineral density on CT is closely related to the sever- 52.2% of emphysema index shows 93.0 HU on thoracic
ity of emphysema. (a) CT image of a 59-year-old male vertebra
with 6.2% of emphysema index shows 206.7 HU on tho-
104 S.M. Lee et al.

a b

c d

Fig. 8.16  Thoracic muscle mass on CT image. Thoracic 7.4% of emphysema index shows more prominent tho-
muscle mass and intercostal fat are associated with sever- racic muscle mass than CT images of a 62-year-old male
ity of COPD. (a, b) CT images of a 75-year-old male with with 54.8% of emphysema index

association of the total amount of calcification in its relative contributions vary in each COPD
coronary artery, thoracic aorta, mitral and aortic patient. Previous studies already have suggested
annuli, and the extent of emphysema on QCT and that different spirometric response patterns to
lung physiology [65]. Calcium score as a mea- bronchodilator exist in patients with obstructive
surement for degree of atherosclerosis shows lung disease showing improvement in expiratory
weak but significant correlation with volume flow (FEV1) or lung volume (FVC) [66, 67].
fraction of emphysema on QCT, FEV1/FVC, and Therefore, QCT measurements can be used as a
diffusion capacity, independent of age, BMI, and longitudinal evaluation of treatment monitoring
smoking amount. The degree of atherosclerosis is based on the fact of a significant correlation
associated with impaired lung function and the between QCT measurement indices and lung
extent of emphysema. physiologic indices. Notably, in the assessment
of different spirometric response patterns to
bronchodilator treatment, the extent of emphy-
I maging Studies of Treatment sema in QCT measurement shows a significant
Monitoring and Disease Progression negative correlation with postbronchodilator
FEV1 change and the E/I-ratioMLD also shows a
 redicting Tool in Treatment Outcome
P significant positive correlation with postbron-
COPD is a heterogeneous condition featuring chodilator FVC change [68]. In case of lung vol-
both parenchymal destruction (emphysema) and ume reduction therapy in regions with severe
small airway disease (obstructive bronchiolitis); emphysema, QCT can be used as a predictor of
8  Imaging of COPD 105

improvement of lung function after surgical or This method allows us to distinguish the relative
bronchoscopic approaches [34]. contributions of functional small airway disease
component and emphysema in COPD in the
Disease Progression course of disease progression [49].
There have been several efforts to evaluate the
progression of emphysema using QCT measure-
ment and QCT is useful in demonstrating the Quality Control and Standardization
change in extent of emphysema directly
(Fig. 8.17). However, the main drawback of QCT Emphysema Quantification
measurement is a variation of inspiration level on There are several sources of variation in quantifi-
each CT scan. Studies showed that the change in cation of emphysema in COPD including scan-
the 15th percentile CT density after the correc- ner, software, and patient factors. Thinner slice
tion of lung volume difference was found to be thickness and the lower CT dose setting result in
more sensitive as an index of progression com- overestimation of emphysema extent on QCT
pared with measures of physiology or healthy due to increasing image noise [4]. Inter- and
status [69]. Recently, new single unified approach intra-scanner variation due to calibration error
using a voxel-wise imaging analysis can be used and beam hardening effects should also be con-
in diagnosing disease extent and phenotype of sidered, and there have been attempt to demon-
COPD, detailed spatial distribution and location. strate that density correction of volumetric CT

a b

c d

Fig. 8.17  Follow-up emphysema quantification (LAA, coding image (b). Three years hence, EI has increased and
%) in same subject. Initial emphysema index (EI, %) was measured to 56.2% on CT image (c) and matched color
measured to 41.8% on CT image (a) and matched color coding image (d)
106 S.M. Lee et al.

data based on air (reference value: −1000 HU in Magnetic Resonance Imaging (MRI)
tracheal air or outside patient air) will improve
the correlation between emphysema quantifica- The MRI can obtain morphologic and functional
tion and pulmonary function test [70, 71]. This imaging without ionizing radiation. However,
correction method may be useful to decrease the there are some reasons of difficulty in imaging the
variation of measured results when multiscanners lung with MRI. The lung consists of the low den-
are involved. Second, a smooth reconstruction sity tissue, so it contains a relatively small number
algorithm is usually recommended for the of protons which generate signal in MRI. In the
emphysema quantification using QCT because a lung, these are fast decay of signal due to suscepti-
strong or overenhancing algorithm can result in bility artifacts from countless air-­tissue interfaces.
overestimation of emphysema [6]. Regarding The fast imaging, triggering, and gating techniques
patient factors, variation in lung volume, which is are needed due to respiratory, vascular, and cardiac
influenced by degree of inspiration, can be a motions. The major advantage of MRI is combina-
major source of variation in clinical practice. tion of morphological and functional lung imag-
Measurements of emphysema can be different at ing, such as perfusion, ventilation, blood flow, gas
varying inspiration levels. However, quantitative exchange, and respiratory motion, with high spa-
measurement of differences in emphysema would tial and temporal resolution [75].
not be significant when scans are obtained above
90% of vital capacity [72]. Current smoking sta-
tus or coexisting air trapping or parenchymal Morphologic Evaluation of COPD
fibrosis can also alter the quantitative emphy-
sema measurements. Measured extent of emphy- Simply speaking, there are two different types in
sema in current smokers appears to be lower than COPD. The airway type relates to chronic bron-
those in former smokers, probably due to chitis and airflow obstruction. The emphysema
increased attenuation induced by s­ moking-­related type shows the parenchymal destruction with
infiltration of inflammatory cells in the lungs in severe airflow obstruction and distal airspace
current smokers [73, 74]. Therefore, for accurate enlargement. The morphologic evaluation in
and precise quantitative assessment of emphy- COPD using MRI is always compared to
sema extent, it is important to consider and con- CT. MRI is technically more challenging due to
trol all the factors discussed above. hyperinflation and loss of lung tissue [76]. So,
there are only a few publications on morphologic
Airway Quantification evaluation of COPD using MRI.
In quantitative airway measurements, consider-
ing factors about the source of variation are simi- Imaging Technique
lar to emphysema measurements. Because airway Many MRI sequences can be used for visualiza-
is small, it is easily influenced by partial volume tion of the lung: balanced steady-state free pre-
averaging and reconstruction algorithm than in cession (bSSFP), volumetric interpolated
quantitative measurement of emphysema. breath-hold examination (VIBE), half-Fourier-­
Appropriate radiation dose to overcome image acquired single-shot turbo spin-echo (HASTE),
noise during reconstruction and submillimeter and ultra-short echo time (UTE) [77–79]. Three-­
section will reduce the variation in airway mea- dimensional (3D) T1-weighted gradient-echo
surements. In addition, there are varieties of sug- sequences are used for the assessment of medias-
gested software algorithms to measure airway tinum and lung parenchymal tissues. T2-weighted
dimension, resulting in different measurement fast spin-echo with half-Fourier acquisition
values. Accordingly, usage of same software sequence can visualize bronchial wall thickening
method is essential in multicenter trial or for fol- and mucus plugging. Respiratory, vascular, and
lowing up the patients. cardiac motions can be overcome by using fast
8  Imaging of COPD 107

imaging, gating, and triggering techniques. Half-­ airway inflammation can be visualized by high
Fourier acquisition or ultra-short echo times are signal on T2-weighted and contrast-­enhanced
recommended. T1-weighted sequences [90]. In contrast,
mucus plugging on peripheral bronchi shows
Emphysema high signal intensity of fluid content on
It is a major role of T1- and T2-weighted images T2-weighted sequence without contrast enhance-
to differentiate inflammation from muscular ment on T1-weighted sequence on MRI.
hypertrophy, edema, and mucus plugging in
bronchial wall [80]. Emphysematous change of
lung cannot be easily diagnosed by a loss of sig- Perfusion MRI
nal. However, hyperinflation can be easily
detected by increased lung volume and reduced Using perfusion MRI, perfusion information of
blood volume. There is one study about the lung can be acquired without ionizing radia-
change of signal intensity of lung parenchyma tion. One of the advantages using perfusion
between inspiration and expiration MRIs, which MRI in COPD is combination of perfusion and
is correlated with FEV1 [81]. The MR signal can morphologic information about parenchymal
be improved and emphysema can be quantified destruction and cause of perfusion changes.
by using the UTE pulse sequences [82]. Using Several imaging techniques have been
fast radiofrequency (RF) excitation pulses, com- introduced.
pressed sensing and parallel imaging in UTE
pulse sequence, MR signal decay, and motion Imaging Techniques
artifacts can be minimized [83]. UTE pulse For assessment of pulmonary vasculature and
sequences improve contrast-to-noise ratio, perfusion, both of non-contrast-enhanced and
signal-­to-noise ratio, and signal intensity with contrast-enhanced sequences are available. Non-­
strong relationship between signal intensity and contrast-­enhanced perfusion MRI can be acquired
tissue density. Using UTE pulse sequence, pul- using arterial spin labeling technique, which is to
monary emphysema [84, 85], lobar fissures and mark a specific part of spins magnetically using
airways [86], inflammation and peribronchial radiofrequency (RF) excitation [91]. Using the
abnormalities [87] can be estimated. electrocardiogram (ECG) gating technique, sig-
nal differences between systolic phase and dia-
Airway stolic phase can make perfusion images of the
There are several factors, such as bronchial lung without contrast injection [92]. However,
level, diameter, wall thickness, and signal from one of the limitations of ECG-gated perfusion
bronchus, to detect bronchiectasis [88]. MRI MRI is that the image subtraction process is sen-
usually visualizes central and peripheral bron- sitive to misregistration due to bulk respiratory
chiectasis and central bronchi, whereas poorly motion [91]. Contrast-enhanced 2D and 3D MRI,
visualizes normal peripheral bronchi. Using which is based on dynamic acquisition of lung
3D volume interpolated gradient-echo tissue during contrast injection, can assess lung
sequence (VIBE) with high spatial resolution, perfusion and quantify pulmonary perfusion [93,
the airway can be visualized [89]. T2-weighted 94]. The advantage of contrast-enhanced perfu-
sequences can visualize inflammation, mucus, sion MRI is high signal-to-noise ratio [95]. For
edema, and fluid collections. Active inflamma- evaluation of whole lung perfusion during peak
tion can be represented by increased fluid, enhancement period, 3D technique should be
which shows high signal of the bronchial wall needed. For improvement of spatial resolution
on T2-weighted sequences. Inflammatory and reduce of the total acquisition time, k-space
activity has relation with contrast enhancement sampling techniques such as parallel imaging
of thickened bronchial wall on contrast-­ techniques or echo sharing techniques can be
enhanced T1-weighted sequence. Therefore, used [96, 97].
108 S.M. Lee et al.

Quantification (PBF, mL/100 mL lung tissue/min), pulmonary

Using MR perfusion technique, pulmonary blood blood volume (PBV, mL/100 mL lung tissue), and
flow can be assessed quantitatively [98, 99]. The mean transit time (MTT, s0) is as follows:
indicator-dilution theory using the maximum of sig-
nal intensity and the temporal course of the signal MTT =
change is base of quantification of pulmonary per- PBF
fusion. If linear relation can be assumed between Normalizing the area under the tissue
the concentration of contrast agent and the signal, concentration-­time curve to the integral of the arte-
concentration-time curves can be made by conver- rial input function can calculate PBV. Figure 8.18
sion of signal-time curves. The relationship among shows quantification images of lung perfusion using
the perfusion parameters—pulmonary blood flow perfusion MRI technique.

a b

50 50

100 100

150 150

200 200

250 250
50 100 150 200 250 50 100 150 200 250

52.5888 122.993 193.397 263.801 334.20 6.24252 12.8351 19.4277 26.0203 32.612





50 100 150 200 250

1.80897 3.61794 5.42692 7.23589 9.0448

Fig. 8.18  Lung perfusion quantification image. Using perfusion MRI technique, PBF map (a), a PBV map (b), and an
MTT map (c) are generated
8  Imaging of COPD 109

 OPD Studies Using Perfusion MRI

C HASTE sequence and single-shot rapid acquisi-
In COPD, chronic inflammation is thought to tion with relaxation enhancement (RARE) or
lead to intimal wall thickening and smooth HASTE sequences, can make oxygen-enhanced
muscle hypertrophy of pulmonary arteries.
­ MRI [111]. Respiratory gating techniques are
Hyperinflation and air trapping can make hypoxic preferred because pulmonary physiology and
vasoconstriction. Perfusion MRI shows high physiopathology can be affected by breath hold
accuracy in detecting perfusion abnormalities in despite decreased misregistration. Oxygen-­
patients with emphysema [100, 101]. The perfu- enhanced MRI is used to assess ventilation
sion abnormalities in COPD usually show a low abnormalities in pulmonary emphysema [112].
degree of inhomogeneous contrast enhancement, Regional changes in ventilation on oxygen-­
especially in the area of severe emphysema [102] enhanced MRI reflect the regional lung function
and decreased peak signal intensity. In COPD [113]. Figure 8.19 shows ventilation of patients
patients with severe emphysema, visual assess- with severe COPD using dynamic oxygen-­
ment of perfusion using 3D perfusion MRI shows enhanced MRI. FEV1 and DLco well correlate
high agreement with parenchymal destruction with slope of oxygen-enhancement time-course
[103]. With quantitative analysis, decreased per- curve and degree of oxygen-enhancement,
fusion parameters on MRI correlate with worsen- respectively. Dynamic oxygen-enhanced MRI
ing of FEV1/FVC and increased emphysema reflects DLco and provides diffusing capacity
index on CT [104]. Quantitative perfusion MRI maps [112].
in COPD shows decreased value and heteroge-
neous change in mean pulmonary blood flow  yperpolarized Noble Gas MRI
(PBF), pulmonary blood volume (PBV), and Using hyperpolarized noble gas MRI with 3He or
mean transit time (MTT) than those in normal 129
Xe gas, ventilation MRI can be acquired [110,
volunteer [105]. 114, 115]. These techniques visualize the 3He or
Xe gas in airway and airspaces, so it can be
used for regional mapping of airflow and assess-
Ventilation MRI ment of diffusion in airspace [116, 117]. For
these techniques, specialized laser equipment
For assessment of lung ventilation, several meth- and specialized RF transmitter and receiver coils
ods of MRI, such as oxygen-enhanced MRI and are mandatory. Four different techniques are used
hyperpolarized noble gas MRI, are developed. for hyperpolarized 3He MRI [118]. Static ventila-
Repeated or time-resolved measurements of lung tion imaging generally uses 2D or 3D fast low-­
dynamics can be obtained using ventilation MRI. angle single-shot or bSSFP sequences [119].
Diffusion imaging uses gradient-echo pulse
Oxygen-Enhanced MRI sequence with bipolar diffusion-sensitizing gra-
Oxygen-enhanced MRI can visualize lung venti- dient waveform between the excitation RF pulse
lation. Oxygen couples to hemoglobin and is and data acquisition [120]. Dynamic ventilation
present as dissolved oxygen in blood during oxy- imaging uses the ultrafast pulse sequences such
gen exchange between capillary beds and alveoli as interleaved spiral pulse sequence with good
[106]. Paramagnetic property of deoxyhemoglo- balance between spatial and temporal resolution
bin makes little T1 shortening effect with T2* [121]. Oxygen partial pressure imaging uses the
shortening effect [107, 108]. Dissolved oxygen paramagnetic effect of oxygen on polarization of
makes shortening of T1 relaxation time of blood 3
He [122]. Single-acquisition and single breath-­
in pulmonary vein due to paramagnetic property hold technique improves temporal resolution and
of oxygen [107]. This shortening leads to reduces error due to second breath hold. MRI
increased signal intensity on oxygen-enhanced with this technique can directly measure the
MRI [109, 110]. Several sequences, such as cen- regional ventilation and perfusion distribution
trically reordered phase-encoding scheme on [123]. Because 129Xe is naturally richer than 3He
110 S.M. Lee et al.

a b

c d

Fig. 8.19  Oxygen-enhanced MRI in COPD patients. (a, smoker with severe COPD (c: thin-section coronal MPR
b) A 75-year-old female smoker with mild COPD (a: thin-­ CT image demonstrates panlobular emphysema in bilat-
section coronal multiplanar reformatted (MPR) CT image eral upper and middle lung field. d: RER map from oxy-
shows centrilobular emphysema in the left upper lung gen O2-enhanced MRI demonstrates heterogeneously
field, b: relative enhancement (RER) map from oxygen decreased oxygen-enhancement in the both lung, espe-
O2-enhanced MRI demonstrates heterogeneously cially upper lung fields.) Image courtesy of Yoshiharu
decreased oxygen-enhancement in the both lung, espe- Ohno, Kobe University Graduate School of Medicine
cially left upper lung filed.) (c, d) a 56-year-old female

and 129Xe MRI shows comparable quality to 3He from 3He MRI due to difference of gas distribu-
MRI with recent advances in polarization and tion. In stage III COPD, ventilation defect vol-
imaging methods; many methods of 3He are ume (VDV) was sensitive to minimal changes
translated for 129Xe. However, 129Xe MRI differs during short-term follow-up [114]. Percentage of
8  Imaging of COPD 111

ventilation volume was significantly different respiratory cycle can be assessed. In emphysema
among three groups (healthy volunteers, healthy patients, motion of the diaphragm and chest wall
asymptomatic smokers, and COPD patients). is reduced, irregular or asynchronous [133].
COPD patients are separated from healthy sub- Emphysema in lower lung shows significant cor-
jects by apparent diffusion coefficient (ADC) relation with diaphragmatic flattening, abnormal
map [124]. Using diffusion-weighted hyperpo- chest wall motion, and severe airflow limitation
larized 129Xe MRI, ADC was significantly corre- [134]. Although both normal and paradoxical
lated with PFTs (FEV1, FEV1/FVC, and DLco) diaphragmatic motion is restricted by severe
[115]. Ventilation defect percentages in 3He MRI hyperinflation, the paradoxical diaphragmatic
show significant correlations with ventilation motion shows significant correlation with hyper-
defect percentages in 129Xe MRI and ventilation inflation [135].
defect percentages show strong correlations with
FEV1 [125]. ADC can measure airspace size sen-
sitively. In COPD, airspace dimensions are Other Imaging for COPD
increased compared to non-smokers [126]. ADCs
in emphysema show regional variations and sig- Dual-Energy CT
nificantly larger than those of healthy volunteers,
which is homogeneous [127]. I ntroduction of Dual-Energy CT
Dual-energy CT refers to CT that uses two differ-
ent energy spectra (usually 80 and 140 kVp).
 ourier Decomposition MRI
F With the knowledge about the X-ray attenuation
for Combined V-Q Imaging change of a particular substance at two different
X-ray energies, the material differentiation and
Non-contrast-enhanced ventilation and perfusion elemental decomposition of tissues are possible
MRI, known as Fourier decomposition MRI, uses in dual-energy CT [136, 137]. Therefore, with
a short echo dynamic SSFP acquisition with sub- single contrast CT scanning at two different ener-
sequent compensation for respiratory motion by gies, the CT images used for structural evaluation
using non-rigid image registration [128, 129]. are created by combination of the CT images
Peaks at the respiratory and cardiac frequencies from the low- and high-energy CT data, and the
can be identified by spectral analysis of the image material-specific images such as iodine map or
time series. Deformation of lung parenchyma and xenon map for functional evaluation are created
pulmonary blood flow leads to regional proton by the material-decomposition algorithms in the
density change, which is related to amplitude of postproccessing of dual-energy datasets
these peaks [130]. With image post-processing, (Fig.  8.20). Three different designs of dual-­
ventilation- and perfusion-weighted maps are energy CT for the acquisition of dual-energy data
generated for regional assessment of lung func- have been proposed. Firstly, dual-source CT sys-
tion from a single acquisition series [131]. tem has two separate X-ray tubes and two corre-
sponding detectors, which are placed with an
angular off-set of 90° on the rotating gantry. Each
Dynamic Respiration MRI X-ray tube can be operated at different kilovolt-
age and milliamperage settings. Secondly, in
Diaphragmatic geometry is affected by hyperin- rapid kVp switching, the single X-ray source is
flation of the lung. Dynamic respiration MRI used and X-ray tube electronically switches the
with fast-acquisition technique can visualize tube voltage between higher energy and lower
complex interaction between chest wall and dia- energy in about 0.5 ms. Lastly, the energy-­
phragmatic motion with high spatial and tempo- sensitive sandwich detector is now commercially
ral resolution [132]. Using dynamic respiration available. This system uses a layered detector
MRI, the change in lung volume during the and single X-ray tube with the polychromatic
112 S.M. Lee et al.

A-tube: 140 kVp, 50 eff.mAs B-tube: 80 kVp, 210 eff.mAs

Two tubes run simultaneously

Single scanning after contrast Average Post-processing

Average image: Fusion image Iodine image:

Conventional CT image Parenchymal perfusion

Fig. 8.20  Iodine perfusion imaging using dual-source tion of the 140 and 80 kVp datasets. Iodine image is
CT. CT image data is generated in dual-energy acquisition obtained with extraction of iodine component with
mode of dual-source CT. Axial CT image is obtained at material-­decomposition theory in post-processing. Fusion
140 kVp and 50 mAs from A-tube and at 80 kVp and image with conventional CT image and iodine image is
210 mAs from B-tube. And conventional CT image generated for the evaluation of lung parenchymal
(approximate 120 kVp image) is generated from combina- perfusion

spectrum. The layered detector comprises two as alveolar ventilation changes or parenchymal
layers: a thin top scintillator that absorbs low perfusion changes.
energy photons and a bottom scintillator that
absorbs the higher mean energy photons.  erfusion Dual-Energy CT
For COPD patients, which is characterized by Dual-energy CT-derived iodine map represents
airway obstruction and emphysematous alveolar the iodine content of the capillary bed, i.e., pul-
destruction, pulmonary vascular changes are also monary blood volume at the time of CT scanning
produced, which are characterized by hypoxic rather than pulmonary blood flow. However, it
vasoconstriction, numeric reduction, and endo- has been demonstrated that the similarity of the
thelial dysfunction of the small pulmonary arter- dynamic CT-derived pulmonary blood flow and
ies [138–140]. These characteristic anatomic dual-energy CT-derived pulmonary blood vol-
changes influence and impair alveolar gas ume [143]. Therefore, pulmonary blood volume,
exchange, and the uneven distribution of alveolar assessed with dual-energy CT, can be used for the
ventilation and pulmonary blood flow (V/Q mis- evaluation of lung perfusion as a surrogate for
match) is the most important cause of arterial dynamic CT-derived pulmonary blood flow with
hypoxemia in the COPD patients [141, 142]. simpler protocol while maintaining quantitative
Therefore, evaluating COPD patients should similarity (Fig. 8.21).
focus on not only the extent and severity of ana- In COPD, the regional lung perfusion
tomic destruction of the lung parenchyma but impairment occurs due to the hypoxic vasocon-
also the functional changes and impairment such striction of the area with decreased ventilation
8  Imaging of COPD 113

a b

c d

e f

Fig. 8.21  Iodine perfusion image with dual-energy CT in be assessed on high-resolution conventional CT image (a,
COPD patient. (a–c) Dual-energy CT in COPD patient d), and anatomically matched parenchymal perfusion
with mild emphysema. (d–f) Dual-energy CT in COPD information can be evaluated on perfusion map (b, c, e, f).
patient with severe emphysema. From iodine perfusion In the area of confluent emphysema of both lower lobes,
CT using dual-energy CT, (a, d) weighted conventional parenchymal perfusion is decreased (displayed with blue
CT image and (b, c, e, f) iodine perfusion map are gener- color)
ated. Lung parenchymal destruction in COPD patient can
114 S.M. Lee et al.

and reduction of the pulmonary capillary by the tigraphy [149]. Park et al. used dual-energy CT
chronic inflammation of pulmonary artery with lung perfusion imaging for target lobe selec-
[144]. Moreover, alveolar surface destruction is tion of bronchoscopic lung volume reduction by
­accompanied by the reduction of the pulmonary endobronchial valves. In that study, the target
capillary bed. Thus, considerable attention has lobe was selected, if it was most hyperinflated
been paid to the evaluation of lung perfusion and least perfused, and if it had no collateral ven-
alterations in COPD patients because the sever- tilation with other lobes on perfusion image with
ity of parenchymal destruction and the altera- dual-energy CT [150].
tion of lung perfusion determine the functional
effect of emphysematous changes. Lung paren-  entilation Dual-Energy CT
chymal perfusion has been assessed by perfu- In clinical practice, the evaluation of COPD
sion scintigraphy, single-photon emission CT severity is based on the result of pulmonary func-
and MR. However, the distribution of perfusion tion test; however, pulmonary function test pro-
impairment does not match with the area of vides global status of lung function and does not
parenchymal destruction [145]. Dynamic multi-­ show the regional distribution of functional
detector CT perfusion imaging also can provide abnormality. For evaluation of regional lung
the regional perfusion. It has been demonstrated parenchymal ventilation, radionuclide scintigra-
that smokers with subtle CT findings of centri- phy or MR is used, but it is limited by its low
lobular emphysema and normal findings at spi- spatial resolution. CT also can be used to depict
rometry has increased regional heterogeneity of lung ventilation with inhalation of xenon gas
lung perfusion compared with never smoked [151, 152]. Xenon is a radio-opaque gas and
subjects and smokers with normal CT image xenon gas concentration in alveolar space can be
[146]. However, dynamic multi-detector CT measured based on the attenuation changes on
perfusion imaging necessitates a central high- CT image (Fig. 8.22). However, because of vari-
pressure bolus of contrast material and scan- ability in baseline lung attenuation between
ning a limited axial extent of the lung during a images due to misregistration artifacts and differ-
cardiac-gated scan. ent respiration levels, the accurate measurement
Pansini et al. have demonstrated that regional of lung ventilation function is limited, triggering
alteration of lung perfusion can be assessed by great attention in the simultaneous structural and
dual-energy CT, matching parenchymal destruc- functional evaluation with single CT scanning
tion in 47 smokers with predominant emphysema accessible to dual-energy CT. Two stable gases,
[147]. Moreover, Lee et al. have shown that the xenon and krypton, with high atomic numbers (54
contrast-enhanced dual-energy CT can be used for xenon and 36 for krypton) are eligible for ven-
for the quantification of emphysema and regional tilation imaging with dual-energy CT. For xenon
perfusion evaluation by using the virtual non-­ ventilation imaging with dual-energy CT, the
contrast images and iodine map, simultaneously patient usually inhales 30% stable xenon (a mix-
[148]. Assessing the distribution of pulmonary ture of 30% xenon and 70% oxygen) for 1 min to
emphysema and anatomically matched paren- 1 min 30 s with use of a xenon gas inhalation sys-
chymal perfusion information is particularly tem (Zetron V; Anzai Medical, Tokyo, Japan).
applicable in the patient and target lobe selection The first clinical report with xenon ventilation
for lung volume reduction surgery or broncho- imaging with dual-energy CT was reported by
scopic lung volume reduction. Data from Chae et al., investigating eight healthy volunteers
National Emphysema Treatment Trial with more and four patients with COPD. The authors have
than 1000 patients undergoing lung volume demonstrated the direct visualization of the
reduction surgery showed that lung volume degree of xenon gas enhancement in the lung
reduction surgery reduces mortality in patients parenchyma as a color overlay on a conventional
with upper-lobe predominant emphysema only if thin-section chest CT image by material decom-
there is low perfusion to the upper lobe on scin- position [153]. Park et al. performed two phase
8  Imaging of COPD 115

a b

c d

e f

Fig. 8.22  Xenon ventilation image with dual-energy CT weighted average image (a, d), centrilobular emphysema
in COPD patient. (a–c) Dual-energy CT in COPD patient and bronchial wall thickening in both lower lobes are
with mild emphysema. (d–f) Dual-energy CT in COPD noted. (b, c, e, f) In right upper lobe and left upper lobe,
patient with severe emphysema. (a, d) Conventional CT decreased xenon ventilation is identified, while xenon
image and (b, c, e, f) xenon ventilation map image ventilation is preserved in left upper lobe posterior
obtained from xenon ventilation dual-energy CT. On segment
116 S.M. Lee et al.

(wash-in and wash-out phase) xenon ventilation  ombined Ventilation and Perfusion

dual-energy CT in 32 patients with COPD. And Assessment with Dual-Energy CT
regional quantified value of xenon enhancement Pulmonary parenchymal perfusion change or
in abnormally low attenuating lung area on ventilation impairment was evaluated with dual-­
xenon-enhanced images with wash-out phase energy CT separately in COPD patients.
showed inverse correlation with pulmonary func- However, the pulmonary parenchymal ventila-
tion test, and it showed better correlation with tion and perfusion are changed concurrently, and
pulmonary function test than CT attenuation the imbalance between ventilation and perfusion
parameters [154]. Similar approaches were also is the important characteristics in the patients
reported in asthma patients. Investigating 22 with COPD. Investigating ten patients with vari-
asthma patients, Chae et al. have demonstrated ous diseases from an anesthesiological intensive
the ventilation defects that appeared on xenon care unit, Thieme et al. have reported the poten-
ventilation imaging with dual-energy CT in tial of dual-energy CT to provide both pulmonary
asthma patients with severe airflow limitation ventilation and perfusion imaging [161]. Zhang
and airway wall thickening. And the extent of the et al. applied combined ventilation and perfusion
ventilation defects on xenon-enhanced CT imaging with dual-energy CT in patients with
showed correlations with parameters of pulmo- suspected pulmonary embolism [162]. Combined
nary function test [155]. Demonstration of the ventilation and perfusion imaging with dual-­
reversibility of airflow obstruction after inhala- energy CT in patients with COPD has not yet
tion of bronchodilator has been reported, and it been reported (Fig. 8.23).
suggested that xenon-enhanced dual-energy CT
may be feasible for visualizing the changes of
airflow in response to drugs in asthma patients Nuclear Medicine Imaging
Stable krypton can be an alternative to xenon  cintigraphy, Single-Photon Emission
for ventilation imaging with dual-energy CT due Computed Tomography (SPECT)
to its high atomic number, and lack of toxicity Both planar scintigraphy and SPECT image the
and anesthetic properties. Hachulla et al. have distribution of radiotracer which is introduced
firstly reported the krypton ventilation imaging into the body, and the emitted radiation from
using dual-energy CT in COPD patients. Single radiotracer is detected by external detectors. In
CT acquisition covering the whole thorax was contrast to scintigraphy which forms a single
performed after inhalation of a mixture of 80% two-dimensional image, analogous to a planar
krypton and 20% oxygen with five respiratory X-ray scan, SPECT provides three-dimensional
maneuvers through the mask. The maximum imaging about the distribution of a radiotracer by
level of krypton enhancement within the lung combining scintigraphic and computed tomo-
was 18.5 HU, which is lower than that reported graphic technique, and allows the functional
with xenon, with an average maximum degree of information from SPECT to be easily combined
xenon enhancement of 23.78 HU, is sufficient to with the high-resolution anatomic information
detect ventilation abnormalities [159]. This sin- from CT. Perfusion scanning is generally per-
gle static approach in phantoms and volunteers formed using 99 m-technetium-labeled macroag-
also has been reported recently using xenon gas gregated albumin (99 m Tc-MAA), which lodges
after a single vital capacity inhalation [160]. The in the pulmonary circulation after peripheral
single static evaluation delivers a lower radiation injection. In an animal study using pigs, perfu-
dose and a few or single inhalation method is sion SPECT has been shown to be more sensitive
more easily implementable for radiologists and than HRCT to detect mild physiologic changes of
patients, without side effects. However, different elastase-induced pulmonary emphysema [163].
ventilation dynamics can be evaluated regarding Moreover, Suga et al. have been demonstrated
the scanning method and gas inhalation method. that perfusion abnormalities on breath-hold
8  Imaging of COPD 117

a b

c d

Fig. 8.23  Combined ventilation and perfusion assess- and perfusion map (d) are generated. Lung parenchymal
ment with dual-energy CT. From combined xenon ventila- ventilation, perfusion and ventilation-perfusion imbal-
tion and iodine perfusion CT using dual-energy CT, the ance with high-resolution anatomic CT information can
conventional virtual non-contrast CT image (a), xenon be simultaneously evaluated with combined ventilation
ventilation/iodine perfusion map (b), ventilation map (c) and perfusion dual-energy CT

SPECT-CT fusion image can better reflect the from endobronchial valve therapy [165].
lung pathophysiology than the emphysema index Although scintigraphy and SPECT have constitu-
on morphologic CT scan [164]. And there have tional problems, low spatial resolution and long
been several intervention studies in COPD image acquisition time, to date, they are widely
patients using perfusion scintigraphy and SPECT applied due to their availability in many centers.
to predict and measure clinical success. As men- Ventilation scintigraphy and SPECT use two
tioned earlier, assessing the lung parenchymal types of inhalation radiotracers: gaseous radio-
perfusion has been particularly applicable in tar- isotopes or radiolabeled particulate aerosols.
get lobe selection for lung volume reduction sur- 81 m Kr and 133 Xe as gaseous radioisotopes
gery or bronchoscopic lung volume reduction. have been used for ventilation scintigraphy and
Perfusion scintigraphy has been used in National SPECT, and several studies with gaseous radio-
Emphysema Treatment Trial with more than isotopes have demonstrated ventilation heteroge-
1000 patients undergoing lung volume reduction neity in COPD [166, 167]. And for radiolabeled
surgery for selection of target lobe [149]. And particulate aerosol, Technegas (99 m Tc-labeled,
perfusion scintigraphy was also useful for selec- aerosolized ultrafine carbon particle, approxi-
tion of target lobe for endobronchial valve ther- mately 200 nm diameter) is usually used in
apy in advanced emphysema patients, and patients with COPD due to its small particle
patients having heterogeneous emphysema with size [168], even in the presence of severe air-
a low baseline target lobe perfusion benefited flow obstruction [169]. With Technegas, the
118 S.M. Lee et al.

inhomogeneities in ventilation on scintigraphy  ositron Emission Tomography (PET)

and SPECT in COPD patients have been visual- Regional ventilation and perfusion also can be
ized and quantified [170, 171]. evaluated with PET using isotope 13N2 gas dis-
Ventilation/perfusion SPECT can also be solved in saline solution. Brudin et al. have reported
applied for the evaluation of the imbalance that high V/Q tended to be more common in sub-
between ventilation and perfusion in the patients jects with an emphysema dominant subtype,
with COPD. Jogi et al. have reported significant whereas low V/Q was more common in those with
correlation between total reduction in lung func- a bronchial inflammation dominant subtype.
tion assessed with ventilation/perfusion SPECT Spatial heterogeneity of lung perfusion also has
and spirometric lung function and emphysema been described with 13N2 saline PET, and the
severity on CT in patients with COPD [172] regional heterogeneity in perfusion has been
(Fig. 8.24). Suga et al. have described that quan- increased in patients with mild COPD compared to
titative analysis of V/Q distribution by SPECT healthy controls, after adjusting for regional
and the standard deviation and kurtosis of the changes in lung tissue density and ventilation.
V/Q profile could be adequate indicator for the [174]. These results suggest that regional perfusion
severity of lung V-Q imbalance causing gas changes may precede lung parenchymal destruc-
exchange impairment in patients with emphy- tion in COPD. Therefore, this imaging method
sema [173]. may serve as an early biomarker for COPD.



Fig. 8.24  Ventilation/perfusion SPECT. Patient with cen- to areas which are well ventilated (black arrows) and to
trilobular emphysema on HRCT. Ventilation/perfusion better preserved parenchyma on HRCT (white arrows).
SPECT shows uneven ventilation and perfusion. Extensive Other areas appearing as hot spots on ventilation images
areas with reverse as well as matched ventilation/perfu- are poorly perfused (dotted arrows). (Reprinted with per-
sion defects are found. Better perfused areas correspond mission, from reference 172)
8  Imaging of COPD 119

Recent studies have been focused on systemic negative correlation with the subject’s FEV1 than
inflammation as a consequence of COPD, and CT, and had greater sensitivity in detecting
patients with COPD are found to have higher changes in wall measurement than CT measure-
levels of systemic biomarkers of inflammation ments. And OCT also provided data on airway
and worse cardiovascular risk profile and progno- wall morphology and subepithelial remodeling
sis [175–177]. Fluorine-18-fluorodeoxyglucose and collagen deposition. While OCT is not
(18F-FDG) is the most commonly used PET widely applicable to date, the novel ability on
radioisotope, and it depicts increased glucose directly assessing small airway disease in COPD
metabolism. In COPD patients, 18F-FDG has patients may allow increasing utilization of OCT
been used to demonstrate both pulmonary and in research and clinical practice.
systemic inflammation. The pulmonary inflam-
mation as 18F-FDG uptake in COPD patients was
significantly higher than in asthmatic and normal References
subjects [178]. Pulmonary 18F-FDG uptake in
usual (PiM phenotype) COPD patients was 1. Larke FJ, Kruger RL, Cagnon CH, Flynn MJ,
greater compared to patients with alpha-1 anti- McNitt-Gray MM, Wu X, et al. Estimated radiation
dose associated with low-dose chest CT of average-­
trypsin deficiency-associated emphysema (PiZ size participants in the National Lung Screening
phenotype) and healthy controls, and it was cor- Trial. AJR Am J Roentgenol. 2011;197(5):1165–9.
related with FEV1 [179]. Coulson et al. have doi:10.2214/AJR.11.6533.
evaluated the aortic inflammation as a measure 2. Newell JD Jr, Sieren J, Hoffman EA. Development of
quantitative computed tomography lung protocols.
of systemic inflammation using 18F-FDG PET in J Thorac Imaging. 2013;28(5):266–71. doi:10.1097/
seven COPD patients, five metabolic syndrome RTI.0b013e31829f6796.
patients, and seven ex-smokers. Aortic inflam- 3. Mayo JR. CT evaluation of diffuse infiltrative lung
mation in COPD patients was intermediate disease: dose considerations and optimal technique.
J Thorac Imaging. 2009;24(4):252–9. doi:10.1097/
between ex-smokers and metabolic syndrome RTI.0b013e3181c227b2.
patients [180]. 4. Madani A, De Maertelaer V, Zanen J, Gevenois
PA. Pulmonary emphysema: radiation dose and sec-
tion thickness at multidetector CT quantification—
comparison with macroscopic and microscopic
Optical Coherence Tomography (OCT) morphometry. Radiology. 2007;243(1):250–7.
OCT is an imaging method utilizing the refrac- 5. Kemerink GJ, Kruize HH, Lamers RJ, van
tion of light waves as it passes through tissues, Engelshoven JM. CT lung densitometry: depen-
dence of CT number histograms on sample volume
and measures the lung structure with endobron- and consequences for scan protocol comparability.
chial approach. A fiberoptic probe with a near-­ J Comput Assist Tomogr. 1997;21(6):948–54.
infrared optical probe is introduced into the 6. Boedeker KL, McNitt-Gray MF, Rogers SR, Truong
airways via a bronchoscope and the reflected DA, Brown MS, Gjertson DW, et al. Emphysema:
effect of reconstruction algorithm on CT imag-
light by the tissue is detected and reconstructed ing measures. Radiology. 2004;232(1):295–301.
into an image. The spatial resolution of OCT is doi:10.1148/radiol.2321030383.
much higher than CT with the ability to resolve 7. Choo JY, Goo JM, Lee CH, Park CM, Park SJ,
structures up to micrometers and distinguish Shim MS. Quantitative analysis of emphysema
and airway measurements according to iterative
between different tissue types within the airways. reconstruction algorithms: comparison of filtered
Thus, the main strength of OCT in COPD is the back projection, adaptive statistical iterative recon-
ability assessing small airway morphology struction and model-based iterative reconstruction.
in vivo. Coxson et al. have demonstrated an Eur Radiol. 2014;24(4):799–806. doi:10.1007/
excellent correlation between airway lumen and 8. Mets OM, Willemink MJ, de Kort FP, Mol CP,
wall area measured by OCT and by CT [181]. Leiner T, Oudkerk M, et al. The effect of iterative
Furthermore, OCT airway dimensions measured reconstruction on computed tomography assessment
at the fifth generation bronchi showed a stronger of emphysema, air trapping and airway dimensions.
120 S.M. Lee et al.

Eur Radiol. 2012;22(10):2103–9. doi:10.1007/ Eur Radiol. 2009;19(7):1686–92. doi:10.1007/

s00330-012-2489-z. s00330-009-1320-y.
9. Nishio M, Matsumoto S, Ohno Y, Sugihara N, 21. Bankier AA, De Maertelaer V, Keyzer C, Gevenois
Inokawa H, Yoshikawa T, et al. Emphysema quan- PA. Pulmonary emphysema: subjective visual grad-
tification by low-dose CT: potential impact of adap- ing versus objective quantification with macroscopic
tive iterative dose reduction using 3D processing. morphometry and thin-section CT densitometry.
AJR Am J Roentgenol. 2012;199(3):595–601. Radiology. 1999;211(3):851–8. doi:10.1148/radiolo
doi:10.2214/AJR.11.8174. gy.211.3.r99jn05851.
10. Schroeder JD, McKenzie AS, Zach JA, Wilson CG, 22. Group COCW, Barr RG, Berkowitz EA, Bigazzi
Curran-Everett D, Stinson DS, et al. Relationships F, Bode F, Bon J, et al. A combined pulmonary-­
between airflow obstruction and quantitative CT radiology workshop for visual evaluation of COPD:
measurements of emphysema, air trapping, and air- study design, chest CT findings and concordance
ways in subjects with and without chronic obstruc- with quantitative evaluation. COPD. 2012;9(2):151–
tive pulmonary disease. AJR Am J Roentgenol. 9. doi:10.3109/15412555.2012.654923.
2013;201(3):W460–70. doi:10.2214/AJR.12.10102. 23. Hayhurst MD, MacNee W, Flenley DC, Wright D,
11. Stockley RA, Mannino D, Barnes PJ. Burden and McLean A, Lamb D, et al. Diagnosis of pulmonary
pathogenesis of chronic obstructive pulmonary emphysema by computerised tomography. Lancet.
disease. Proc Am Thorac Soc. 2009;6(6):524–6. 1984;2(8398):320–2.
doi:10.1513/pats.200904-016DS. 24. Muller NL, Staples CA, Miller RR, Abboud RT.
12. Roy K, Smith J, Kolsum U, Borrill Z, Vestbo J, “Density mask”. An objective method to quantitate
Singh D. COPD phenotype description using prin- emphysema using computed tomography. Chest.
cipal components analysis. Respir Res. 2009;10:41. 1988;94(4):782–7.
doi:10.1186/1465-9921-10-41. 25. Madani A, Zanen J, de Maertelaer V, Gevenois
13. Bergin C, Muller N, Nichols DM, Lillington G, PA. Pulmonary emphysema: objective quanti-
Hogg JC, Mullen B, et al. The diagnosis of emphy- fication at multi-detector row CT—comparison
sema. A computed tomographic-pathologic correla- with macroscopic and microscopic morphometry.
tion. Am Rev Respir Dis. 1986;133(4):541–6. Radiology. 2006;238(3):1036–43. doi:10.1148/
14. Bafadhel M, Umar I, Gupta S, Raj JV, Vara DD, radiol.2382042196.
Entwisle JJ, et al. The role of CT scanning in 26. Dirksen A, Friis M, Olesen KP, Skovgaard LT,
multidimensional phenotyping of COPD. Chest. Sorensen K. Progress of emphysema in severe
2011;140(3):634–42. doi:10.1378/chest.10-3007. alpha 1-antitrypsin deficiency as assessed by annual
15. O'Brien C, Guest PJ, Hill SL, Stockley CT. Acta Radiol. 1997;38(5):826–32.
RA. Physiological and radiological characterisa- 27. Dirksen A. Monitoring the progress of emphy-
tion of patients diagnosed with chronic obstruc- sema by repeat computed tomography scans with
tive pulmonary disease in primary care. Thorax. focus on noise reduction. Proc Am Thorac Soc.
2000;55(8):635–42. 2008;5(9):925–8. doi:10.1513/pats.200804-033QC.
16. McDonough JE, Yuan R, Suzuki M, Seyednejad N, 28. Heussel CP, Herth FJ, Kappes J, Hantusch R, Hartlieb
Elliott WM, Sanchez PG, et al. Small-airway obstruc- S, Weinheimer O, et al. Fully automatic quantitative
tion and emphysema in chronic obstructive pulmo- assessment of emphysema in computed tomography:
nary disease. N Engl J Med. 2011;365(17):1567–75. comparison with pulmonary function testing and
doi:10.1056/NEJMoa1106955. normal values. Eur Radiol. 2009;19(10):2391–402.
17. Lehouck A, Boonen S, Decramer M, Janssens doi:10.1007/s00330-009-1437-z.
W. COPD, bone metabolism, and osteoporo- 29. Stolk J, Dirksen A, van der Lugt AA, Hutsebaut
sis. Chest. 2011;139(3):648–57. doi:10.1378/ J, Mathieu J, de Ree J, et al. Repeatability of lung
chest.10-1427. density measurements with low-dose computed
18. Miller RR, Muller NL, Vedal S, Morrison NJ, tomography in subjects with alpha-1-antitrypsin
Staples CA. Limitations of computed tomography in deficiency-associated emphysema. Investig Radiol.
the assessment of emphysema. Am Rev Respir Dis. 2001;36(11):648–51.
1989;139(4):980–3. doi:10.1164/ajrccm/139.4.980. 30. Gietema HA, Muller NL, Fauerbach PV, Sharma
19. Kim SS, Seo JB, Lee HY, Nevrekar DV, Forssen S, Edwards LD, Camp PG, et al. Quantifying the
AV, Crapo JD, et al. Chronic obstructive pulmonary extent of emphysema: factors associated with radi-
disease: lobe-based visual assessment of volumetric ologists’ estimations and quantitative indices of
CT by using standard images—comparison with emphysema severity using the ECLIPSE cohort.
quantitative CT and pulmonary function test in the Acad Radiol. 2011;18(6):661–71. doi:10.1016/j.
COPDGene study. Radiology. 2013;266(2):626–35. acra.2011.01.011.
doi:10.1148/radiol.12120385. 31. Parr DG, Stoel BC, Stolk J, Stockley RA. Pattern
20. Cavigli E, Camiciottoli G, Diciotti S, Orlandi of emphysema distribution in alpha1-antitrypsin
I, Spinelli C, Meoni E, et al. Whole-lung densi- deficiency influences lung function impairment.
tometry versus visual assessment of emphysema.
8  Imaging of COPD 121

Am J Respir Crit Care Med. 2004;170(11):1172–8. correlation with CT indices and pulmonary function
doi:10.1164/rccm.200406-761OC. test. Eur J Radiol. 2011;80(3):e531–5. doi:10.1016/j.
32. Nakano Y, Sakai H, Muro S, Hirai T, Oku Y, Nishimura ejrad.2010.12.062.
K, et al. Comparison of low attenuation areas on com- 43. Boiselle PM, Litmanovich DE, Michaud G, Roberts
puted tomographic scans between inner and outer seg- DH, Loring SH, Womble HM, et al. Dynamic expi-
ments of the lung in patients with chronic obstructive ratory tracheal collapse in morbidly obese COPD
pulmonary disease: incidence and contribution to lung patients. COPD. 2013;10(5):604–10. doi:10.3109/1
function. Thorax. 1999;54(5):384–9. 5412555.2013.781149.
33. Chae EJ, Seo JB, Song JW, Kim N, Park BW, Lee 44. Regan EA, Hokanson JE, Murphy JR, Make B, Lynch
YK, et al. Slope of emphysema index: an objective DA, Beaty TH, et al. Genetic epidemiology of COPD
descriptor of regional heterogeneity of emphysema (COPDGene) study design. COPD. 2010;7(1):32–
and an independent determinant of pulmonary func- 43. doi:10.3109/15412550903499522.
tion. AJR Am J Roentgenol. 2010;194(3):W248–55. 45. Matsuoka S, Kurihara Y, Yagihashi K, Nakajima
doi:10.2214/AJR.09.2672. Y. Quantitative assessment of peripheral airway
34. Nakano Y, Coxson HO, Bosan S, Rogers RM, obstruction on paired expiratory/inspiratory thin-­
Sciurba FC, Keenan RJ, et al. Core to rind distri- section computed tomography in chronic obstruc-
bution of severe emphysema predicts outcome of tive pulmonary disease with emphysema. J Comput
lung volume reduction surgery. Am J Respir Crit Assist Tomogr. 2007;31(3):384–9. doi:10.1097/01.
Care Med. 2001;164(12):2195–9. doi:10.1164/ rct.0000243457.00437.10.
ajrccm.164.12.2012140. 46. Matsuoka S, Kurihara Y, Yagihashi K, Hoshino M,
35. Martinez-Garcia MA, Soler-Cataluna JJ, Donat Sanz Watanabe N, Nakajima Y. Quantitative assessment
Y, Catalan Serra P, Agramunt Lerma M, Ballestin of air trapping in chronic obstructive pulmonary
Vicente J, et al. Factors associated with bronchiecta- disease using inspiratory and expiratory volumetric
sis in patients with COPD. Chest. 2011;140(5):1130– MDCT. AJR Am J Roentgenol. 2008;190(3):762–9.
7. doi:10.1378/chest.10-1758. doi:10.2214/AJR.07.2820.
36. Grydeland TB, Thorsen E, Dirksen A, Jensen R, 47. Yamashiro T, Matsuoka S, Bartholmai BJ, San Jose
Coxson HO, Pillai SG, et al. Quantitative CT mea- Estepar R, Ross JC, Diaz A, et al. Collapsibility of
sures of emphysema and airway wall thickness are lung volume by paired inspiratory and expiratory
related to D(L)CO. Respir Med. 2011;105(3):343– CT scans: correlations with lung function and mean
51. doi:10.1016/j.rmed.2010.10.018. lung density. Acad Radiol. 2010;17(4):489–95.
37. Nakano Y, Muro S, Sakai H, Hirai T, Chin K, Tsukino doi:10.1016/j.acra.2009.11.004.
M, et al. Computed tomographic measurements of 48. Lee YK, Oh YM, Lee JH, Kim EK, Lee JH, Kim
airway dimensions and emphysema in smokers. N, et al. Quantitative assessment of emphysema, air
Correlation with lung function. Am J Respir Crit trapping, and airway thickening on computed tomog-
Care Med. 2000;162(3 Pt 1):1102–8. doi:10.1164/ raphy. Lung. 2008;186(3):157–65. doi:10.1007/
ajrccm.162.3.9907120. s00408-008-9071-0.
38. Washko GR, Dransfield MT, Estepar RS, Diaz A, 49. Galban CJ, Han MK, Boes JL, Chughtai KA,
Matsuoka S, Yamashiro T, et al. Airway wall attenu- Meyer CR, Johnson TD, et al. Computed tomogra-
ation: a biomarker of airway disease in subjects phy-based biomarker provides unique signature for
with COPD. J Appl Physiol. 2009;107(1):185–91. diagnosis of COPD phenotypes and disease progres-
doi:10.1152/japplphysiol.00216.2009. sion. Nat Med. 2012;18(11):1711–5. doi:10.1038/
39. Nakano Y, Wong JC, de Jong PA, Buzatu L, Nagao nm.2971.
T, Coxson HO, et al. The prediction of small air- 50. Barbosa EM Jr, Song G, Tustison N, Kreider M,
way dimensions using computed tomography. Gee JC, Gefter WB, et al. Computational analysis
Am J Respir Crit Care Med. 2005;171(2):142–6. of thoracic multidetector row HRCT for segmenta-
doi:10.1164/rccm.200407-874OC. tion and quantification of small airway air trapping
40. Achenbach T, Weinheimer O, Biedermann A, Schmitt and emphysema in obstructive pulmonary disease.
S, Freudenstein D, Goutham E, et al. MDCT assess- Acad Radiol. 2011;18(10):1258–69. doi:10.1016/j.
ment of airway wall thickness in COPD patients acra.2011.06.004.
using a new method: correlations with pulmonary 51. Ginsburg SB, Lynch DA, Bowler RP, Schroeder
function tests. Eur Radiol. 2008;18(12):2731–8. JD. Automated texture-based quantification of cen-
doi:10.1007/s00330-008-1089-4. trilobular nodularity and centrilobular emphysema in
41. Han MK, Kazerooni EA, Lynch DA, Liu LX, chest CT images. Acad Radiol. 2012;19(10):1241–
Murray S, Curtis JL, et al. Chronic obstructive pul- 51. doi:10.1016/j.acra.2012.04.020.
monary disease exacerbations in the COPDGene 52. Park YS, Seo JB, Kim N, Chae EJ, Oh YM, Lee SD,
study: associated radiologic phenotypes. Radiology. et al. Texture-based quantification of pulmonary
2011;261(1):274–82. doi:10.1148/radiol.11110173. emphysema on high-resolution computed tomog-
42. Lee HJ, Seo JB, Chae EJ, Kim N, Lee CW, Oh YM, raphy: comparison with density-based quantifica-
et al. Tracheal morphology and collapse in COPD: tion and correlation with pulmonary function test.
122 S.M. Lee et al.

Investig Radiol. 2008;43(6):395–402. doi:10.1097/ a population-based study and a systematic review

RLI.0b013e31816901c7. of the literature. Chest. 2005;127(6):1952–9.
53. Barr RG, Mesia-Vela S, Austin JH, Basner RC, doi:10.1378/chest.127.6.1952.
Keller BM, Reeves AP, et al. Impaired flow-mediated 64. Sin DD, Man SF. Chronic obstructive pulmonary
dilation is associated with low pulmonary function disease as a risk factor for cardiovascular morbidity
and emphysema in ex-smokers: the emphysema and and mortality. Proc Am Thorac Soc. 2005;2(1):8–11.
cancer action project (EMCAP) study. Am J Respir doi:10.1513/pats.200404-032MS.
Crit Care Med. 2007;176(12):1200–7. doi:10.1164/ 65. Chae EJ, Seo JB, Oh YM, Lee JS, Jung Y, Lee
rccm.200707-980OC. SD. Severity of systemic calcified atherosclerosis
54. Ross JC, Estepar RS, Diaz A, Westin CF, Kikinis is associated with airflow limitation and emphy-
R, Silverman EK, et al. Lung extraction, lobe sema. J Comput Assist Tomogr. 2013;37(5):743–9.
­segmentation and hierarchical region assessment for doi:10.1097/RCT.0b013e318299f9e7.
quantitative analysis on high resolution computed 66. Ayres SM, Griesbach SJ, Reimold F, Evans
tomography images. Med Image Comput Comput RG. Bronchial component in chronic obstructive
Assist Interv. 2009;12(Pt 2):690–8. lung disease. Am J Med. 1974;57(2):183–91.
55. Xiao C, Staring M, Shamonin D, Reiber JH, Stolk 67. Pare PD, Lawson LM, Brooks LA. Patterns of
J, Stoel BC. A strain energy filter for 3D vessel response to inhaled bronchodilators in asthmatics.
enhancement with application to pulmonary CT Am Rev Respir Dis. 1983;127(6):680–5.
images. Med Image Anal. 2011;15(1):112–24. 68. Lee JS, Huh JW, Chae EJ, Seo JB, Ra SW, Lee
doi:10.1016/ JH, et al. Response patterns to bronchodilator and
56. Matsuoka S, Washko GR, Dransfield MT, Yamashiro quantitative computed tomography in chronic
T, San Jose Estepar R, Diaz A, et al. Quantitative CT obstructive pulmonary disease. Clin Physiol
measurement of cross-sectional area of small pulmo- Funct Imaging. 2012;32(1):12–8. doi:10.1111/j.
nary vessel in COPD: correlations with emphysema 1475-097X.2011.01046.x.
and airflow limitation. Acad Radiol. 2010;17(1):93– 69. Dirksen A, Piitulainen E, Parr DG, Deng C,
9. doi:10.1016/j.acra.2009.07.022. Wencker M, Shaker SB, et al. Exploring the
57. Harrison RA, Siminoski K, Vethanayagam D, role of CT densitometry: a randomised study
Majumdar SR. Osteoporosis-related kyphosis and of augmentation therapy in alpha1-antitrypsin
impairments in pulmonary function: a systematic deficiency. Eur Respir J. 2009;33(6):1345–53.
review. J Bone Miner Res Off J Am Soc Bone Miner doi:10.1183/09031936.00159408.
Res. 2007;22(3):447–57. doi:10.1359/jbmr.061202. 70. Kim SS, Seo JB, Kim N, Chae EJ, Lee YK, Oh YM,
58. Ohara T, Hirai T, Muro S, Haruna A, Terada K, et al. Improved correlation between CT emphysema
Kinose D, et al. Relationship between pulmonary quantification and pulmonary function test by den-
emphysema and osteoporosis assessed by CT in sity correction of volumetric CT data based on air
patients with COPD. Chest. 2008;134(6):1244–9. and aortic density. Eur J Radiol. 2014;83(1):57–63.
doi:10.1378/chest.07-3054. doi:10.1016/j.ejrad.2012.02.021.
59. Kiyokawa H, Muro S, Oguma T, Sato S, Tanabe N, 71. Parr DG, Stoel BC, Stolk J, Nightingale PG, Stockley
Takahashi T, et al. Impact of COPD exacerbations RA. Influence of calibration on densitometric studies
on osteoporosis assessed by chest CT scan. COPD. of emphysema progression using computed tomog-
2012;9(3):235–42. doi:10.3109/15412555.2011.650 raphy. Am J Respir Crit Care Med. 2004;170(8):883–
243. 90. doi:10.1164/rccm.200403-326OC.
60. Marquis K, Debigare R, Lacasse Y, LeBlanc P, 72. Madani A, Van Muylem A, Gevenois PA. Pulmonary
Jobin J, Carrier G, et al. Midthigh muscle cross-­ emphysema: effect of lung volume on objec-
sectional area is a better predictor of mortality than tive quantification at thin-section CT. Radiology.
body mass index in patients with chronic obstruc- 2010;257(1):260–8. doi:10.1148/radiol.10091446.
tive pulmonary disease. Am J Respir Crit Care Med. 73. Grydeland TB, Dirksen A, Coxson HO, Pillai SG,
2002;166(6):809–13. doi:10.1164/rccm.2107031. Sharma S, Eide GE, et al. Quantitative computed
61. Soler-Cataluna JJ, Sanchez-Sanchez L, tomography: emphysema and airway wall thickness by
Martinez-Garcia MA, Sanchez PR, Salcedo sex, age and smoking. Eur Respir J. 2009;34(4):858–
E, Navarro M. Mid-arm muscle area is a bet- 65. doi:10.1183/09031936.00167908.
ter predictor of mortality than body mass index in 74. Camiciottoli G, Cavigli E, Grassi L, Diciotti S,
COPD. Chest. 2005;128(4):2108–15. doi:10.1378/ Orlandi I, Zappa M, et al. Prevalence and correlates
chest.128.4.2108. of pulmonary emphysema in smokers and former
62. Park MJ, Cho JM, Jeon KN, Bae KS, Kim HC, Choi smokers. A densitometric study of participants in
DS, et al. Mass and fat infiltration of intercostal the ITALUNG trial. Eur Radiol. 2009;19(1):58–66.
muscles measured by CT histogram analysis and doi:10.1007/s00330-008-1131-6.
their correlations with COPD severity. Acad Radiol. 75. Ley-Zaporozhan J, Ley S, Kauczor HU. Morphological
2014;21(6):711–7. doi:10.1016/j.acra.2014.02.003. and functional imaging in COPD with CT and MRI:
63. Sin DD, Wu L, Man SF. The relationship between present and future. Eur Radiol. 2008;18(3):510–21.
reduced lung function and cardiovascular ­mortality: doi:10.1007/s00330-007-0772-1.
8  Imaging of COPD 123

76. Bankier AA, O’Donnell CR, Mai VM, Storey P, asthma in mice: inflammation and bronchial remod-
De Maertelaer V, Edelman RR, et al. Impact of eling assessment. NMR Biomed. 2013;26(11):1451–
lung volume on MR signal intensity changes of 9. doi:10.1002/nbm.2975.
the lung parenchyma. J Magn Reson Imaging. 88. Puderbach M, Eichinger M, Gahr J, Ley S,
2004;20(6):961–6. doi:10.1002/jmri.20198. Tuengerthal S, Schmahl A, et al. Proton MRI
77. Rajaram S, Swift AJ, Capener D, Telfer A, Davies appearance of cystic fibrosis: comparison to CT. Eur
C, Hill C, et al. Lung morphology assessment with Radiol. 2007;17(3):716–24. doi:10.1007/s00330-
balanced steady-state free precession MR imaging 006-0373-4.
compared with CT. Radiology. 2012;263(2):569–77. 89. Biederer J, Both M, Graessner J, Liess C, Jakob P,
doi:10.1148/radiol.12110990. Reuter M, et al. Lung morphology: fast MR imaging
78. Ley-Zaporozhan J, Ley S, Eberhardt R, Kauczor assessment with a volumetric interpolated breath-­
HU, Heussel CP. Visualization of morphological hold technique: initial experience with patients.
parenchymal changes in emphysema: compari- Radiology. 2003;226(1):242–9. doi:10.1148/
son of different MRI sequences to 3D-HRCT. Eur radiol.2261011974.
J Radiol. 2010;73(1):43–9. doi:10.1016/j.ejrad. 90. Beckmann N, Cannet C, Zurbruegg S, Rudin M, Tigani
2008.09.029. B. Proton MRI of lung parenchyma reflects allergen-
79. Ley-Zaporozhan J, Ley S, Kauczor HU. Proton induced airway remodeling and endotoxin-­ aroused
MRI in COPD. COPD. 2007;4(1):55–65. hyporesponsiveness: a step toward ventilation studies
doi:10.1080/15412550701198719. in spontaneously breathing rats. Magn Reson Med.
80. Vogel-Claussen J, Renne J, Hinrichs J, Schonfeld C, 2004;52(2):258–68. doi:10.1002/mrm.20127.
Gutberlet M, Schaumann F, et al. Quantification of 91. Ogasawara N, Suga K, Zaki M, Okada M,
pulmonary inflammation after segmental allergen Kawakami Y, Matsunaga N. Assessment of lung
challenge using turbo-inversion recovery-­magnitude perfusion impairment in patients with pulmonary
magnetic resonance imaging. Am J Respir Crit artery-­
occlusive and chronic obstructive pulmo-
Care Med. 2014;189(6):650–7. doi:10.1164/rccm. nary diseases with noncontrast electrocardiogram-­
201310-1825OC. gated fast-spin-echo perfusion MR imaging. J Magn
81. Iwasawa T, Takahashi H, Ogura T, Asakura A, Gotoh Reson Imaging. 2004;20(4):601–11. doi:10.1002/
T, Kagei S, et al. Correlation of lung parenchy- jmri.20150.
mal MR signal intensity with pulmonary function 92. Suga K, Ogasawara N, Okada M, Hara A, Matsunaga
tests and quantitative computed tomography (CT) N. Potential of noncontrast electrocardiogram-gated
evaluation: a pilot study. J Magn Reson Imaging. half-fourier fast-spin-echo magnetic resonance
2007;26(6):1530–6. doi:10.1002/jmri.21183. imaging to monitor dynamically altered perfusion in
82. Mayo JR, MacKay A, Muller NL. MR imaging of the regional lung. Investig Radiol. 2002;37(11):615–25.
lungs: value of short TE spin-echo pulse sequences. doi:10.1097/01.RLI.0000031079.78361.7D.
AJR Am J Roentgenol. 1992;159(5):951–6. 93. Hatabu H, Tadamura E, Levin DL, Chen Q, Li W,
doi:10.2214/ajr.159.5.1414805. Kim D, et al. Quantitative assessment of pulmo-
83. Bergin CJ, Pauly JM, Macovski A. Lung paren- nary perfusion with dynamic contrast-enhanced
chyma: projection reconstruction MR imaging. MRI. Magn Reson Med. 1999;42(6):1033–8.
Radiology. 1991;179(3):777–81. doi:10.1148/ 94. Levin DL, Chen Q, Zhang M, Edelman RR, Hatabu
radiology.179.3.2027991. H. Evaluation of regional pulmonary perfusion using
84. Ohno Y, Koyama H, Yoshikawa T, Matsumoto K, ultrafast magnetic resonance imaging. Magn Reson
Takahashi M, Van Cauteren M, et al. T2* mea- Med. 2001;46(1):166–71.
surements of 3-T MRI with ultrashort TEs: capa- 95. Keilholz SD, Mai VM, Berr SS, Fujiwara N, Hagspiel
bilities of pulmonary function assessment and KD. Comparison of first-pass Gd-DOTA and
clinical stage classification in smokers. AJR Am FAIRER MR perfusion imaging in a rabbit model
J Roentgenol. 2011;197(2):W279–85. doi:10.2214/ of pulmonary embolism. J Magn Reson Imaging.
AJR.10.5350. 2002;16(2):168–71. doi:10.1002/jmri.10138.
85. Takahashi M, Togao O, Obara M, van Cauteren M, 96. Korosec FR, Frayne R, Grist TM, Mistretta
Ohno Y, Doi S, et al. Ultra-short echo time (UTE) CA. Time-resolved contrast-enhanced 3D MR angi-
MR imaging of the lung: comparison between nor- ography. Magn Reson Med. 1996;36(3):345–51.
mal and emphysematous lungs in mutant mice. 97. Fink C, Ley S, Kroeker R, Requardt M, Kauczor HU,
J Magn Reson Imaging. 2010;32(2):326–33. Bock M. Time-resolved contrast-enhanced three-­
doi:10.1002/jmri.22267. dimensional magnetic resonance angiography of the
86. Johnson KM, Fain SB, Schiebler ML, Nagle S. Optimized chest: combination of parallel imaging with view
3D ultrashort echo time pulmonary MRI. Magn sharing (TREAT). Investig Radiol. 2005;40(1):40–8.
Reson Med. 2013;70(5):1241–50. doi:10.1002/ 98. Fink C, Ley S, Risse F, Eichinger M, Zaporozhan J,
mrm.24570. Buhmann R, et al. Effect of inspiratory and expira-
87. Bianchi A, Ozier A, Ousova O, Raffard G, tory breathhold on pulmonary perfusion: assessment
Cremillieux Y. Ultrashort-TE MRI longitudinal by pulmonary perfusion magnetic resonance imag-
study and characterization of a chronic model of ing. Investig Radiol. 2005;40(2):72–9.
124 S.M. Lee et al.

99. Dehnert C, Risse F, Ley S, Kuder TA, Buhmann R, Eur J Radiol. 2007;64(3):320–8. doi:10.1016/j.
Puderbach M, et al. Magnetic resonance imaging of ejrad.2007.08.006.
uneven pulmonary perfusion in hypoxia in humans. 112. Ohno Y, Hatabu H, Takenaka D, Van Cauteren M,
Am J Respir Crit Care Med. 2006;174(10):1132–8. Fujii M, Sugimura K. Dynamic oxygen-enhanced
doi:10.1164/rccm.200606-780OC. MRI reflects diffusing capacity of the lung. Magn
100. Sergiacomi G, Sodani G, Fabiano S, Manenti G, Reson Med. 2002;47(6):1139–44. doi:10.1002/
Spinelli A, Konda D, et al. MRI lung perfusion mrm.10168.
2D dynamic breath-hold technique in patients with 113. Ohno Y, Hatabu H, Takenaka D, Adachi S, Van
severe emphysema. In Vivo. 2003;17(4):319–24. Cauteren M, Sugimura K. Oxygen-enhanced
101. Molinari F, Fink C, Risse F, Tuengerthal S, Bonomo MR ventilation imaging of the lung: prelimi-
L, Kauczor HU. Assessment of differential pulmo- nary clinical experience in 25 subjects. AJR Am
nary blood flow using perfusion magnetic resonance J Roentgenol. 2001;177(1):185–94. doi:10.2214/
imaging: comparison with radionuclide perfusion ajr.177.1.1770185.
scintigraphy. Investig Radiol. 2006;41(8):624–30. 114. Kirby M, Mathew L, Wheatley A, Santyr GE,
doi:10.1097/01.rli.0000225399.65609.45. McCormack DG, Parraga G. Chronic obstructive
102. Morino S, Toba T, Araki M, Azuma T, Tsutsumi S, pulmonary disease: longitudinal hyperpolarized (3)
Tao H, et al. Noninvasive assessment of pulmonary He MR imaging. Radiology. 2010;256(1):280–9.
emphysema using dynamic contrast-enhanced mag- doi:10.1148/radiol.10091937.
netic resonance imaging. Exp Lung Res. 2006;32(1– 115. Kaushik SS, Cleveland ZI, Cofer GP, Metz G, Beaver
2):55–67. doi:10.1080/01902140600691548. D, Nouls J, et al. Diffusion-weighted hyperpolarized
103. Ley-Zaporozhan J, Ley S, Eberhardt R, Weinheimer 129Xe MRI in healthy volunteers and subjects with
O, Fink C, Puderbach M, et al. Assessment of the chronic obstructive pulmonary disease. Magn Reson
relationship between lung parenchymal destruc- Med. 2011;65(4):1154–65. doi:10.1002/mrm.22697.
tion and impaired pulmonary perfusion on a lobar 116. Altes TA, Salerno M. Hyperpolarized gas MR imag-
level in patients with emphysema. Eur J Radiol. ing of the lung. J Thorac Imaging. 2004;19(4):250–8.
2007;63(1):76–83. doi:10.1016/j.ejrad.2007.01.020. 117. de Lange EE, Altes TA, Patrie JT, Gaare JD, Knake JJ,
104. Jang YM, Oh YM, Seo JB, Kim N, Chae EJ, Lee Mugler JP III, et al. Evaluation of asthma with hyper-
YK, et al. Quantitatively assessed dynamic contrast-­ polarized helium-3 MRI: correlation with clinical
enhanced magnetic resonance imaging in patients severity and spirometry. Chest. 2006;130(4):1055–
with chronic obstructive pulmonary disease: cor- 62. doi:10.1378/chest.130.4.1055.
relation of perfusion parameters with pulmonary 118. van Beek EJ, Wild JM, Kauczor HU, Schreiber W,
function test and quantitative computed tomography. Mugler JP III, de Lange EE. Functional MRI of the
Investig Radiol. 2008;43(6):403–10. doi:10.1097/ lung using hyperpolarized 3-helium gas. J Magn
RLI.0b013e31816901ab. Reson Imaging. 2004;20(4):540–54. doi:10.1002/
105. Ohno Y, Hatabu H, Murase K, Higashino T, Kawamitsu jmri.20154.
H, Watanabe H, et al. Quantitative assessment of 119. Frahm J, Haase A, Matthaei D. Rapid NMR imaging
regional pulmonary perfusion in the entire lung of dynamic processes using the FLASH technique.
using three-dimensional ultrafast dynamic contrast- Magn Reson Med. 1986;3(2):321–7.
enhanced magnetic resonance imaging: preliminary 120. Bock M. Simultaneous T2* and diffusion measure-
experience in 40 subjects. J Magn Reson Imaging. ments with 3He. Magn Reson Med. 1997;38(6):890–5.
2004;20(3):353–65. doi:10.1002/jmri.20137. 121. Salerno M, Altes TA, Brookeman JR, de Lange EE,
106. Tadamura E, Hatabu H, Li W, Prasad PV, Edelman Mugler JP III. Dynamic spiral MRI of pulmonary
RR. Effect of oxygen inhalation on relaxation gas flow using hyperpolarized (3)He: preliminary
times in various tissues. J Magn Reson Imaging. studies in healthy and diseased lungs. Magn Reson
1997;7(1):220–5. Med. 2001;46(4):667–77.
107. Brooks RA, Di Chiro G. Magnetic resonance 122. Deninger AJ, Eberle B, Ebert M, Grossmann T,
imaging of stationary blood: a review. Med Phys. Heil W, Kauczor H, et al. Quantification of regional
1987;14(6):903–13. intrapulmonary oxygen partial pressure evolu-
108. Thulborn KR, Waterton JC, Matthews PM, Radda tion during apnea by (3)He MRI. J Magn Reson.
GK. Oxygenation dependence of the transverse relax- 1999;141(2):207–16. doi:10.1006/jmre.1999.1902.
ation time of water protons in whole blood at high 123. Deninger AJ, Eberle B, Bermuth J, Escat B,
field. Biochim Biophys Acta. 1982;714(2):265–70. Markstaller K, Schmiedeskamp J, et al. Assessment
109. Ohno Y, Chen Q, Hatabu H. Oxygen-enhanced mag- of a single-acquisition imaging sequence for
netic resonance ventilation imaging of lung. Eur oxygen-sensitive (3)He-MRI. Magn Reson Med.
J Radiol. 2001;37(3):164–71. 2002;47(1):105–14.
110. Bauman G, Eichinger M. Ventilation and perfu- 124. van Beek EJ, Dahmen AM, Stavngaard T, Gast KK,
sion magnetic resonance imaging of the lung. Pol Heussel CP, Krummenauer F, et al. Hyperpolarised
J Radiol. 2012;77(1):37–46. 3He MRI versus HRCT in COPD and normal volun-
111. Ohno Y, Hatabu H. Basics concepts and clinical teers: PHIL trial. Eur Respir J. 2009;34(6):1311–21.
applications of oxygen-enhanced MR imaging. doi:10.1183/09031936.00138508.
8  Imaging of COPD 125

125. Kirby M, Svenningsen S, Owrangi A, Wheatley A, 136. Millner MR, McDavid WD, Waggener RG, Dennis
Farag A, Ouriadov A, et al. Hyperpolarized 3He MJ, Payne WH, Sank VJ. Extraction of informa-
and 129Xe MR imaging in healthy volunteers and tion from CT scans at different energies. Med Phys.
patients with chronic obstructive pulmonary dis- 1979;6(1):70–1.
ease. Radiology. 2012;265(2):600–10. doi:10.1148/ 137. Rutherford RA, Pullan BR, Isherwood I. X-ray ener-
radiol.12120485. gies for effective atomic number determination.
126. Swift AJ, Wild JM, Fichele S, Woodhouse N,
Neuroradiology. 1976;11(1):23–8.
Fleming S, Waterhouse J, et al. Emphysematous 138. Barbera JA, Riverola A, Roca J, Ramirez J, Wagner
changes and normal variation in smokers and PD, Ros D, et al. Pulmonary vascular abnor-
COPD patients using diffusion 3He MRI. Eur malities and ventilation-perfusion relationships in
J Radiol. 2005;54(3):352–8. doi:10.1016/j. mild chronic obstructive pulmonary disease. Am
ejrad.2004.08.002. J Respir Crit Care Med. 1994;149(2 Pt 1):423–9.
127. Salerno M, de Lange EE, Altes TA, Truwit JD, doi:10.1164/ajrccm.149.2.8306040.
Brookeman JR, Mugler JP III. Emphysema: hyper- 139. Hale KA, Niewoehner DE, Cosio MG. Morphologic
polarized helium 3 diffusion MR imaging of the changes in the muscular pulmonary arteries: relation-
lungs compared with spirometric indexes—ini- ship to cigarette smoking, airway disease, and emphy-
tial experience. Radiology. 2002;222(1):252–60. sema. Am Rev Respir Dis. 1980;122(2):273–8.
doi:10.1148/radiol.2221001834. 140. Peinado VI, Barbera JA, Ramirez J, Gomez FP,
128. Hermosillo G, Chefd'Hotel C, Faugeras Roca J, Jover L, et al. Endothelial dysfunction in
O. Variational methods for multimodal image match- pulmonary arteries of patients with mild COPD. Am
ing. Int J Comput Vision. 2002;50(3):329–43. doi:10 J Phys. 1998;274(6 Pt 1):L908–13.
.1023/A:1020830525823. 141. Barbera JA, Roca J, Ferrer A, Felez MA, Diaz
129. Bauman G, Puderbach M, Deimling M, Jellus O, Roger N, et al. Mechanisms of worsen-
V, Chefd'hotel C, Dinkel J, et al. Non-contrast-­ ing gas exchange during acute exacerbations of
enhanced perfusion and ventilation assessment of chronic obstructive pulmonary disease. Eur Respir
the human lung by means of fourier decomposition J. 1997;10(6):1285–91.
in proton MRI. Magn Reson Med. 2009;62(3):656– 142. Sandek K, Bratel T, Hellstrom G, Lagerstrand
64. doi:10.1002/mrm.22031. L. Ventilation-perfusion inequality and carbon
130. Suga K, Ogasawara N, Okada M, Tsukuda T, dioxide sensitivity in hypoxaemic chronic obstruc-
Matsunaga N, Miyazaki M. Lung perfusion tive pulmonary disease (COPD) and effects of 6
impairments in pulmonary embolic and air- months of long-term oxygen treatment (LTOT). Clin
way obstruction with noncontrast MR imaging. Physiol. 2001;21(5):584–93.
J Appl Physiol. 2002;92(6):2439–51. doi:10.1152/ 143. Fuld MK, Halaweish AF, Haynes SE, Divekar AA,
japplphysiol.00900.2001. Guo J, Hoffman EA. Pulmonary perfused blood vol-
131. Bauman G, Lutzen U, Ullrich M, Gaass T, Dinkel ume with dual-energy CT as surrogate for pulmo-
J, Elke G, et al. Pulmonary functional imaging: nary perfusion assessed with dynamic multidetector
qualitative comparison of Fourier decomposition CT. Radiology. 2013;267(3):747–56. doi:10.1148/
MR imaging with SPECT/CT in porcine lung. radiol.12112789.
Radiology. 2011;260(2):551–9. doi:10.1148/ 144. Barbera JA, Peinado VI, Santos S. Pulmonary hyper-
radiol.11102313. tension in chronic obstructive pulmonary disease.
132. Gierada DS, Curtin JJ, Erickson SJ, Prost RW, Eur Respir J. 2003;21(5):892–905.
Strandt JA, Goodman LR. Diaphragmatic motion: 145. Ley S, Puderbach M, Fink C, Eichinger M, Plathow
fast gradient-recalled-echo MR imaging in C, Teiner S, et al. Assessment of hemodynamic
healthy subjects. Radiology. 1995;194(3):879–84. changes in the systemic and pulmonary arterial cir-
doi:10.1148/radiology.194.3.7862995. culation in patients with cystic fibrosis using phase-­
133. Suga K, Tsukuda T, Awaya H, Takano K, Koike S, contrast MRI. Eur Radiol. 2005;15(8):1575–80.
Matsunaga N, et al. Impaired respiratory mechan- doi:10.1007/s00330-005-2721-1.
ics in pulmonary emphysema: evaluation with 146. Alford SK, van Beek EJ, McLennan G, Hoffman
dynamic breathing MRI. J Magn Reson Imaging. EA. Heterogeneity of pulmonary perfusion as a
1999;10(4):510–20. mechanistic image-based phenotype in emphy-
134. Iwasawa T, Takahashi H, Ogura T, Asakura A, sema susceptible smokers. Proc Natl Acad Sci U
Gotoh T, Shibata H, et al. Influence of the distri- S A. 2010;107(16):7485–90. doi:10.1073/pnas.
bution of emphysema on diaphragmatic motion in 0913880107.
patients with chronic obstructive pulmonary dis- 147. Pansini V, Remy-Jardin M, Faivre JB, Schmidt B,
ease. Jpn J Radiol. 2011;29(4):256–64. doi:10.1007/ Dejardin-Bothelo A, Perez T, et al. Assessment of
s11604-010-0552-8. lobar perfusion in smokers according to the pres-
135. Iwasawa T, Yoshiike Y, Saito K, Kagei S, Gotoh ence and severity of emphysema: preliminary
T, Matsubara S. Paradoxical motion of the hemi- experience with dual-energy CT angiography.
diaphragm in patients with emphysema. J Thorac Eur Radiol. 2009;19(12):2834–43. doi:10.1007/
Imaging. 2000;15(3):191–5. s00330-009-1475-6.
126 S.M. Lee et al.

148. Lee CW, Seo JB, Lee Y, Chae EJ, Kim N, Lee HJ, 159. Hachulla AL, Pontana F, Wemeau-Stervinou L,
et al. A pilot trial on pulmonary emphysema quanti- Khung S, Faivre JB, Wallaert B, et al. Krypton ven-
fication and perfusion mapping in a single-step using tilation imaging using dual-energy CT in chronic
contrast-enhanced dual-energy computed tomogra- obstructive pulmonary disease patients: initial expe-
phy. Investig Radiol. 2012;47(1):92–7. doi:10.1097/ rience. Radiology. 2012;263(1):253–9. doi:10.1148/
RLI.0b013e318228359a. radiol.12111211.
149. Chandra D, Lipson DA, Hoffman EA, Hansen-­ 160. Honda N, Osada H, Watanabe W, Nakayama M,
Flaschen J, Sciurba FC, Decamp MM, et al. Nishimura K, Krauss B, et al. Imaging of ventilation
Perfusion scintigraphy and patient selection for with dual-energy CT during breath hold after single
lung volume reduction surgery. Am J Respir Crit vital-capacity inspiration of stable xenon. Radiology.
Care Med. 2010;182(7):937–46. doi:10.1164/ 2012;262(1):262–8. doi:10.1148/radiol.11110569.
rccm.201001-0043OC. 161. Thieme SF, Hoegl S, Nikolaou K, Fisahn J,
150. Park TS, Hong Y, Lee JS, Lee SM, Seo JB, Oh Irlbeck M, Maxien D, et al. Pulmonary ventila-
YM, et al. Efficacy of bronchoscopic lung volume tion and perfusion imaging with dual-energy
reduction by endobronchial valves in patients with CT. Eur Radiol. 2010;20(12):2882–9. doi:10.1007/
heterogeneous emphysema: report on the first Asian s00330-010-1866-8.
cases. J Korean Med Sci. 2014;29(10):1404–10. 162. Zhang LJ, Zhou CS, Schoepf UJ, Sheng HX, Wu
doi:10.3346/jkms.2014.29.10.1404. SY, Krazinski AW, et al. Dual-energy CT lung ven-
151. Gur D, Drayer BP, Borovetz HS, Griffith BP, Hardesty tilation/perfusion imaging for diagnosing pulmo-
RL, Wolfson SK. Dynamic computed tomography nary embolism. Eur Radiol. 2013;23(10):2666–75.
of the lung: regional ventilation measurements. doi:10.1007/s00330-013-2907-x.
J Comput Assist Tomogr. 1979;3(6):749–53. 163. Noma S, Moskowitz GW, Herman PG, Khan A,
152. Herbert DL, Gur D, Shabason L, Good WF, Rojas KA. Pulmonary scintigraphy in elastase-­
Rinaldo JE, Snyder JV, et al. Mapping of human induced emphysema in pigs. Correlation with high-­
local pulmonary ventilation by xenon enhanced resolution computed tomography and histology.
computed tomography. J Comput Assist Tomogr. Investig Radiol. 1992;27(6):429–35.
1982;6(6):1088–93. 164. Suga K, Kawakami Y, Iwanaga H, Hayashi N, Seto
153. Chae EJ, Seo JB, Goo HW, Kim N, Song KS, Lee A, Matsunaga N. Assessment of anatomic relation
SD, et al. Xenon ventilation CT with a dual-energy between pulmonary perfusion and morphology in
technique of dual-source CT: initial experience. pulmonary emphysema with breath-hold SPECT-CT
Radiology. 2008;248(2):615–24. doi:10.1148/ fusion images. Ann Nucl Med. 2008;22(5):339–47.
radiol.2482071482. doi:10.1007/s12149-007-0137-5.
154. Park EA, Goo JM, Park SJ, Lee HJ, Lee CH, Park 165. Argula RG, Strange C, Ramakrishnan V, Goldin
CM, et al. Chronic obstructive pulmonary disease: J. Baseline regional perfusion impacts exercise
quantitative and visual ventilation pattern analysis response to endobronchial valve therapy in advanced
at xenon ventilation CT performed by using a dual-­ pulmonary emphysema. Chest. 2013;144(5):1578–
energy technique. Radiology. 2010;256(3):985–97. 86. doi:10.1378/chest.12-2826.
doi:10.1148/radiol.10091502. 166. Stavngaard T, Sogaard LV, Mortensen J, Hanson
155. Chae EJ, Seo JB, Lee J, Kim N, Goo HW, Lee LG, Schmiedeskamp J, Berthelsen AK, et al.
HJ, et al. Xenon ventilation imaging using dual-­ Hyperpolarized 3He MRI and 81mKr SPECT in
energy computed tomography in asthmatics: initial chronic obstructive pulmonary disease. Eur J Nucl
experience. Investig Radiol. 2010;45(6):354–61. Med Mol Imaging. 2005;32(4):448–57. doi:10.1007/
doi:10.1097/RLI.0b013e3181dfdae0. s00259-004-1691-x.
156. Goo HW, Yu J. Redistributed regional ventilation 167. Suga K, Nishigauchi K, Kume N, Koike S,
after the administration of a bronchodilator demon- Takano K, Tokuda O, et al. Dynamic pulmonary
strated on xenon-inhaled dual-energy CT in a patient SPECT of xenon-133 gas washout. J Nucl Med.
with asthma. Korean J Radiol. 2011;12(3):386–9. 1996;37(5):807–14.
doi:10.3348/kjr.2011.12.3.386. 168. Amis TC, Crawford AB, Davison A, Engel
157. Kim WW, Lee CH, Goo JM, Park SJ, Kim JH, LA. Distribution of inhaled 99mtechnetium labelled
Park EA, et al. Xenon-enhanced dual-energy ultrafine carbon particle aerosol (Technegas) in
CT of patients with asthma: dynamic ventilation human lungs. Eur Respir J. 1990;3(6):679–85.
changes after methacholine and salbutamol inhala- 169. Crawford AB, Davison A, Amis TC, Engel
tion. AJR Am J Roentgenol. 2012;199(5):975–81. LA. Intrapulmonary distribution of 99mtechne-
doi:10.2214/AJR.11.7624. tium labelled ultrafine carbon aerosol (Technegas)
158. Park SJ, Lee CH, Goo JM, Kim JH, Park EA, in severe airflow obstruction. Eur Respir
Jung JW, et al. Quantitative analysis of dynamic J. 1990;3(6):686–92.
airway changes after methacholine and salbuta- 170. Cukic V, Begic A. Potential role of lung ventila-
mol inhalation on xenon-enhanced chest CT. Eur tion scintigraphy in the assessment of COPD. Acta
Radiol. 2012;22(11):2441–50. doi:10.1007/ Inform Med. 2014;22(3):170–3. doi:10.5455/
s00330-012-2516-0. aim.2014.22.170-173.
8  Imaging of COPD 127

171. Xu J, Moonen M, Johansson A, Gustafsson A, Bake mation in stable and exacerbation phases of
B. Quantitative analysis of inhomogeneity in ventila- COPD. Lung. 2008;186(6):403–9. doi:10.1007/
tion SPET. Eur J Nucl Med. 2001;28(12):1795–800. s00408-008-9106-6.
doi:10.1007/s002590100649. 177. Sinden NJ, Stockley RA. Systemic inflammation
172. Jogi J, Ekberg M, Jonson B, Bozovic G, Bajc and comorbidity in COPD: a result of ‘overspill’
M. Ventilation/perfusion SPECT in chronic obstruc- of inflammatory mediators from the lungs? Review
tive pulmonary disease: an evaluation by reference of the evidence. Thorax. 2010;65(10):930–6.
to symptoms, spirometric lung function and emphy- doi:10.1136/thx.2009.130260.
sema, as assessed with HRCT. Eur J Nucl Med 178. Jones HA, Marino PS, Shakur BH, Morrell NW. In
Mol Imaging. 2011;38(7):1344–52. doi:10.1007/ vivo assessment of lung inflammatory cell activ-
s00259-011-1757-5. ity in patients with COPD and asthma. Eur Respir
173. Suga K, Kawakami Y, Koike H, Iwanaga H, Tokuda J. 2003;21(4):567–73.
O, Okada M, et al. Lung ventilation-perfusion 179. Subramanian DR, Jenkins L, Edgar R, Quraishi N,
imbalance in pulmonary emphysema: assess- Stockley RA, Parr DG. Assessment of pulmonary
ment with automated V/Q quotient SPECT. Ann neutrophilic inflammation in emphysema by quan-
Nucl Med. 2010;24(4):269–77. doi:10.1007/ titative positron emission tomography. Am J Respir
s12149-010-0369-7. Crit Care Med. 2012;186(11):1125–32. doi:10.1164/
174. Vidal Melo MF, Winkler T, Harris RS, Musch G, rccm.201201-0051OC.
Greene RE, Venegas JG. Spatial heterogeneity of 180. Coulson JM, Rudd JH, Duckers JM, Rees JI,

lung perfusion assessed with (13)N PET as a vascu- Shale DJ, Bolton CE, et al. Excessive aortic
lar biomarker in chronic obstructive pulmonary dis- inflammation in chronic obstructive pulmonary
ease. J Nucl Med. 2010;51(1):57–65. doi:10.2967/ disease: an 18F-­ FDG PET pilot study. J Nucl
jnumed.109.065185. Med. 2010;51(9):1357–60. doi:10.2967/jnumed.
175. Gan WQ, Man SF, Senthilselvan A, Sin 110.075903.
DD. Association between chronic obstructive pul- 181. Coxson HO, Quiney B, Sin DD, Xing L,
monary disease and systemic inflammation: a McWilliams AM, Mayo JR, et al. Airway wall
systematic review and a meta-analysis. Thorax. thickness assessed using computed tomography
2004;59(7):574–80. and optical coherence tomography. Am J Respir
176. Karadag F, Karul AB, Cildag O, Yilmaz M, Crit Care Med. 2008;177(11):1201–6. ­doi:10.1164/
Ozcan H. Biomarkers of systemic inflam- rccm.200712-1776OC.
Biomarkers of COPD
Ho Il Yoon

Introduction this endpoint unsuitable for drug discovery in

COPD. The discovery of a validated, reliable,
Chronic obstructive pulmonary disease (COPD) robust, and reproducible blood biomarker would
is a major health burden across the world. provide a major boost to the development of
Globally, more than 300 million people suffer novel compounds because it would allow investi-
from COPD [1] and nearly three million die each gators (and companies) to demonstrate the thera-
year from this disease [2]. COPD mortality con- peutic promise of a drug in small (usually phase
tinues to climb at an alarming rate, such that by II) trials before proceeding to a much more
2030, nearly nine million people will die annu- expensive and logistically difficult phase III tri-
ally from COPD [3]. The economic burden of als. Without such data, pharmaceutical compa-
COPD is also enormous. In the United States nies are hesitant to invest millions of dollars on
alone, COPD accounted for $20.9 billion USD in large phase III studies to bring compounds to
direct and $7.4 billion USD in indirect costs in market. For this reason, some international com-
2004 [4]. Regrettably, the pipeline for new drugs panies have recently abandoned COPD drug
for COPD is relatively dry compared to other development altogether, while many others have
major causes of mortality such as HIV/AIDs, scaled back their efforts significantly. The pur-
cancer, and diabetes [5]. One major barrier to pose of this chapter is to review potential bio-
drug discovery in COPD is the paucity of well-­ markers of COPD, especially those that could be
accepted and well-validated biomarkers. used in predicting treatment responses.
Currently, the only “marker” that is widely
accepted by regulatory agencies for new drug
approval in COPD is FEV1 (forced expiratory Sources of Biomarkers
volume in 1 s), which is a robust measure of lung
function. However, COPD is defined operation- Biomarkers can originate from any organ.
ally as a chronic respiratory condition that results However, because COPD is predominantly a lung
in airflow limitation which is progressive and not disease, the airways are the most logical source
fully reversible [6]. In other words, COPD is for identifying novel biomarkers in COPD. There
defined by limited reversibility of FEV1, making are three major sources of airway samples. They
include sputum (spontaneous or induced), exhaled
gases or condensates, or bronchial washes or
HI. Yoon brushes. The latter source requires bronchoscopy,
Seoul National University Bundang Hospital, which is invasive. Thus, repeated measurements
Seongnam, South Korea
e-mail: are usually not feasible. Furthermore, bronchial

© Springer-Verlag Berlin Heidelberg 2017 129

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_9
130 HI. Yoon

sampling is limited to one or two airways, which and (5) molecules variably associated with COPD
may result in sampling error because the human and/or tobacco smoke.
lung contains more than 40,000 terminal bronchi- In addition, it would be crucial to differentiate
oles [7]. For these and other reasons, broncho- biomarkers in terms of their reproducibility
scopic samples have not been used for biomarker within and across laboratories and platforms to
discovery in COPD and will not be discussed any identify its relevance despite technical differ-
further in this chapter. ences [12].
Both spontaneous and induced sputum and
exhaled gases and condensates also have method-
ological challenges that limit their utility as a Sputum Neutrophils
source of biomarker discovery. For example,
induced sputum is hard to process and may contain By far, induced sputum has been the most popu-
contaminants such as saliva that makes it difficult lar source of biomarker discovery in
to relate sputum findings to the pathophysiological COPD. Sputum contains a mixture of mucins
processes that are occurring in the lungs [8]. (approximately 2–4% of the total weight), salts,
Exhaled gases and more recently condensates have tissue plasma, lipids, and inflammatory cells
been used to identify novel biomarkers in from tissue and airway lumen, microbial prod-
COPD. However, the data to date have been vari- ucts, cellular debris, and inhaled particles from
able and inconsistent, owing to lack of standardiza- the environment [8]. The predominant mucins in
tion of methods and heterogeneity in the techniques sputum are MUC5B and MUC5AC [8]. Induced
employed for sample collection and analysis [9] sputum is an attractive source of biomarker dis-
and the fact that most proteins are unpredictable or covery because it is relatively noninvasive to
below the lower limit of quantification [10]. obtain even in subjects who do not spontaneously
Nevertheless, these sources are commonly used for expectorate and may reflect the inflammatory
biomarker discovery in COPD because they are process in the airways. However, there are sev-
readily accessible and noninvasive. eral challenges in identifying possible biomark-
Other than its sources, different criteria can be ers in sputum. These include (1) the use of
used to classify biomarkers of COPD. These can liquefaction agents such as DTT (dithiothreitol)
be by the biological process where they are modi- or NAC (N-acetylcysteine), which may disturb or
fied, by purposes of which they can be used in the interfere with protein measurements in sputum;
clinical field, and sometimes by the association (2) the lack of consistent marker that can be used
with tobacco smoke. In effect, the respiratory to normalize sputum data. This makes it difficult
biomarkers could be classified to five categories to compare results across subjects and across
in relation to their association with exposure to studies; (3) contamination of sputum by saliva,
tobacco smoke and/or with COPD [11]: (1) bio- which may obscure biomarkers in sputum sam-
markers of response to tobacco smoke exposure, ples (largely by dilution); and (4) lack of stan-
associated with COPD activity, which are dardization in the collection and processing of
expressed at higher levels in healthy smokers samples and in data analysis. Furthermore, the
than in non-smokers and at higher levels in source of the induced sputum is likely the large
COPD than in healthy smokers; (2) biomarkers (central) airways and not the small peripheral air-
of COPD inflammation whose expression level is ways, which are thought to be the major site of
not associated with tobacco smoke exposure, as it disease in COPD [7]. The use of spontaneous
is higher in COPD patients than in controls, with sputum has similar issues but is more likely to
no significant differences between healthy smok- reflect increased inflammation and can be sam-
ers and non-smokers; (3) biomarkers of response pled on a sequential basis reducing variability
to tobacco smoke exposure, not associated with [13].
COPD activity; (4) biomarkers negatively associ- Notwithstanding these limitations, sputums
ated with COPD and/or tobacco smoke exposure; have been used for biomarker discovery because
9  Biomarkers of COPD 131

they can be procured noninvasively and some inhaled corticosteroid therapy for at least 6 weeks
components of sputum may reflect the active results in a significant reduction in sputum neu-
inflammatory process in the airways of COPD trophil count in patients with stable COPD [25].
patients. One such biomarker is the neutrophil Together, these studies validate the concept of
count in induced sputum. It is an attractive bio- using sputum neutrophilia as a biomarker to eval-
marker in COPD because neutrophils are thought uate “neutrophilic” airway inflammation in
to play an important role in the pathogenesis of patients with COPD.
COPD [14, 15], secreting various cytokines that Recently, several investigators have applied
provoke and perpetuate lung inflammation [16, this biomarker to evaluate emerging (new) thera-
17]. Furthermore, neutrophils are easily measur- pies in COPD. For instance, He and colleagues
able in sputum of COPD patients [18]. With used sputum neutrophil count to assess the pos-
smoking and with COPD progression, the spu- sible beneficial effects of erythromycin on airway
tum contains increasing percentages of neutro- inflammation [10]. In this randomized controlled
phils [19]. Interestingly, once COPD is firmly trial, they found that erythromycin (at 125 mg
established, neutrophil count remains elevated three times per day) significantly reduced sputum
even following smoking cessation [20]. Most neutrophilia by 3 months and interestingly also
importantly, elevated expression of sputum neu- reduced the risk of exacerbations suggesting anti-­
trophils has been associated with rapid decline in inflammatory and/or antibacterial effects [10].
FEV1, highlighting the importance of neutrophils Ford and colleagues used sputum neutrophils to
in COPD pathogenesis [21]. However, the rela- assess the possible benefits of low-dose theophyl-
tionship between the neutrophil count and per- line in the management of COPD patients. Thirty
cent neutrophils is exponential and absolute patients with COPD were treated with placebo
count may therefore be more relevant. theophylline capsules plus either inhaled flutica-
Neutrophil counts in sputum have also been sone propionate (500 microg bid) or inhaled pla-
used as a biomarker to evaluate well-established cebo for 4 weeks in a double-dummy, randomized,
COPD drugs. For instance, in the study by Barnes double-blind, parallel study. Then following a
et al., sputum neutrophil count was used as a co-­ two-week-washout period, patients were given
primary endpoint to evaluate the possible effi- active theophylline. However, contrary to expec-
cacy of fluticasone/salmeterol combination in the tations, they found that combination treatment
treatment of COPD [22]. Over 3 months, they did with fluticasone and theophylline did not reduce
not find a significant change in sputum neutrophil total sputum neutrophils [13]. However, in
count between fluticasone/salmeterol combina- another study, 8 weeks of low-dose theophylline
tion and placebo [22]. This was surprising as therapy was associated with reduced sputum neu-
fluticasone/salmeterol combination has been trophil count [26]. Sputum neutrophil count has
shown to reduce the rate of COPD progression also been used as a primary endpoint in clinical
(albeit marginally) in patients with established studies evaluating promising but not yet approved
COPD [23] suggesting the study may have been COPD drugs. For example, Gronke et al. used
underpowered [13]. Similarly, Laperre et al. eval- sputum neutrophils as a proof of concept that leu-
uated the effects of fluticasone alone and in com- kotriene B4 receptor antagonist, which in vitro
bination with salmeterol on a variety of different and in animal studies decreased neutrophil
outcomes included sputum neutrophil count. recruitment to the lungs, unfortunately the study
They showed that at 30 months of therapy, the proved negative in COPD [27].
group that received fluticasone had significantly Although neutrophil counts in sputum have
lower sputum neutrophil count than did the pla- been widely used to assess possible therapeutic
cebo group [24]. On the other hand, salmeterol benefits of drugs for COPD, it has not been fully
by itself had no significant effect on sputum neu- validated as a biomarker in COPD because
trophilia [24]. These data are consistent with a ­sputum neutrophil measurements are only weakly
meta-analysis published in 2005, showing that associated with FEV1 or with health status [28]
132 HI. Yoon

and there have been no large-scale studies that diagnostic marker in asthma [32] and a useful
have demonstrated its utility in predicting biomarker in assessing severity and control [33]
important health outcomes in COPD such as and in evaluating treatment responses, espe-
exacerbation or mortality [28]. Furthermore, cially to inhaled corticosteroids [34, 35]. The
despite the implication of neutrophils in the data in COPD are more scarce and generally
pathogenesis of emphysema, sputum neutro- less impressive. Levels of FENO are usually
phils do not appear to correlate with the extent normal or only modestly elevated in COPD,
of emphysema in COPD patients [28] which except during exacerbations [36–38] and have
seems to be natural because sputum neutrophil not been demonstrated to predict important
mainly reflects conditions of large airways not health outcomes in COPD. Thus, FENO is not a
of lung parenchyma. promising biomarker in COPD. However, with
recent innovations in FENO measurements,
there is renewed optimism of using FENO as a
Other Sputum Biomarkers biomarker in COPD. One such modification is
multiple exhalation flow technique (MEFT),
There are other cellular elements in induced spu- which allows separate measurement of NO
tum that have been used as possible biomarkers derived from small airways and alveoli (called
in COPD. These include total cell count and spu- corrected alveolar nitric oxide (CALV)) from
tum eosinophil and lymphocyte counts [25, 29]. those of larger airways. The subdivision of
The long-term use of inhaled corticosteroids has FENO into small and large airway compart-
been shown to reduce total cell and lymphocyte ments is very attractive in COPD as COPD is
counts, and eosinophilia in induced sputum [25]. thought to involve largely the small airways and
However, as with sputum neutrophilia, none of lung parenchyma. A recent study showed that
these measurements has been consistently shown CALV was elevated in COPD patients but disap-
to predict important health outcomes in pointingly, it was not associated with COPD
COPD. As such, their usefulness as biomarkers severity or smoking status of the patients [39].
in COPD remains uncertain. Furthermore, CALV was not affected by inhaled
Some investigators have used supernatants corticosteroid treatment [39]. Additional work
from induced sputum to measure levels of inflam- will be needed to understand the value, if any,
matory and oxidative molecules, which have that FENO has in COPD.
been implicated in COPD pathogenesis. These
include interleukin (IL)-8, IL-6, myeloperoxi-
dase, and matrix metalloproteinases (MMP)-9. Other Exhaled Biomarkers
However, as with cellular components of induced
sputum, the relationship of these inflammatory There are a number of other potential biomarkers
and oxidative stress molecules in sputum to from exhaled breath condensate (EBC). EBC pH
health outcomes such as COPD progression and is an acidification marker and was reported to be
mortality has not been well established. lower in smokers and COPD than in control non-­
smokers. But it failed to differentiate COPD from
smokers without COPD, to relate to disease
 raction of Exhaled Nitric Oxide
F severity, and to reflect response to corticosteroids
(FENO) [40]. Concentrations of leukotriene B4 (LTB4)
and 8-isoprostane, markers of inflammation and
Given these limitations of induced sputum, oxidative stress, respectively, were known to
there has been growing enthusiasm for develop- increase in COPD exacerbations and decrease
ing exhaled gases as biomarkers in COPD [30]. after antibiotic treatment [41]. LTB4 was further
Of these, the best studied has been FENO [31]. reported to contribute to neutrophil chemotactic
In general, FENO shows good specificity as a activity [42]. Another oxidative stress marker,
9  Biomarkers of COPD 133

hydrogen peroxide, reflected level of oxidative changes have also been noted in mice exposed to
stress in patients with COPD after exercise, but acute cigarette smoke [52]. Interestingly, mice
its performance in terms of therapeutic response which are deficient in TNF-α or its receptors are
biomarker is unknown yet [43]. protected against elastic fiber degradation and
their urinary desmosine excretion is relatively
normal in the presence of cigarette smoke. These
Desmosine/Isodesmosine mice are also relatively protected against smok-
ing-induced emphysematous changes in the
While the pathogenesis of COPD is complex and lungs [52, 53].
poorly understood, there is some agreement that Blood desmosine levels were reported to be
excess breakdown and turnover of extracellular elevated in stable COPD patients compared to
matrix in the lungs is likely to be very important healthy control and asthma patients. It was found
for the emphysema phenotype of COPD [44]. to be in negative relation with lung diffusing
The extracellular matrix in the lungs acts as a capacity. Moreover, urinary desmosine levels
three-dimensional scaffold, providing strength were reported to be elevated in exacerbations.
and support for the alveolar units. The extracel- [54] Human data, recently reviewed by Luisetti
lular matrix consists mostly of collagens (type 1 et al. [49], have been more mixed and less opti-
and 3), which provide tensile strength, elastin, mistic with some study showing a significant
which confers flexibility to the lung tissues and relationship of urinary (or plasma) desmosine/
glycosaminoglycans [45]. In the normal lungs, isodesmosine to COPD, while others failing to
the collagen fibers tightly wrap around elastic demonstrate this relationship. The heterogeneity
bands in an orderly manner. However, in COPD, in the human data reflects in part major differ-
this structure is disrupted, leading to a “random” ences in the modalities of measurement (e.g.,
distribution of collagen and elastic fibers [46]. immunoassays versus high-performance liquid
Furthermore, with COPD disease progression, chromatography), the biological source of assays
the ratio of collagen to elastin surrounding the (e.g., urine versus plasma), and the underlying
alveolar units becomes distorted, owing largely clinical characteristics of the patients across the
to a severe loss in elastin [46]. The functional studies. Furthermore, most of the human studies
consequence is a loss in elastic recoil pressure to date have been relatively small in scope and
and “floppy” airways. Consistent with these did not consider the large variations in the pheno-
observations, deletion of the elastin gene in mice types of the patients. Recently, ECLIPSE
results in emphysematous changes in the lungs (Evaluation of COPD Longitudinally to Identify
and mutations in the elastin gene in humans (e.g., Predictive Surrogate Endpoints) investigators
cutix laxa) have been associated with severe early reported association between serum desmosine
onset emphysema [47, 48]. level and cardiovascular mortality in COPD, not
Based on these and other findings, there has emphysema progression [55]. Notwithstanding
been some interest in using markers of elastin these limitations, the totality of data suggests that
turnover as possible biomarkers in COPD. Of desmosine or isodesmosine is elevated in the
these, the best studied have been desmosine and urine of COPD patients [56–58].
isodesmosine [49]. Because these proteins are Very few studies have evaluated these markers
present only in mature elastin, it is thought that in therapeutic trials, demonstrating mixed results.
their expression in blood, urine, or sputum mostly For instance, a study by Stone and colleagues
reflects elastin degradation [50]. Consistent with showed that short-term treatment with alpha-­1-­
this theory, injection of pancreatic elastase into antrypsin replacement therapy led to significant
animals leads to an acute loss of elastin and reductions in urinary excretion of desmosine in
emphysematous changes in the lungs. Degradation two patients with severe COPD secondary to
of elastic fibers in the lungs in turn can be detected alpha-1-antitrypsin deficiency [59]. However, in
in the urine in the form of desmosine [51]. These another study, the use of alpha-1-antitrypsin
134 HI. Yoon

replacement therapy did not modify urinary in bronchoalveolar lavage samples of patients
excretion of desmosine in these patients [60]. Ma with COPD but not in control individuals [63].
and colleagues evaluated plasma, urine, and spu- Similarly, these markers are detectable in sputum
tum expression of desmosine and isodesmosine samples of COPD patients but not in those of
related to the use of tiotropium, which is a long-­ control subjects [69]. In serum, PGP levels are
acting muscarinic bronchodilator, and found that also significantly higher in COPD patients than
after 2 months of therapy, their expression in controls [69]. Interestingly, patients with cys-
decreased in COPD patients [61]. However, all of tic fibrosis also have increased PGP expression in
these studies contained very small sample sizes, sputum as in COPD, likely related to the intense
which reduces the reliability of these reports. neutrophilic inflammation observed in these con-
Thus, the possible role of desmosine or isodes- ditions [64]. However, to date, PGP or N-α-PGP
mosine as biomarkers in COPD remains obscure. has not been associated with important health
In a phase II trial, oral matrix metalloproteinase outcomes in COPD, and they have not been used
(MMP)-9 and -12 inhibitor has been shown to in therapeutic trials. Thus, their possible role in
reduce urinary desmosine excretion [62] evaluating new products in COPD is uncertain. In
one study, levels of PGP in sputum were highest
during exacerbation and azithromycin treatment
PGP (Proline-Glycine-Proline), lowered its levels [70]. Similarly, roflumilast has
N-α-PGP been reported to reduce sputum AcPGP by more
than 50% in COPD patients with chronic bron-
It is increasingly recognized that breakdown chitis [71].
products of collagen in the lungs may be pro-­
inflammatory and exacerbate the inflammatory
process in COPD, creating a vicious cycle. One Leukotriene A4 Hydrolase (LTA4H)
such molecule produced by collagen metabolism
is PGP. PGP shares structural homology with Leukotriene A4 hydrolase (LTA4H) is an enzyme
alpha chemokines and similarly to these chemo- released from neutrophils and epithelial cells that
kines induces the recruitment of neutrophils into generates leukotriene A4 (LTA4) via its hydrolase
lungs by stimulating CXCR-1 and most impor- activity in the cytosol. LTA4 is a potent chemoat-
tantly, CXCR-2 receptors [63], which in turn tractant for neutrophils. LTA4H also has a pepti-
augment the inflammatory cascade in the COPD dase activity. In the extracellular milieu, it
lungs [64–66]. In mouse models, instillation of degrades PGP or N-α-PGP, which, in turn, down-­
N-acetyl-Proline-Glycine-Proline (N-α-PGP) regulates neutrophilic inflammation. Interestingly,
into the lungs of mice causes a marked recruit- cigarette smoking inhibits the peptidase but not
ment of neutrophils to the airways and chronic the hydrolase activity of LTA4H, thus up-regulat-
airway exposure causes COPD-like pathology in ing neutrophilic inflammation [72]. The balance
the lungs characterized by alveolar enlargement between these two enzymatic activities of LTA4H
and right ventricular hypertrophy [63]. Blockage may be important in determining the extent and
of PGP using monoclonal antibody, on the other duration of neutrophilic inflammation in the lungs
hand, suppresses both the in vitro chemotactic of COPD patients.
activity and in vivo recruitment of inflammatory
cells into the lungs related to cigarette smoke [63,
67]. Additionally, treatment of these mice with a Markers of Systemic Inflammation
complementary peptide, L-arginine-threonine-­
arginine (RTR), which binds to PGP sequences, Although biomarkers can be obtained from any
inhibits neutrophil infiltration and prevents devel- organ, the most successful ones are from blood
opment of pulmonary emphysema [68]. because blood is easy to obtain and its measure-
Could PGP or N-α-PGP be a useful biomarker ments can be standardized. Furthermore, with
in COPD? On the affirmative side, PGP is detectable modern techniques, high throughput (and
9  Biomarkers of COPD 135

relatively inexpensive) assays can be developed CRP is relatively insensitive to the effects of
for blood-based protein measurements. In inhaled glucocorticoids [84, 85]. Thus, the role of
COPD, there are no such blood biomarkers cur- CRP as a biomarker for therapeutic drugs in
rently approved for use. Moreover, because COPD remains uncertain.
blood samples are usually obtained in periph-
eral veins (e.g., antecubital fossa) and not from
pulmonary veins or arteries, there is concern Fibrinogen
that blood biomarkers in COPD may not ade-
quately reflect the disease activity (or severity) Fibrinogen is another widely used biomarker of
in the lungs, which are the primary sites of dis- systemic inflammation. Plasma fibrinogen levels
ease in COPD. Nevertheless, there is increasing strongly associate with coronary heart disease
evidence that the inflammatory process in the (CHD), stroke, other vascular mortality, and non-
lungs “spill-over” into the systemic circulation, vascular mortality [86]. They also predict future
which may result in an “inflammatory signa- risk of moderate and severe exacerbations [87]
ture” in blood related to COPD [73]. and hospitalizations from COPD [88]. Moreover,
serum levels of fibrinogen have been found to be
related to mortality in COPD patients [89, 90].
C-reactive Protein (CRP) However, plasma fibrinogen levels are not easily
modifiable with medications [91, 92].
It is now well accepted that lung inflammation
is an important component in the pathogenesis
of COPD, and this inflammatory process carries IL-6
over into the systemic circulation [73].
Accordingly, many groups of investigators are The main up-stream regulator of CRP, fibrinogen,
endeavoring to identify plasma proteins that are and other acute phase proteins such as serum
specific to the inflammatory process in COPD amyloid A is interleukin (IL)-6 (93). IL-6, along
and relate these biomarkers to salient clinical with other primary inflammatory cytokines like
health outcomes such as exacerbations and TNF-α and IL-1β, responds to an acute insult by
mortality. Of these, the best studied to date is orchestrating and unleashing a cocktail of inflam-
C-reactive protein (CRP). CRP is a sensitive matory mediators from the liver and other organs
but not a specific marker of systemic inflamma- to contain the insult and protect the body from
tion and tissue damage. It has been extensively harm [93]. In the lungs, IL-6 is synthesized pre-
studied in patients with ischemic heart disease dominantly by alveolar macrophages and bron-
and other cardiovascular diseases (CVD) [74, chial epithelial cells and is released in response to
75] and has been shown to predict CVD mor- environmental triggers such as cigarette smoke
bidity and mortality. Importantly, CRP has also and air pollution particles [94]. IL-6 in turn is
been shown to be a useful biomarker in guiding accompanied by inflammatory cells including
statin therapy for patients who do not have neutrophils into the lungs and limits the effects of
raised serum cholesterols [76]. the environment trigger. The mechanism of neu-
CRP may also be useful in COPD [77]. Serum trophil recruitment of lung mediated by IL-6 is
levels of CRP are associated with health status of not fully elucidated, but there are data suggesting
COPD patients [78] and with all-cause, cardio- transcriptional activation of local chemokines
vascular, and cancer-specific mortality [79]. such as IL-8 through signal transducer and activa-
Elevated levels are also associated with increased tor of transcription-3 (STAT3). However, in cer-
risk for hospitalization, comorbidity, and mortal- tain circumstances, the IL-6 response can become
ity from COPD [80, 81]. During acute exacerba- dysregulated and spill into the systemic circula-
tions, CRP levels rise [82]. Oral glucocorticoid tion, causing acute deleterious effects in the vas-
therapy (e.g., prednisone at 30 mg/d for 2 weeks) cular system [95] and contribute to cardiovascular
reduces serum CRP levels [83]. However, serum dysfunction related to acute lung injury [95].
136 HI. Yoon

Given these properties of IL-6, IL-6 has been family of carbohydrate-binding proteins [104].
investigated as a possible biomarker in SP-D is composed of three polypeptide chains of
COPD. COPD patients in general have higher 43 kDa monomers, which aggregate to form a
serum IL-6 levels than those without COPD [96, stable helical trimeric structure. Each trimeric
97]. Serum levels of IL-6 were significantly structure contains four major domains: a cysteine-­
related to mortality in COPD patients [98]. containing cross-linking domain, a carbohydrate
However, serum IL-6 as a biomarker is limited in recognition domain, a collagenous domain, and a
that it is not lung specific and is widely expressed peptide linking domain [105]. The main colla-
throughout the body. Thus, serum IL-6 levels are gens in the trimeric complex are hydroxylysine
poorly responsive to COPD medications and as and hydroxylysyl glycosides. In most cases, the
such are less than ideal biomarker in COPD. trimeric structures are further modified into a
complex quaternary cruciate structure, consisting
of four trimeric subunits that undergo disulfide
Serum Amyloid Protein A (SAA) cross-linking within their amino-terminal
domain, which results in a dodecamer. Under
Serum amyloid protein A (SAA) is another promis- oxidizing conditions (e.g., in the presence of cig-
ing biomarker of COPD exacerbation. This protein arette smoke), cysteine residues in the N-terminus
is known to be elevated in patients with exacerba- can become nitrosylated, leading to the disrup-
tion and decrease over time with recovery [99, 100]. tion of the multimeric structure into smaller tri-
mers or monomers [106], which unlike the
dodecamers are thought to be pro-inflammatory
Lung Predominant Proteins and are more likely to pass into the systemic
To enhance the specificity of biomarkers, recent SP-D translocates from the lung into systemic
investigations have focused on proteins that are circulation, which is dependent on several factors
mostly synthesized in the lungs. These include including rate of synthesis and the permeability
surfactant proteins and Clara cell protein-16 of the alveolar-capillary barrier (i.e., lung perme-
(CC-16). ability) [103]. With cigarette smoking, lung per-
meability increases. Thus, in smokers, the serum
level of SP-D is elevated compared to non-­
Surfactant Protein D smokers but the concentration in the bronchoal-
veolar lavage (BAL) fluid is reduced [107, 108].
Surfactant proteins are produced predominantly Similarly, with COPD, lung permeability
by type II pneumocytes. Together with phospho- increases independent of cigarette smoking
lipids they form pulmonary surfactants which act [109]. Thus, with COPD progression, lung
to reduce surface tension of alveoli preventing expression of SP-D decreases but the ratio of
atelectasis. There are four kinds of surfactant plasma SP-D to BAL SP-D increases [110].
proteins, A, B, C, and D. Of these, surfactant pro- Importantly, treatment of COPD patients with
tein D (SP-D) has been the most widely explored inhaled glucocorticoids with or without long-­
biomarker in COPD because, dissimilar to other acting beta-2 agonists or oral glucocorticoids for
surfactant proteins, it is hydrophilic and is well 4 weeks significantly decreased measurable
expressed in plasma [101]. serum SP-D levels [109], which in turn is associ-
Although it has some minor role in regulating ated with improved health status of these patients.
surface tension of alveoli, SP-D’s main function Serum SP-D levels may also identify patients at
is to modulate innate immunity [102, 103]. SP-D high risk of recurrent exacerbations [111, 112].
is a large multimeric calcium-dependent, collag- Together, these data suggest that SP-D is a prom-
enous glycoprotein that is part of the collectin ising modifiable biomarker in COPD.
9  Biomarkers of COPD 137

Clara Cell-Derived Protein (CC-16) response to the therapy more precisely. This is
especially true when thinking about the enor-
Clara cell secretory protein-16 (CCSP, cc-16, mous complexity of COPD in its pathogenesis
cc-10, uteroglobin) is a member of the secreto- and phenotype; the success of BODE index
globin family of secreted disulfide-bridged where multiple clinical factors were combined to
dimeric proteins [113]. It is produced almost predict its prognosis further raises this possibility
exclusively by non-ciliated Clara cells [114, [129]. In a recent study, selected panel of 24 bio-
115], and its main function is to protect the lungs markers correlated with some clinical markers of
against oxidative stress and carcinogenesis [114]. COPD [130]. But there is no report with regard to
Serum levels of CC-16 become elevated after validity of multiple biomarkers in evaluating
acute exposure to various environmental triggers drug response in this field as yet.
such as cigarette smoke, chlorine, and lipopoly-
saccharides [116]. They can also rise after ozone
exposure and can be suppressed by inhaled glu- Molecular Biomarkers
cocorticoids [117]. Interestingly, serum CC-16
levels are relatively low in obliterative bronchiol- Finally, there is a growing effort to find useful
itis, asthma, and in healthy smokers [118–120]. biomarker using gene expression profile. This
In patients with COPD, serum CC-16 levels are approach has already been successful in some
lower compared to smokers without COPD [121, other fields of medicine [131, 132]. In a recent
122]. It is uncertain, however, whether in COPD, publication, a set of 220 biomarkers was identi-
serum CC-16 levels are modifiable. fied and predicted disease in an independent data
set with 97% accuracy [133]. The ability to iden-
tify COPD-related molecular processes in gene
 ulmonary and Activation-­
P expression raises the possibility of serving as bio-
Regulated Chemokine (PARC/ markers of therapeutic response for novel and
CCL-18) existing COPD therapies [134].
Future research effort and regulatory approval
PARC/CCL-18 is protein that is mostly produced should include (1) the whole-OMIC high
by monocytes, macrophages, and dendritic cells throughput technique to identify and validate
in lungs [123]. Serum levels of PARC/CCL-18 useful biomarkers and (2) the evaluation of assays
are elevated in acute coronary syndrome and in validation of biomarkers in various cells/tis-
idiopathic pulmonary fibrosis [124–126]. Serum sues/samples and evaluation of bioinformatics
PARC/CCL-18 levels are also significantly ele- tools as suggested by Cazzola [12].
vated in COPD and associated with the risk of
cardiovascular hospitalization and mortality in Conclusion
mild-to-moderate disease and with total mortality To date, there is no universally accepted bio-
in more advanced diseases [127]. Importantly, marker in COPD. However, with the assembly
short-term uses of oral but not inhaled corticoste- of large cohorts and infusion of capital from
roids can down-regulated systemic expression of various sources including government agencies
PARC/CCL-18 levels in COPD [127, 128]. and industry, there is renewed hope of finding a
modifiable biomarker that will be useful in the
discovery of novel compounds to treat
Multiple Biomarkers COPD. Since COPD is a heterogeneous disor-
der with multiple different (but related) pheno-
Finally, it seems very likely that multiple bio- types, it is essential that future endeavors in
markers, compared to individual ones, can reflect biomarker discovery consider these pheno-
pathogenesis, natural course and prognosis, and types. Although biomarkers can originate from
138 HI. Yoon

Table 9.1  Summary of serum/plasma biomarkers of COPD

Biomarkers Lung function Exacerbation Mortality Modified by drug (s)
C-reactive protein Yes [135] Yes [82] Yes [79, 80] Yes [83]
Interleukin 6 (IL-6) Yes [129] Yes [130] Not known Yes [136]
Fibrinogen Yes [129, 137] Yes [138] Not known Not known
Tumor necrosis factor Not known Yes [82] Not known Not known
receptor 1
Eotaxin 2 Not known Yes [82] Not known Not known
Serum amyloid A Not known Yes [99] Not known Yes [99]
Interleukin 1 (IL-1) Not known Yes [82] Not known Not known
Surfactant protein D Yes [139] Yes [109, 111] Not known Yes [85, 109]
Clara cell protein Not known Not known Not known Yes [117]
Pulmonary and Yes [140] Yes [82] Yes [127] Yes [127]
chemokine (PARC)

any source, sputum and blood are the most 7. Hogg JC. Pathophysiology of airflow limitation
readily available sources and hence appear to in chronic obstructive pulmonary disease. Lancet.
be the most promising sites for large-­scale bio- 8. Dragonieri S, Tongoussouva O, Zanini A, Imperatori
marker discovery in COPD. However, clearer A, Spanevello A. Markers of airway inflammation
understanding of the pathophysiological role of in pulmonary diseases assessed by induced sputum.
the putative biomarkers, their variability, and Monaldi Arch Chest Dis. 2009;71(3):119–26.
9. Borrill ZL, Roy K, Singh D. Exhaled breath condensate
importantly their relationship to disease pro- biomarkers in COPD. Eur Respir J. 2008;32(2):472–86.
gression are crucial defining factors (Table 9.1). 10. He ZY, Ou LM, Zhang JQ, Bai J, Liu GN, Li MH,
et al. Effect of 6 months of erythromycin treatment
on inflammatory cells in induced sputum and exac-
erbations in chronic obstructive pulmonary disease.
Respiration. 2010;80(6):445–52.
References 11. Comandini A, Rogliani P, Nunziata A, Cazzola M,
Curradi G, Saltini C. Biomarkers of lung damage
1. Buist AS, McBurnie MA, Vollmer WM, Gillespie associated with tobacco smoke in induced sputum.
S, Burney P, Mannino DM, et al. International Respir Med. 2009;103(11):1592–613.
variation in the prevalence of COPD (the BOLD 12. Cazzola M, Novelli G. Biomarkers in COPD. Pulm
study): a population-based prevalence study. Lancet. Pharmacol Ther. 23(6):493–500.
2007;370(9589):741–50. 13. Ford PA, Durham AL, Russell RE, Gordon F,
2. Murray CJ, Lopez AD. Alternative projections of mor- Adcock IM, Barnes PJ. Treatment effects of low-
tality and disability by cause 1990-­2020: global burden dose theophylline combined with an inhaled cortico-
of disease study. Lancet. 1997;349(9064):1498–504. steroid in COPD. Chest. 2010;137(6):1338–44.
3. Mathers CD, Loncar D. Projections of global mor- 14. Sharafkhaneh A, Hanania NA, Kim V. Pathogenesis
tality and burden of disease from 2002 to 2030. of emphysema: from the bench to the bedside. Proc
PLoS Med. 2006;3(11):e442. Am Thorac Soc. 2008;5(4):475–7.
4. National Heart L, and Blood Institute. Morbidity and 15. Cowburn AS, Condliffe AM, Farahi N, Summers C,
mortality: chartbook on cardiovascular, lung, and Chilvers ER. Advances in neutrophil biology. Chest.
blood diseases. In: US Department of Health and 2008;134(3):606–12.
Human Services PHS, editor. National Institutes of 16. O'Donnell RA, Peebles C, Ward JA, Daraker A, Angco
Health; 2004. G, Broberg P, et al. Relationship between peripheral air-
5. Sin DD, Man SF. Steroids in COPD: still up in the way dysfunction, airway obstruction, and neutrophilic
air? Eur Respir J. 2010;35(5):949–51. inflammation in COPD. Thorax. 2004;59(10):837–42.
6. Rabe KF, Hurd S, Anzueto A, Barnes PJ, Buist SA, 17. Keatings VM, Collins PD, Scott DM, Barnes
Calverley P, et al. Global strategy for the diagnosis, PJ. Differences in interleukin-8 and tumor necrosis
management, and prevention of chronic obstructive factor-alpha in induced sputum from patients with
pulmonary disease: GOLD executive summary. Am chronic obstructive pulmonary disease or asthma.
J Respir Crit Care Med. 2007;176(6):532–55. Am J Respir Crit Care Med. 1996;153(2):530–4.
9  Biomarkers of COPD 139

18. Williams TJ, Jose PJ. Neutrophils in chronic obstruc- 32. Dupont LJ, Demedts MG, Verleden GM. Prospective
tive pulmonary disease. Novartis Found Symp. evaluation of the validity of exhaled nitric oxide for
2001;234:136–41, discussion 41–8. the diagnosis of asthma*. Chest. 2003;123(3):751–6.
19. Perng DW, Huang HY, Chen HM, Lee YC, Perng 33. Van Den Toorn LM, Overbeek SE, De Jongste
RP. Characteristics of airway inflammation and JC, Leman K, Hoogsteden HC, Prins J-B. Airway
bronchodilator reversibility in COPD: a potential inflammation is present during clinical remission
guide to treatment. Chest. 2004;126(2):375–81. of atopic asthma. Am J Respir Crit Care Med.
20. Willemse BW, ten Hacken NH, Rutgers B, Lesman-­ 2001;164(11):2107–13.
Leegte IG, Postma DS, Timens W. Effect of 1-year 34. Kharitonov SA, Donnelly LE, Montuschi P, Corradi
smoking cessation on airway inflammation in M, Collins JV, Barnes PJ. Dose-dependent onset and
COPD and asymptomatic smokers. Eur Respir cessation of action of inhaled budesonide on exhaled
J. 2005;26(5):835–45. nitric oxide and symptoms in mild asthma. Thorax.
21. Stanescu D, Sanna A, Veriter C, Kostianev S, 2002;57(10):889–96.
Calcagni PG, Fabbri LM, et al. Airways obstruction, 35. Jones SL, Kittelson J, Cowan JO, Flannery EM,
chronic expectoration, and rapid decline of FEV1 in Hancox RJ, McLachlan CR, et al. The predic-
smokers are associated with increased levels of spu- tive value of exhaled nitric oxide measurements in
tum neutrophils. Thorax. 1996;51(3):267–71. assessing changes in asthma control. Am J Respir
22. Barnes NC, Qiu YS, Pavord ID, Parker D, Davis Crit Care Med. 2001;164(5):738–43.
PA, Zhu J, et al. Antiinflammatory effects of sal- 36. Kharitonov SA, Barnes PJ. Exhaled markers of
meterol/fluticasone propionate in chronic obstruc- pulmonary disease. Am J Respir Crit Care Med.
tive lung disease. Am J Respir Crit Care Med. 2001;163(7):1693–722.
2006;173(7):736–43. 37. Verleden GM, Dupont LJ, Verpeut AC, Demedts
23. Celli BR, Thomas NE, Anderson JA, Ferguson MG. The effect of cigarette smoking on exhaled
GT, Jenkins CR, Jones PW, et al. Effect of phar- nitric oxide in mild steroid-naive asthmatics*. Chest.
macotherapy on rate of decline of lung function in 1999;116(1):59–64.
chronic obstructive pulmonary disease: results from 38. Kharitonov SA, Robbins RA, Yates D, Keatings V,
the TORCH study. Am J Respir Crit Care Med. Barnes PJ. Acute and chronic effects of cigarette
2008;178(4):332–8. smoking on exhaled nitric oxide. Am J Respir Crit
24. Lapperre TS, Snoeck-Stroband JB, Gosman MM, Care Med. 1995;152(2):609–12.
Jansen DF, van Schadewijk A, Thiadens HA, et al. 39. Brindicci C, Ito K, Resta O, Pride NB, Barnes
Effect of fluticasone with and without salmeterol on PJ, Kharitonov SA. Exhaled nitric oxide from
pulmonary outcomes in chronic obstructive pulmo- lung periphery is increased in COPD. Eur Respir
nary disease: a randomized trial. Ann Intern Med. J. 2005;26(1):52–9.
2009;151(8):517–27. 40. MacNee W, Rennard SI, Hunt JF, Edwards LD,
25. Gan WQ, Man SF, Sin DD. Effects of inhaled cor- Miller BE, Locantore NW, et al. Evaluation of
ticosteroids on sputum cell counts in stable chronic exhaled breath condensate pH as a biomarker for
obstructive pulmonary disease: a systematic review COPD. Respir Med. 2011;105(7):1037–45.
and a meta-analysis. BMC Pulm Med. 2005;5:3. 41. Biernacki WA, Kharitonov SA, Barnes PJ. Increased
26. Kanehara M, Yokoyama A, Tomoda Y, Shiota N, leukotriene B4 and 8-isoprostane in exhaled breath
Iwamoto H, Ishikawa N, et al. Anti-inflammatory condensate of patients with exacerbations of
effects and clinical efficacy of theophylline and COPD. Thorax. 2003;58(4):294–8.
tulobuterol in mild-to-moderate chronic obstruc- 42. Corhay J-L, Henket M, Nguyen D, Duysinx B,
tive pulmonary disease. Pulm Pharmacol Ther. Sele J, Louis R. Leukotriene B4 contributes to
2008;21(6):874–8. exhaled breath condensate and sputum neutrophil
27. Gronke L, Beeh KM, Cameron R, Kornmann O, Beier Chemotaxis in COPD. Chest. 2009;136(4):1047–54.
J, Shaw M, et al. Effect of the oral leukotriene B4 43. Mercken EM, Gosker HR, Rutten EP, Wouters EF,
receptor antagonist LTB019 on inflammatory sputum Bast A, Hageman GJ, et al. Systemic and pulmo-
markers in patients with chronic obstructive pulmonary nary oxidative stress after single-leg exercise in
disease. Pulm Pharmacol Ther. 2008;21(2):409–17. COPD. Chest. 2009;136(5):1291–300.
28. Singh D, Edwards L, Tal-Singer R, Rennard 44. Shifren A, Mecham RP. The stumbling block in lung
S. Sputum neutrophils as a biomarker in COPD: find- repair of emphysema: elastic fiber assembly. Proc
ings from the ECLIPSE study. Respir Res. 11(1):77. Am Thorac Soc. 2006;3(5):428–33.
29. Boorsma M, Lutter R, van de Pol MA, Out TA, 45. Pelosi P, Rocco PR. Effects of mechanical ventila-
Jansen HM, Jonkers RE. Long-term effects of tion on the extracellular matrix. Intensive Care Med.
budesonide on inflammatory status in COPD. COPD. 2008;34(4):631–9.
2008;5(2):97–104. 46. Abraham T, Hogg J. Extracellular matrix remodel-
30. Kharitonov SA, Barnes PJ. Exhaled biomarkers. ing of lung alveolar walls in three dimensional space
Chest. 2006;130(5):1541–6. identified using second harmonic generation and
31. Barnes PJ, Dweik RA, Gelb AF, Gibson PG, George multiphoton excitation fluorescence. J Struct Biol.
SC, Grasemann H, et al. Exhaled nitric oxide in pul- 2010;171(2):189–96.
monary diseases: a comprehensive review. Chest. 47. Kelleher CM, Silverman EK, Broekelmann T, Litonjua
2010;138(3):682–92. AA, Hernandez M, Sylvia JS, et al. A functional
140 HI. Yoon

mutation in the terminal exon of elastin in severe, 61. Ma S, Lin Y, Tartell L, Turino G. The effect of tiotro-
early-onset chronic obstructive pulmonary disease. pium therapy on markers of elastin degradation in
Am J Respir Cell Mol Biol. 2005;33(4):355–62. COPD. Respir Res. 2009;10(1):12.
48. Wendel DP, Taylor DG, Albertine KH, Keating 62. Dahl R, Titlestad I, Lindqvist A, Wielders P, Wray
MT, Li DY. Impaired distal airway development in H, Wang M, et al. Effects of an oral MMP-9 and -12
mice lacking elastin. Am J Respir Cell Mol Biol. inhibitor, AZD1236, on biomarkers in moderate/
2000;23(3):320–6. severe COPD: a randomised controlled trial. Pulm
49. Luisetti M, Ma S, Iadarola P, Stone PJ, Viglio S, Pharmacol Ther. 2012;25(2):169–77.
Casado B, et al. Desmosine as a biomarker of elas- 63. Weathington NM, van Houwelingen AH, Noerager
tin degradation in COPD: current status and future BD, Jackson PL, Kraneveld AD, Galin FS, et al. A
directions. Eur Respir J. 2008;32(5):1146–57. novel peptide CXCR ligand derived from extracellu-
50. Rosenbloom J. Biochemical/immunologic markers lar matrix degradation during airway inflammation.
of emphysema. Ann N Y Acad Sci. 1991;624:7–12. Nat Med. 2006;12(3):317–23.
51. Goldstein RA, Starcher BC. Urinary excretion of 64. Gaggar A, Jackson PL, Noerager BD, O’Reilly PJ,
elastin peptides containing desmosin after intratra- McQuaid DB, Rowe SM, et al. A novel proteolytic
cheal injection of elastase in hamsters. J Clin Invest. cascade generates an extracellular matrix-derived
1978;61(5):1286–90. chemoattractant in chronic neutrophilic inflamma-
52. Churg A, Dai J, Tai H, Xie C, Wright JL. Tumor tion. J Immunol. 2008;180(8):5662–9.
necrosis factor-alpha is central to acute cigarette 65. Laskin DL, Kimura T, Sakakibara S, Riley DJ, Berg
smoke-induced inflammation and connective tis- RA. Chemotactic activity of collagen-like poly-
sue breakdown. Am J Respir Crit Care Med. peptides for human peripheral blood neutrophils.
2002;166(6):849–54. J Leukoc Biol. 1986;39(3):255–66.
53. Churg A, Wang RD, Tai H, Wang X, Xie C, Wright 66. Postlethwaite AE, Kang AH. Collagen-and collagen
JL. Tumor necrosis factor-alpha drives 70% of ciga- peptide-induced chemotaxis of human blood mono-
rette smoke-induced emphysema in the mouse. Am cytes. J Exp Med. 1976;143(6):1299–307.
J Respir Crit Care Med. 2004;170(5):492–8. 67. Houghton AM, Quintero PA, Perkins DL, Kobayashi
54. Huang JT-J, Chaudhuri R, Albarbarawi O, Barton DK, Kelley DG, Marconcini LA, et al. Elastin frag-
A, Grierson C, Rauchhaus P, et al. Clinical valid- ments drive disease progression in a murine model
ity of plasma and urinary desmosine as biomarkers of emphysema. J Clin Invest. 2006;116(3):753–9.
for chronic obstructive pulmonary disease. Thorax. 68. van Houwelingen AH, Weathington NM, Verweij
2012;67(6):502–8. V, Blalock JE, Nijkamp FP, Folkerts G. Induction
55. Rabinovich RA, Miller BE, Wrobel K, Ranjit K, of lung emphysema is prevented by L-arginine-­
Williams MC, Drost E, et al. Circulating desmosine threonine-arginine. FASEB J. 2008;22(9):3403–8.
levels do not predict emphysema progression but are 69. O’Reilly P, Jackson PL, Noerager B, Parker S,
associated with cardiovascular risk and mortality in Dransfield M, Gaggar A, et al. N-alpha-PGP and
COPD. Eur Respir J. 2016;47(5):1365–73. PGP, potential biomarkers and therapeutic targets
56. Fiorenza D, Viglio S, Lupi A, Baccheschi J, Tinelli for COPD. Respir Res. 2009;10:38.
C, Trisolini R, et al. Urinary desmosine excretion in 70. O’Reilly PJ, Jackson PL, Wells JM, Dransfield MT,
acute exacerbations of COPD: a preliminary report. Scanlon PD, Blalock JE. Sputum PGP is reduced
Respir Med. 2002;96(2):110–4. by azithromycin treatment in patients with COPD
57. Ma S, Lin YY, Turino GM. Measurements of des- and correlates with exacerbations. BMJ Open.
mosine and isodesmosine by mass spectrometry in 2013;3(12):e004140.
COPD*. Chest. 2007;131(5):1363–71. 71. Wells JM, Jackson PL, Viera L, Bhatt SP, Gautney J,
58. Cocci F, Miniati M, Monti S, Cavarra E, Gambelli Handley G, et al. A randomized, placebo-controlled
F, Battolla L, et al. Urinary desmosine excretion is trial of Roflumilast. Effect on proline-glycine-­
inversely correlated with the extent of emphysema in proline and neutrophilic inflammation in chronic
patients with chronic obstructive pulmonary disease. obstructive pulmonary disease. Am J Respir Crit
Int J Biochem Cell Biol. 2002;34(6):594–604. Care Med. 2015;192(8):934–42.
59. Stone PJ, Morris TA 3rd, Franzblau C, Snider 72. Snelgrove RJ, Jackson PL, Hardison MT, Noerager
GL. Preliminary evidence that augmentation therapy BD, Kinloch A, Gaggar A, et al. A critical role for
diminishes degradation of cross-linked elastin in LTA4H in limiting chronic pulmonary neutrophilic
alpha-1-antitrypsin-deficient humans. Respiration. inflammation. Science. 330(6000):90–4.
1995;62(2):76–9. 73. van Eeden SF, Sin DD. Chronic obstructive pulmo-
60. Gottlieb DJ, Luisetti M, Stone PJ, Allegra L, nary disease: a chronic systemic inflammatory dis-
Cantey-Kiser JM, Grassi C, et al. Short-term supple- ease. Respiration. 2008;75(2):224–38.
mentation therapy does not affect elastin degrada- 74. Danesh J, Collins R, Appleby P, Peto R. Association of
tion in severe alpha(1)-antitrypsin deficiency. The fibrinogen, C-reactive protein, albumin, or leukocyte
American-Italian AATD study group. Am J Respir count with coronary heart disease: meta-analyses of
Crit Care Med. 2000;162(6):2069–72. prospective studies. JAMA. 1998;279(18):1477–82.
9  Biomarkers of COPD 141

75. Kaptoge S, Di Angelantonio E, Lowe G, Pepys MB, markers of inflammation and incidence of hospitali-
Thompson SG, Collins R, et al. C-reactive protein sations for COPD: results from a population-based
concentration and risk of coronary heart disease, cohort study. Thorax. 2009;64(3):211–5.
stroke, and mortality: an individual participant meta-­ 89. Duvoix A, Dickens J, Haq I, Mannino D, Miller B,
analysis. Lancet. 2010;375(9709):132–40. Tal-Singer R, et al. Blood fibrinogen as a biomarker
76. Ridker PM, Danielson E, Fonseca FA, Genest of chronic obstructive pulmonary disease. Thorax.
J, Gotto AM Jr, Kastelein JJ, et al. Rosuvastatin 2013;68(7):670–6.
to prevent vascular events in men and women 90. Valvi D, Mannino DM, Müllerova H, Tal-Singer
with elevated C-reactive protein. N Engl J Med. R. Fibrinogen, chronic obstructive pulmonary dis-
2008;359(21):2195–207. ease (COPD) and outcomes in two United States
77. Karadag F, Kirdar S, Karul AB, Ceylan E. The value cohorts. Int J COPD. 2012;7:173–82.
of C-reactive protein as a marker of systemic inflam- 91. Dentener MA, Creutzberg EC, Pennings HJ, Rijkers
mation in stable chronic obstructive pulmonary dis- GT, Mercken E, Wouters EF. Effect of infliximab on
ease. Eur J Intern Med. 2008;19(2):104–8. local and systemic inflammation in chronic obstruc-
78. Garrod R, Marshall J, Barley E, Fredericks S, tive pulmonary disease: a pilot study. Respiration.
Hagan G. The relationship between inflamma- 2008;76(3):275–82.
tory markers and disability in chronic obstructive 92. Kaczmarek P, Sladek K, Skucha W, Rzeszutko M,
pulmonary disease (COPD). Prim Care Respir Iwaniec T, Dziedzina S, et al. The influence of simv-
J. 2007;16(4):236–40. astatin on selected inflammatory markers in patients
79. Man SFP, Connett JE, Anthonisen NR, Wise RA, with chronic obstructive pulmonary disease. Pol
Tashkin DP, Sin DD. C-reactive protein and mortal- Arch Med Wewn. 2010;120(1–2):11–7.
ity in mild to moderate chronic obstructive pulmo- 93. Kishimoto T. IL-6: from its discovery to clinical
nary disease. Thorax. 2006;61(10):849–53. applications. Int Immunol. 2010;22(5):347–52.
80. Dahl M, Vestbo J, Lange P, Bojesen SE, Tybjaerg-­ 94. Ishii H, Hayashi S, Hogg JC, Fujii T, Goto Y,
Hansen A, Nordestgaard BG. C-reactive protein Sakamoto N, et al. Alveolar macrophage-epithelial
as a predictor of prognosis in chronic obstructive cell interaction following exposure to atmospheric
pulmonary disease. Am J Respir Crit Care Med. particles induces the release of mediators involved
2007;175(3):250–5. in monocyte mobilization and recruitment. Respir
81. Thomsen M, Dahl M, Lange P, Vestbo J, Nordestgaard Res. 2005;6:87.
BG. Inflammatory biomarkers and comorbidities in 95. Kido T, Tamagawa E, Bai N, Suda K, Yang HH, Li
chronic obstructive pulmonary disease. Am J Respir Y, et al. Particulate matter induces translocation of
Crit Care Med. 2012;186(10):982–8. IL-6 from the lung to the systemic circulation. Am
82. Hurst JR, Donaldson GC, Perera WR, Wilkinson J Respir Cell Mol Biol. 2011;44(2):197–204.
TMA, Bilello JA, Hagan GW, et al. Use of plasma 96. Thorleifsson SJ, Margretardottir OB, Gudmundsson
biomarkers at exacerbation of chronic obstructive G, Olafsson I, Benediktsdottir B, Janson C, et al.
pulmonary disease. Am J Respir Crit Care Med. Chronic airflow obstruction and markers of sys-
2006;174(8):867–74. temic inflammation: results from the BOLD study in
83. Sin DD, Lacy P, York E, Man SFP. Effects of flutica- Iceland. Respir Med. 2009;103(10):1548–53.
sone on systemic markers of inflammation in chronic 97. Walter RE, Wilk JB, Larson MG, Vasan RS, Keaney
obstructive pulmonary disease. Am J Respir Crit JF Jr, Lipinska I, et al. Systemic inflammation
Care Med. 2004;170(7):760–5. and COPD: the Framingham heart study. Chest.
84. Perng DW, Tao CW, Su KC, Tsai CC, Liu LY, Lee 2008;133(1):19–25.
YC. Anti-inflammatory effects of salmeterol/fluti- 98. Celli BR, Locantore N, Yates J, Tal-Singer R,
casone, tiotropium/fluticasone or tiotropium in Miller BE, Bakke P, et al. Inflammatory biomarkers
COPD. Eur Respir J. 2009;33(4):778–84. improve clinical prediction of mortality in chronic
85. Sin DD, Man SF, Marciniuk DD, Ford G, FitzGerald obstructive pulmonary disease. Am J Respir Crit
M, Wong E, et al. The effects of fluticasone with or Care Med. 2012;185(10):1065–72.
without salmeterol on systemic biomarkers of inflam- 99. Bozinovski S, Hutchinson A, Thompson M,
mation in chronic obstructive pulmonary disease. MacGregor L, Black J, Giannakis E, et al. Serum
Am J Respir Crit Care Med. 2008;177(11):1207–14. amyloid a is a biomarker of acute exacerbations of
86. Fibrinogen SC. Plasma fibrinogen level and the risk chronic obstructive pulmonary disease. Am J Respir
of major cardiovascular diseases and nonvascular Crit Care Med. 2008;177(3):269–78.
mortality. JAMA. 2005;294(14):1799–809. 100. Koutsokera A, Kiropoulos TS, Nikoulis DJ, Daniil
87. Groenewegen KH, Postma DS, Hop WCJ, Wielders ZD, Tsolaki V, Tanou K, et al. Clinical, functional and
PLML, Schölsser NJ, Wouters EF. Increased sys- biochemical changes during recovery from COPD
temic inflammation is a risk factor for COPD exac- exacerbations. Respir Med. 2009;103(6):919–26.
erbations*. Chest. 2008;133(2):350–7. 101. Cole FS. Surfactant protein B: unambiguously nec-
88. Engstrom G, Segelstorm N, Ekberg-Aronsson M, essary for adult pulmonary function. Am J Physiol
Nilsson PM, Lindgarde F, Lofdahl CG. Plasma Lung Cell Mol Physiol. 2003;285(3):L540–L2.
142 HI. Yoon

102. Hartl D, Griese M. Surfactant protein D in human tects against ozone-induced airway inflamma-
lung diseases. Eur J Clin Invest. 2006;36(6):423–35. tion and modified immune cell activation markers
103. Sin DD, Pahlavan PS, Man SFP. Review: surfactant in healthy volunteers. Environ Health Perspect.
protein D: a lung specific biomarker in COPD? Ther 2008;116(6):799–805.
Adv Respir Dis. 2008;2(2):65–74. 118. Shijubo N, Itoh Y, Yamaguchi T, Shibuya Y, Morita
104. Kishore U, Greenhough TJ, Waters P, Shrive AK, Y, Hirasawa M, et al. Serum and BAL Clara cell 10
Ghai R, Kamran MF, et al. Surfactant proteins SP-A kDa protein (CC10) levels and CC10-positive bron-
and SP-D: structure, function and receptors. Mol chiolar cells are decreased in smokers. Eur Respir
Immunol. 2006;43(9):1293–315. J. 1997;10(5):1108–14.
105. Crouch EC. Surfactant protein-D and pulmonary 119. Shijubo N, Itoh Y, Yamaguchi T, Imada A, Hirasawa
host defense. Respir Res. 2000;1(2):93–108. M, Yamada T, et al. Clara cell protein-positive epi-
106. Guo CJ, Atochina-Vasserman EN, Abramova E, thelial cells are reduced in small airways of asthmat-
Foley JP, Zaman A, Crouch E, et al. S-nitrosylation ics. Am J Respir Crit Care Med. 1999;160(3):930–3.
of surfactant protein-D controls inflammatory func- 120. Mattsson J, Remberger M, Andersson O, Sundberg
tion. PLoS Biol. 2008;6(11):e266. B, Nord M. Decreased serum levels of clara cell
107. More JM, Voelker DR, Silveira LJ, Edwards MG, secretory protein (CC16) are associated with bron-
Chan ED, Bowler RP. Smoking reduces surfactant chiolitis obliterans and may permit early diagnosis
protein D and phospholipids in patients with and in patients after allogeneic stem-cell transplantation.
without chronic obstructive pulmonary disease. Transplantation. 2005;79(10):1411–6.
BMC Pulm Med. 2010;10:53. 121. Tsoumakidou M, Bouloukaki I, Thimaki K, Tzanakis
108. Honda Y, Takahashi H, Kuroki Y, Akino T, Abe N, Siafakas NM. Innate immunity proteins in chronic
S. Decreased contents of surfactant proteins a obstructive pulmonary disease and idiopathic pul-
and D in BAL fluids of healthy smokers. Chest. monary fibrosis. Exp Lung Res. 2010;36(6):373–80.
1996;109(4):1006–9. 122. Lomas DA, Silverman EK, Edwards LD, Miller BE,
109. Lomas DA, Silverman EK, Edwards LD, Locantore Coxson HO, Tal-Singer R, et al. Evaluation of serum
NW, Miller BE, Horstman DH, et al. Serum sur- CC-16 as a biomarker for COPD in the ECLIPSE
factant protein D is steroid sensitive and associ- cohort. Thorax. 2008;63(12):1058–63.
ated with exacerbations of COPD. Eur Respir 123. Günther C, Bello-Fernandez C, Kopp T, Kund J,
J. 2009;34(1):95–102. Carballido-Perrig N, Hinteregger S, et al. CCL18
110. Tkacova R, McWilliams A, Lam S, Sin DD. Integrating is expressed in atopic dermatitis and mediates skin
lung and plasma expression of pneumo-proteins in homing of human memory T cells. J Immunol.
developing biomarkers in COPD: a case study of surfac- 2005;174(3):1723–8.
tant protein D. Med Sci Monit. 2010;16(11):CR540–4. 124. Prasse A, Probst C, Bargagli E, Zissel G, Toews GB,
111. Shakoori TA, Sin DD, Ghafoor F, Bashir S, Bokhari Flaherty KR, et al. Serum CC-chemokine ligand
SN. Serum surfactant protein D during acute exac- 18 concentration predicts outcome in idiopathic
erbations of chronic obstructive pulmonary disease. pulmonary fibrosis. Am J Respir Crit Care Med.
Dis Markers. 2009;27(6):287–94. 2009;179(8):717–23.
112. Hurst JR, Vestbo J, Anzueto A, Locantore N, 125. Prasse A, Pechkovsky DV, Toews GB, Schäfer M,
Mullerova H, Tal-Singer R, et al. Susceptibility to Eggeling S, Ludwig C, et al. CCL18 as an indica-
exacerbation in chronic obstructive pulmonary dis- tor of pulmonary fibrotic activity in idiopathic inter-
ease. N Engl J Med. 2010;363(12):1128–38. stitial pneumonias and systemic sclerosis. Arthritis
113. Umland TC, Swaminathan S, Singh G, Warty V, Rheum. 2007;56(5):1685–93.
Furey W, Pletcher J, et al. Structure of a human Clara 126. Kraaijeveld AO, de Jager SCA, de Jager WJ,
cell phospholipid-binding protein-ligand complex at Prakken BJ, McColl SR, Haspels I, et al. CC che-
1.9 A resolution. Nat Struct Biol. 1994;1(8):538–45. mokine ligand-5 (CCL5/RANTES) and CC chemo-
114. Broeckaert F, Bernard A. Clara cell secretory kine ligand-18 (CCL18/PARC) are specific markers
protein (CC16): characteristics and perspectives of refractory unstable angina pectoris and are tran-
as lung peripheral biomarker. Clin Exp Allergy. siently raised during severe ischemic symptoms.
2000;30(4):469–75. Circulation. 2007;116(17):1931–41.
115. Yoneda K. Ultrastructural localization of phospho- 127. Sin DD, Miller B, Duvoix A, Man SFP, Zhang X,
lipases in the Clara cell of the rat bronchiole. Am Silverman EK, et al. Serum PARC/CCL-18 concen-
J Pathol. 1978;93(3):745–52. trations and health outcomes in chronic obstructive
116. Lakind JS, Holgate ST, Ownby DR, Mansur AH, pulmonary disease. Am J Respir Crit Care Med.
Helms PJ, Pyatt D, et al. A critical review of the use 2011;183(9):1187–92.
of Clara cell secretory protein (CC16) as a biomarker 128. Man SF, Xuekui Z, Vessey R, Walker T, Lee K,
of acute or chronic pulmonary effects. Biomarkers. Park D, et al. The effects of inhaled and oral cor-
2007;12(5):445–67. ticosteroids on serum inflammatory biomarkers in
117. Alexis NE, Lay JC, Haczku A, Gong H, Linn W, COPD: an exploratory study. Ther Adv Respir Dis.
Hazucha MJ, et al. Fluticasone propionate pro- 2009;3(2):73–80.
9  Biomarkers of COPD 143

129. Donaldson GC, Seemungal TAR, Patel IS, Bhowmik between ­pulmonary function and C-reactive p­ rotein
A, Wilkinson TMA, Hurst JR, et al. Airway and sys- in apparently healthy subjects. Am J Respir Crit
temic inflammation and decline in lung function in Care Med. 2006;174(6):626–32.
patients with COPD*. Chest. 2005;128(4):1995–2004. 136. Cosio BG, Iglesias A, Rios A, Noguera A, Sala E,
130. Karadag F, Karul A, Cildag O, Yilmaz M, Ozcan Ito K, et al. Low-dose theophylline enhances the
H. Biomarkers of systemic inflammation in sta- anti-­inflammatory effects of steroids during exacer-
ble and exacerbation phases of COPD. Lung. bations of COPD. Thorax. 2009;64(5):424–9.
2008;186(6):403–9. 137. Dahl M, Tybjoerg-Hansen A, Vestbo J, Lange P,
131. Toedter G, Li K, Marano C, Ma K, Sague S, Huang Nordestgaard BG. Elevated plasma fibrinogen associ-
CC, et al. Gene expression profiling and response ated with reduced pulmonary function and increased
signatures associated with differential responses risk of chronic obstructive pulmonary disease. Am
to infliximab treatment in ulcerative colitis. Am J Respir Crit Care Med. 2001;164(6):1008–11.
J Gastroenterol. 2011;106(7):1272–80. 138. Wedzicha JA, Seemungal TA, MacCallum PK,
132. Elashoff MR, Wingrove JA, Beineke P, Daniels SE, Paul EA, Donaldson GC, Bhowmik A, et al. Acute
Tingley WG, Rosenberg S, et al. Development of a exacerbations of chronic obstructive pulmonary
blood-based gene expression algorithm for assess- disease are accompanied by elevations of plasma
ment of obstructive coronary artery disease in non-­ fibrinogen and serum IL-6 levels. Thromb Haemost.
diabetic patients. BMC Med Genet. 2011;4(1):26. 2000;84(2):210–5.
133. Bhattacharya S, Srisuma S, DeMeo DL, Shapiro SD, 139.
Tkacova R, McWilliams A, Lam S, Sin
Bueno R, Silverman EK, et al. Molecular biomark- DD. Integrating lung and plasma expression of
ers for quantitative and discrete COPD phenotypes. pneumo-proteins in developing biomarkers in
Am J Respir Cell Mol Biol. 2009;40(3):359–67. COPD: a case study of surfactant protein D. Med Sci
134. Silverman EK, Spira A, Pare PD. Genetics and Monit. 16(11):CR540–4.
genomics of chronic obstructive pulmonary disease. 140. Pinto-Plata V, Toso J, Lee K, Park D, Bilello J,
Proc Am Thorac Soc. 2009;6(6):539–42. Mullerova H, et al. Profiling serum biomarkers
135. Aronson D, Roterman I, Yigla M, Kerner A, in patients with COPD: associations with clinical
Avizohar O, Sella R, et al. Inverse association parameters. Thorax. 2007;62(7):595–601.
Part III
Phenotypes of COPD
Jamie Sheth and MeiLan Han

Introduction that describe differences between individuals

with COPD as they relate to clinically meaning-
Significant heterogeneity exists in chronic ful outcomes (symptoms, exacerbations, response
obstructive pulmonary disease (COPD) patients to therapy, rate of disease progression or death)”
with respect to clinical presentation, physiology, [1]. In many cases, the discovery and validation
imaging characteristics, response to therapy, dis- of phenotypes occurs through a multi-step pro-
ease progression, and ultimately survival [1]. cess (Fig. 10.1) where for instance a unique
While FEV1 correlates to many outcomes of group of patients may initially be identified
interest, no single measure adequately reflects through a biological or molecular signature and
this disease complexity. The goal of phenotyping validated by demonstrating a similar response to
is to identify patient subgroups with unique char- therapy [2].
acteristics with the ultimate goal of altering clini- What follows in this chapter is a discussion of
cally meaningful outcomes through targeted the “current state” for putative COPD phenotypes
therapeutic approaches. While various defini- and data to support a relationship between pro-
tions have been described, a consensus definition posed phenotypes and clinically meaningful met-
that was recently proposed defines a phenotype rics including prognosis, symptoms, and
as, “a single or combination of disease attributes outcomes in COPD.

J. Sheth, M.D. (*) • M. Han, M.D., M.S.

Division of Pulmonary and Critical Care,
University of Michigan, Ann Arbor, MI, USA

© Springer-Verlag Berlin Heidelberg 2017 147

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_10
148 J. Sheth and M. Han

Clinical phenotype
defined by similar

Validation of
Symptomatic, physiologic,
molecular marker or
and/or radiologic
determination of
of phenotype
response in target

+/– Development Biologic or molecular

of therapy characterization
of phenotype

Fig. 10.1  Ideal phenotyping construct wherein candidate define a subpopulation that leads to the identification of a
phenotypes are validated once their relevance to clinical biologic target and focused therapy. Alternatively, the pro-
outcomes is established. There are multiple potential cess might begin with the differentiation of subgroups
points of entry into this iterative process of phenotype based on a biologic marker that is then validated by simi-
identification. For instance, similar clinical outcomes may lar clinical response within subgroups [1]

Clinically Defined Phenotypes achieved sustained smoking cessation [6] with

smokers losing lung function in a dose-dependent
 obacco Smoke-Associated, Biomass
T manner [7].
Smoke-Associated, and Non-smoking Compared to nonsmokers, COPD in smokers
COPD has been associated with greater amounts of
emphysema, impaired gas exchange and greater
Based on the GOLD definition, COPD is charac- amounts of chronic sputum production [8].
terized by persistent airflow limitation that is usu- Classically, cigarette smoke-related COPD is
ally progressive and associated with an enhanced manifest by upper lobe predominant centrilobu-
chronic inflammatory response in the airways lar emphysema though a variety of subtypes have
and the lung thought to be the result of exposure been described. As an example, the COPDGene
to noxious particles or gases [3]. Tobacco is the study has identified four subtypes of smokers
principal risk factor and environmental toxin with distinct patterns of airway disease and
responsible for disease in most patients with emphysema that are strongly associated with
COPD. The prevalence of COPD in smokers is COPD-related clinical characteristics including
approximately 20% as compared to approxi- exacerbations and dyspnea [9]. The first cluster is
mately 4% in nonsmokers [4]. Tobacco smoke the “relatively resistant smoker” characterized by
exposure leads to decreases in expiratory flow heavy smoking exposure with no or minimal air-
through two distinct pathophysiological pro- flow obstruction. Second is “mild upper zone-­
cesses: the emphysematous destruction of the predominant emphysema” characterized by mild
lung parenchyma and/or narrowing and oblitera- airflow obstruction and mild emphysema which
tion of the small peripheral airways [5]. The was also found to have a strong genetic associa-
Lung Health Study (an interventional smoking tion with single-nucleotide polymorphism (SNP)
cessation study in smokers with COPD with a rs 1980057 near the HHIP gene. Cluster 3 repre-
mild degree of airflow obstruction) demonstrated sents “airway-predominant disease” character-
that the rate of FEV1 (forced expiratory volume ized by thicker airway walls, lowest average
in one second) decline was greatest in patients emphysema of all clusters and high BMI. Cluster
who smoked the most and least in those who 4, “severe emphysema,” is characterized by high
10  Phenotypes of COPD 149

emphysema, gas trapping, and severe airflow biomass-exposed individuals demonstrate more
obstruction. It is associated with the lowest BMI, air trapping, peri-bronchial thickening and less
highest lifetime pack-years exposure, and oldest emphysema than the tobacco-exposed group,
average age. A strong genetic association is seen suggesting an airway-predominant phenotype
with SNP rs8034191 in the chromosome 15q [12–14]. The early exposure and repeated respi-
locus that includes the nicotinic receptor genes ratory infections associated with biomass smoke
[9]. Clearly, tobacco cessation should be a pre- exposure may alter the structure and function of
dominant focus of treatment plan in all patients the airway walls and may predispose biomass
although we do not know whether the smoking smoke-exposed individuals to a different COPD
cessation approach should be tailored to a phenotype as adults compared to tobacco smok-
patient’s genetic variation in SNPs associated ers who may begin smoking at an older age.
with nicotine dependence. One problem with this Chemical composition, age of exposure, and
type of approach is that it can be difficult to sepa- inhalation pattern have all been suggested as fac-
rate differences due to differences exposure, dis- tors that contribute to variable phenotypes
ease severity, and disease biology. between tobacco smoke- and biomass smoke-­
A substantial proportion of COPD cases can- related COPD [12]. Compared to tobacco smoke
not be explained by smoking, especially among COPD, those with disease related to biomass
younger persons, females, and residents of devel- smoke exposure have more cough, sputum, and
oping countries [10]. In multiple p­ opulation-­based air trapping on CT scan.
studies, approximately 30% of all cases of COPD
occur in never-smokers [8]. Factors indepen-
dently associated with COPD in nonsmokers Alpha-1 Antitrypsin Deficiency
include increasing age, a diagnosis of asthma,
and severe childhood respiratory disease. Severe alpha-1 antitrypsin deficiency (AATD) is a
Secondhand (also known as environmental) and well-established genetic risk factor for COPD in
biomass smoke exposure are additional gender-­ both smokers and nonsmokers. Alpha-1 antitryp-
specific risk factors for women in non-smoking-­ sin (AAT) is a protease that inactivates neutrophil
related COPD [8]. For women in the developing elastase. High-risk genotypes include S, Z, and
world, where fuels such as coal and biomass are null alleles. It has been estimated to be responsi-
used for indoor cooking and heating, several ble for approximately 2–3% of cases of COPD
studies have reported an association between [15]. The classic clinical phenotype is a young
exposure to biomass smoke and COPD [11]. COPD patient, often a smoker, with lower lobe
Alternative risk factors for COPD include genetic predominant emphysema and a family history of
factors, outdoor air pollution, secondhand smoke emphysema. Chest imaging frequently demon-
exposure, occupational exposures, diet, and strates a predominantly lower lobe distribution of
tuberculosis [10]. Respiratory symptoms such as emphysema with a panacinar pattern, different
chronic cough, chronic phlegm, wheeze, and than the more common centriacinar pattern. In
exertional dyspnea are features of COPD regard- subjects homozygous for the AAT Z allele, ciga-
less of smoking status though are more frequent rette smoking leads to a markedly increased risk
in ever-smokers [8]. However, the burden of of COPD and reduced survival. Non-smoking PI
COPD exacerbations has been shown to be Z subjects are also at increased risk for develop-
equally prevalent, occurring in approximately ing COPD although to a lesser degree [16–19].
30% of ever-smokers and never-smokers with While the classic homozygous Z phenotype is
COPD [8]. While nonsmokers exhibit a similar associated with early-onset lower lung predomi-
respiratory symptoms profile to that seen in nant, the heterozygous SZ patients are less sus-
smokers, they demonstrate different radiologic ceptible to tobacco smoke and may have a clinical
and physiologic presentations. phenotype more consistent with “usual” COPD
In studies comparing radiologic phenotypes demonstrating upper lobe predominant emphy-
between tobacco smoked-exposed and biomass sema [20]. Currently, treatment for AATD incor-
smoke-exposed patients matched for lung function, porates not only bronchodilator therapy but also a
150 J. Sheth and M. Han

mechanism-directed treatment approach using and 40%, depending on how it is defined. The
clinical presentation in addition to serum AAT most well-recognized definition for chronic bron-
concentration, AAT protein phenotyping, and chitis is the presence of chronic cough and ­sputum
AAT genotyping to guide decision-­ making for production for 3 months a year, for 2 consecutive
treatment with AAT replacement therapy [21]. years [22]. Chronic bronchitis is also present in
the non-COPD population with cigarette smoking
as a well-established shared risk factor for both
Chronic Bronchitis obstructed and non-obstructed individuals [23].
Table 10.1 summarizes studies examining preva-
Chronic bronchitis in the COPD population has lence of chronic bronchitis in both the COPD and
been estimated to be between approximately 7 non-COPD patient populations.

Table 10.1  Summary of studies estimating the prevalence of chronic bronchitis [33]
Study Patients Findings
Lange et al., 1989 [34] General population, Copenhagen; Bronchial hypersecretion: 10.1%
12,698 adults
Sobradillo et al., 1999 [35] General population, Spain; 4035 Cough: 13.5%
adults aged 40–69 years Expectoration 10.7%
Chronic bronchitis 4.8%
Pallasaho et al., 1999 [36] Random sample, Finland; 8000 Productive cough: 27%
patients aged 20–69 years
Von Hertzen et al., 2000 [37] Random patients, Finland; 7217 Chronic bronchitis and/or emphysema:
patients aged >30 years 22% in men, 7% in women
Cerveri et al., 2001 [38] General population, Europe; 17,966 Chronic bronchitis: 2.6% (range
patients aged 20–44 years 0.7–9.7% across countries)
Janson et al., 2001 [39] Multinational; 18,277 patients aged Productive cough: 10.2%
20–48 years
Huchon et al., 2002 [40] General population, France; 14,076 Chronic bronchitis: 4.1%
patients Chronic cough and/or expectoration:
Lundback et al., 2003 [41] 5892 patients from OLIN study Chronic productive cough: 60% in
cohort COPD patients
Miravitlles et al., 2006 [42] General population, Spain; 6758 Cough: 5% in never-smokers, 11% in
adults aged >40 years smokers or ex-smokers
Expectoration: 4% in never-smokers,
11% in smokers and ex-smokers
Pelkonen et al., 2006 [43] Finnish cohort of 1711 adult men Incidence of chronic productive cough:
aged 40–59 years 42% current smokers, 26% past
smokers, 22% never-smokers
De Marco et al., 2007 [44] International cohort of 5002 patients Chronic cough/phlegm production:
aged 20–44 years with normal lung 9.2%
Miravitlles et al., 2009 [45] Population-based sample, Spain; Chronic cough: 3.4%
4274 adults aged 40–80 years Chronic sputum production: 11.7%
Harmsen et al., 2010 [46] Danish cohort of 29,180 (in 1994) Cumulative prevalence of chronic
and 21,130 (in 2004) twins aged mucus secretion over 10 years of study,
12–41 years 10.7% in women and 8.7% in men
Kim et al., 2011 [47] US cohort of 1061 adults current or Chronic bronchitis: 27.3%
former smokers with COPD
Martinez et al., 2014 [48] US cohort of 5858 adults past or Chronic bronchitis: 34.6%
previous smokers without airflow
10  Phenotypes of COPD 151

Biomass smoke exposure may also contribute these patients are still evolving. Proposed defini-
[12–14]. Genetic factors are likely at play as tions vary in complexity from “airflow obstruc-
genome studies have identified a single-­nucleotide tion that is not completely reversible, accompanied
polymorphism associated with chronic mucus hyper- by symptoms or signs of increased obstruction
secretion on chromosome 3 [24]. Radiographically, reversibility” [49] to the presence of discrete
chronic bronchitis is associated with greater airway major and minor criteria [50]. Clinically, patients
disease manifest by higher bronchial wall thick- with ACOS tend to be younger than those with
ness-to-diameter ratios [25]. COPD but older than those with “pure” asthma
In the COPDGene cohort, in addition to cur- and have less smoking history than typically seen
rent smoking, allergic rhinitis, acute bronchitis, in traditional COPD [51]. They often demonstrate
asthma, male gender, and Caucasian race (as features associated with asthma including wheez-
opposed to African American race) were also ing, atopy, elevated total immunoglobulin (Ig) E
associated with clinical phenotype of chronic levels, and allergic problems such as allergic rhi-
bronchitis [26]. In this analysis, COPD subjects nitis and hay fever [52, 53]. Airway inflamma-
with chronic bronchitis were also more likely to tion tends to be more eosinophilic than the usual
have allergic nasal and ocular symptoms, higher neutrophilic inflammation described in COPD
exacerbation frequency, worse health-related patients [54]. Pulmonary function testing also
quality of life, and lower 6-minute walk distance reveals a higher carbon monoxide diffusing
as compared to COPD patients without chronic capacity (DLCO) and more pronounced acute
bronchitis [26]. Most importantly, identification bronchodilator response [54, 55]. On HRCT,
of the chronic bronchitis phenotype has impor- patients clinically identified with ACOS have
tant therapeutic implications. more air trapping, less emphysema, and greater
Roflumilast, an oral phosphodiesterase-4 bronchial wall thickening [52, 56].
inhibitor, has been found to be most effective in Part of the difficulty in creating criteria to
patients with a chronic bronchitis phenotype and identify these overlap patients is the heterogene-
a history of repeated exacerbations [27, 28]. ity that already exists within both asthma and
PDE-4 inhibitors such as roflumilast are believed COPD. For instance, not all asthmatics fit the
to reduce airway inflammation through a variety typical eosinophilic, steroid-responsive profile
of mechanisms [29–32]. While initial clinical tri- [51]. Smoking asthmatics may fit a profile more
als showed inconsistent effects of PDE-4 inhibi- consistent with traditional COPD including mini-
tors on clinically relevant outcomes such as mal lung function improvement with inhaled cor-
frequency of acute exacerbation in all-comers ticosteroids (ICS) [57–59]. However, in general,
with COPD, follow-up studies have demonstrated ACOS represents a group of COPD patients who
improved lung function and reduction in fre- are more likely to receive greater benefit from
quency of exacerbations in patients specifically ICS, regardless of FEV1 severity and exacerba-
with chronic bronchitis symptoms and severe air- tion history, due to it ascribed features (eosinoph-
flow obstruction [27]. Use of roflumilast in this ila, bronchodilator responsiveness) [51].
patient population results in reduced frequency
of exacerbation, modest improvements in FEV1,
and improved dyspnea scores [27]. Gender

Gender has been linked to disease susceptibil-

 sthma COPD Overlap Syndrome
A ity, symptoms, exacerbations, rate of progres-
(ACOS) sion, extent and distribution of airway
abnormalities, and mortality in COPD. Several
Asthma COPD overlap syndrome (ACOS) is studies have demonstrated that women suffer
becoming increasingly recognized as a unique clin- more severe airflow limitation than men for a
ical subgroup but definitive criteria for identifying given tobacco exposure and are disproportionately
152 J. Sheth and M. Han

represented among COPD patients without a his- COPD population and conditions such as obstruc-
tory of significant smoking [60, 61]. The physi- tive sleep apnea (OSA) that may modify the dis-
ologic changes of COPD may also affect ease course.
women and men differently in terms of symp-
toms and quality of life [11]. For a similar Cardiovascular Disease
degree of physiologic impairment, several stud- Ischemic cardiovascular disease continues to be a
ies suggest women experience more severe dys- leading cause of death in COPD [69]. While
pnea and worse health-­related quality of life tobacco use is a shared risk factor, epidemiologic
than men [62–64]. Additionally, the experience evidence suggests that impaired lung function
of symptoms may be further modified by gen- itself is an independent risk factor for cardiovas-
der-associated COPD comorbidities, with cular mortality, even when adjusted for smoking
women for instance e­ xperiencing higher rates status [70]. Data from the National Health and
of anxiety and depression [11]. Female gender Nutrition Examination Survey demonstrated that
and the presence of anxiety and depression patients in the lowest FEV1 quintile had the high-
have both been independently linked to hospital est risk of cardiovascular mortality (RR 3.36),
readmissions for COPD exacerbations [11, 65]. even after adjustment for smoking status, blood
Several studies have also demonstrated an pressure, BMI, and presence of diabetes [70]. In
increased frequency of exacerbation in women; patients with COPD, FEV1 predicts the presence
however, it remains unclear if this is related to of atherosclerosis [71] as well as cardiovascular
disease biology or difference in reporting pat- mortality [72, 73]. COPD is also a risk factor for
terns [66, 67]. From a physiologic perspective, hospitalization due to cardiovascular events [74].
data from the National Emphysema Trial sug- The characterization of atherosclerosis as a dis-
gest that men and women may respond differ- ease of systemic inflammation helps may explain
ently in the type and location of lung damage the connection to COPD [75]. Elevated C-reactive
due to tobacco exposure. In a population of protein (CRP) levels correlate not only with the
patients with severe emphysema, women dem- presence of COPD but also with the presence of
onstrated less severe overall emphysema though exacerbations, severity of lung function, and risk
on histologic examination had significantly for hospitalization and death [76]. Cardioselective
thicker bronchiole airway walls. Work needs to beta-blockers, frequently used in patients with
be done to understand the therapeutic implica- cardiovascular disease, have traditionally been
tions of these data. We do know that the effects used with caution in patients with COPD due to
of smoking cessation, the most impactful inter- the theoretical risk of worsening bronchospasm.
vention for patients with COPD, vary by on However, more recent data has suggested that
gender. Smoking cessation benefits women beta-blockers may actually reduce all-cause mor-
more in regard to lung function (improvement tality in COPD [77–79]. The use of other cardio-
in FEV1) though men have greater improve- vascular disease medications including statins
ment in symptoms. Unfortunately, multiple alone or in combination with angiotensin-­
studies suggest women have greater difficulty converting inhibitor or angiotensin receptor
in sustaining long-term abstinence from tobacco blockers has also been shown to improve overall
use but it may be more important [68]. mortality and reduce hospitalizations in several
COPD cohort studies [80, 81].

Comorbidities Musculoskeletal Disease

Cachexia defined as a loss of fat and muscle tis-
Accumulating data suggest that comorbidities sues can be seen in severe COPD, with a reported
must be included in the assessment of COPD prevalence of 5–15% [82]. Weight loss and mus-
with conditions such as cardiovascular disease cular wasting play a role in low exercise capacity
and osteoporosis that are more prevalent in the seen in COPD [83]. Additionally, low BMI is an
10  Phenotypes of COPD 153

important independent predictor of increased inactivity [101]. However, low bone mineral den-
mortality in patients with COPD [84]. The prog- sity has also been shown in patients with COPD
nostic effect of BMI is further supported by its even in the absence of systemic steroids [101].
inclusion in the BMI, airflow obstruction, dys- Pulmonary rehabilitation improves the functional
pnea, and exercise capacity (BODE) index, which status of patients with COPD and may diminish
is one of the best mortality prediction indices in fracture risk by decreasing the risk of falls, but
COPD [85]. Several studies have also correlated has not been shown to increase bone mineral den-
low BMI with greater extent of emphysema on sity directly [102].
HRCT, confirming its association with the classic
pink puffer phenotype [86]. Conversely higher Diabetes
BMI has been associated with the presence of Diabetes is another comorbidity with increased
chronic bronchitis and HRCT indicators of prevalence in COPD. Lung function impairment
­airway disease [47]. Body composition is impor- has been associated with the coexistence of meta-
tant in the assessment of COPD patients as the bolic syndrome, insulin resistance, and the devel-
negative effect of low body weight on survival opment of diabetes [103–105]. This association
can be reversed by appropriate therapy in some remains even after adjustment for BMI and
COPD patients [87]. A relationship between smoking. The exact cause of this association is
increased mortality and low fat-free mass index not known, but data implicates inhaled cortico-
(FFMI) has also been demonstrated, even in sub- steroids (ICS). While some studies have sug-
jects with a normal BMI [88]. In fact, patients gested a dose-dependent association between
with low FFMI are often more severely disabled inhaled corticosteroid use and diabetes control
than COPD patients with low BMI and normal and new onset diabetes, [106] a retrospective
FFMI [89]. FFMI has been shown to be a strong analysis of eight COPD trials and 26 asthma tri-
predictor of peripheral skeletal muscle weakness, als found no association between ICS and hyper-
exercise capacity, and reduced health status [90– glycemia [107]. A higher prevalence of diabetes
94]. In regard to exercise capacity, FFMI has has been seen with airway-disease-predominant
been shown to correlate more specifically with COPD phenotype compared with emphysema-­
6-minute walk distance [95, 96]. Greater FFMI predominant phenotype [108]. Patients with
has also been inversely associated with emphy- airway-­predominant disease also demonstrated a
sema extent on HRCT [86]. Assessment of FFM higher BMI and less frequent osteoporosis.
is important in patients referred to pulmonary Certain inflammatory mediators such as interleu-
rehabilitation, as exercise training results in a sig- kin 6 (IL-6) and tumor necrosis factor-alpha
nificant gain in FFM in normal-weight COPD (TNF-α) have been implicated in both bronchial
patients [97]. inflammation and insulin resistance [2]. While
As with low BMI, osteoporosis and low bone data supports a relationship between certain met-
mineral density seems to be more strongly related abolic pathways and COPD phenotypes, the
to the presence and severity of emphysema, but exact pathways and therapeutic implications are
not related to the severity of chronic bronchitis or currently unknown.
airway wall thickness as assessed by HRCT [98].
A clear association has been shown between  astroesophageal Reflux Disease
osteoporosis and COPD with studies suggesting The association between gastroesophageal reflux
a two- to five-fold increase in prevalence of disease (GERD) and COPD is also recognized.
osteoporosis in patients with COPD compared The presence of heartburn, regurgitation, and
with age-matched controls [99, 100]. There are dysphagia has been found with higher frequency
multiple shared risk factors between COPD and in patients with COPD compared with controls
osteoporosis that likely influence this association [109, 110]. Studies including esophageal pH
including oral and inhaled steroid use, smok- monitoring have demonstrated that the actual
ing, low body mass index (BMI), and physical proportion of GERD in the COPD population is
154 J. Sheth and M. Han

higher than estimates based on symptoms alone. depressive symptoms are associated with
Both cross-section and longitudinal studies have increased risk of death [127, 128]. While specific
reported associations between the presence of therapies for anxiety and depression have not
reflux and poor quality of life in COPD [111]. been shown to improve COPD outcomes, pulmo-
GERD has also been identified as a risk factor for nary rehab has been shown to improve anxiety
COPD exacerbations [66, 112] and is specifically and depression in addition to overall quality of
associated with the chronic bronchitic phenotype life and functional capacity [129].
discussed previously. The presence of GERD is Two general patterns of clinical features and
associated with increased symptoms, poorer comorbidities have been proposed which may
quality of life, and increased frequency of exacer- represent unique phenotypes: [1] emphysema,
bations; associations which are maintained after low BMI and osteoporosis and [2] chronic bron-
controlling for PPI use [113]. Unfortunately, only chitis, airway disease, high BMI, OSA, and dia-
limited data suggest that treatment of GERD may betes [2]. However, it is still unclear if these
reduce the risk of exacerbations [114]. associations are related to specific mechanistic
pathways that could lead to development of tar-
 bstructive Sleep Apnea (OSA)
O geted therapies.
In patients with COPD, the prevalence of obstruc-
tive sleep apnea has been estimated to be 16%
[115], compared to an estimated 9% in women Physiologically Defined Phenotypes
and 24% in men in the general population [116].
However, poor sleep quality, decreased sleep effi- Spirometric indices, including FEV1, FVC, and
ciency, and difficulties in initiating and maintain- their ratio, are used to define the presence and
ing sleep have been reported in more than 40% of severity of disease. While pulmonary function
patients with COPD [117]. In general, COPD explains less than 10–25% of disease impact on
patients with OSA are more severely hypercap- symptoms, quality of life, and exercise perfor-
nic, demonstrate more profound and frequent mance [130–132], FEV1 remains an important
nocturnal oxygen desaturation, and have a higher outcome measure in COPD. Moreover, rapid
risk of pulmonary hypertension [118]. Untreated physiologic progression as indicated by a change
OSA is also a risk factor for poor quality of live, in FEV1 may indicate a distinct phenotype. A
acute exacerbations, and increased all-cause more rapid decline in FEV1 has been associated
mortality [119, 120]. Diagnosis of OSA in COPD with morbidity, mortality, and hospitalization
is important as initiation of continuous positive rates in COPD [133] and also has been linked to
airway pressure therapy for patients with overlap distinct plasma biomarker signatures [134]. The
has been associated with both decreased frisk of rate of FEV1 decline has been associated with air-
death and decreased incidence of severe exacer- way reactivity, exacerbations, and possibly a
bations [119]. chronic inflammatory state [135–139]. There has
been conflicting data regarding the role of inhaled
 epression and Anxiety
D corticosteroids in reducing decline in FEV1 [140,
Coexistent depression and anxiety are seen in at 141]. Some genetic associations have also been
least 10% of the general COPD population [121, seen, particularly with accelerated decline in
122]; however, with significantly higher esti- FEV1 in smokers with alpha-1 antitrypsin defi-
mated in patients with severe COPD [123]. Risk ciency [142]. Additionally, a single nuclear poly-
factors for depression in COPD include disease morphism has been identified in the A disintegrin
severity, limited mobility, low BMI, comorbid and metalloprotease 33 (ADAM33) gene, a sus-
conditions, the need for supplemental oxygen, ceptibility gene for asthma, that is associated
and female gender [124, 125]. COPD patients with decline in FEV1 in COPD [143]. While
with comorbid anxiety are at risk for COPD smoking cessation remains the best treatment for
exacerbations and higher morality [126], while preventing FEV1 loss [133], it is unclear whether
10  Phenotypes of COPD 155

other currently available medications are effec- or equal to 12% absolute volume [157]. Alternate
tive in slowing lung function decline although physiological criteria include a greater increase
several studies suggest bronchodilators and in FEV1 or FVC (greater than or equal to 400 mL)
inhaled bronchodilator/corticosteroid combina- and a 12% increase in either variable [147].
tions may be effective, particularly in moderate While bronchodilator reversibility defined as
disease [144, 145]. change in FEV1 may be less common with an
GOLD defines an exacerbation as an event emphysema-dominant phenotype [147, 158],
characterized by a change in the patient’s base- there is a small subgroup of patients with severe
line dyspnea, cough, and/or sputum production emphysema that meets ATS/ERS criteria for
that is beyond normal day-to-day variation, is bronchoreversibility. However, the quantity of
acute in onset, and may warrant a change in regu- emphysema determined by HRCT is a negative
lar medication in a patient with underlying COPD predictor of meeting volume-guided bronchorev-
[146]. In literature, exacerbations are often ersibility criteria [147]. In the severe emphysema
defined functionally (and not biologically) as population, bronchoreversibility as defined by a
“health care utilization” events requiring treat- change in FEV1 is infrequent, varies over time,
ment with antibiotics, systemic corticosteroids, and is more common in males and those with less
or both [147]. The natural history of COPD is severe emphysema. The proportion of patients
punctuated by exacerbations that appear to be meeting bronchoreversibility criteria varies by
associated with accelerated decline in lung func- the physiological criteria used to define a positive
tion [136, 137], reduced physical activity, poorer response, with a propensity for an increase in
quality of life, increased health care utilization, FVC in patients with emphysema [147].
and increased risk of death [148–151]. The The true impact of bronchoreversibility on
Evaluation of COPD Longitudinally to Identify clinically meaningful outcomes such as lung
Predictive Surrogate Endpoints (ECLIPSE) function decline and exacerbation frequency is
cohort study demonstrated that while increased controversial. The ISOLDE (Inhaled Steroids in
exacerbations occur as lung function declines, Obstructive Lung Disease in Europe) and Lung
certain individuals demonstrate a relatively stable Health Study data suggest that bronchoreversibil-
susceptibility phenotype irrespective of lung ity is not associated with accelerated decline in
function [66]. Other factors identified as being pulmonary function [159] or the number of exac-
associated with increased exacerbation frequency erbations [154]. However, in patients with alpha
included a history of GERD, poorer quality of 1 antitrypsin deficiency, bronchoreversibility has
life, and elevated white blood cell count [66]. been associated with a greater rate of decline in
Controlled trials have shown that pharmacother- FEV1 [160]. Studies in relatives of COPD patients
apy can reduce exacerbations [144, 152], which have shown familial clustering of spirometric and
should be applied to all patients with the frequent bronchodilators responsiveness phenotypes
exacerbation phenotype across all disease [161–166] and a genome-wide linkage analysis
severities. has suggested that bronchoreversibility can be
While COPD has been characterized as a dis- linked to a specific COPD genotype [167], which
order with airflow obstruction that is not fully increases its utility as meaningful phenotype.
reversible [146, 153], the presence of bronchodi- In addition to the above, several other physio-
lator reversibility has been confirmed in patients logic parameters have been identified that predict
clinically diagnosed with non-asthmatic COPD clinically meaningful outcomes. Airway hyper-
and may be clinically relevant [154–156]. There responsiveness has been associated with more
are various spirometric criteria for bronchorev- rapid longitudinal decline in lung function [147,
ersibility. The American Thoracic Society (ATS)/ 158], while hyperinflation has been related to
European Respiratory Society (ERS) defines mortality and functional impairments [168, 169].
bronchoreversibility as an increase in FEV1 Diffusing capacity impairment is an independent
greater than or equal to 200 mL and greater than predictor of the extent of radiologic emphysema
156 J. Sheth and M. Han

[170], the presence of resting hypoxemia, development of techniques aimed at decreasing

exercise-­induced arterial oxygen desaturation the markedly increased lung volumes seen in
[171], and functional impairment [172]. Whether emphysema and ultimately lung volume reduc-
these parameters are truly reflective of a unique tion surgery (LVRS) [176–181]. The National
biologic processes is unclear and represent pos- Emphysema Treatment Trial (NETT) was a mul-
sibilities for unique therapeutic interventions is ticenter prospective randomized controlled trial
unknown [173, 174]. that compared optimal medical treatment, includ-
ing pulmonary rehabilitation, with optimal medi-
cal treatment plus LVRS in patients with severe
Radiographically Defined emphysema [158, 182]. NETT demonstrated
Phenotypes how physiological and radiographic phenotypes
can predict patient survival as well as response to
Chest radiography and computed tomography treatment. In COPD patients with upper lobe pre-
(CT) are the two imaging modalities most com- dominant emphysema and a low exercise capac-
monly used in COPD. Chest radiography is nei- ity post-rehabilitation, LVRS improved survival
ther sensitive nor specific for the diagnosis of and quality of life, as compared to those with
COPD, but can reveal some characteristic fea- upper lobe disease and high exercise capacity or
tures including radiolucency, diaphragmatic flat- COPD patients with non-upper lobe predominant
tening, and hyperinflation. Chest CT allows for disease regardless of exercise capacity [183]. CT
better detection and quantification of emphysema imaging is required to assess emphysema extent
and can reveal thickened airways indicative of and distribution for the purposes of lung volume
bronchial thickening. CT has also been used as a reduction surgery [158].
noninvasive tool to investigate the heterogeneous Several studies have demonstrated a strong rela-
manifestations of COPD as they correlate to the tionship between emphysema and both lung func-
clinical phenotypes discussed above. tion decline [184, 185] and mortality [186, 187]
Lung hyperinflation in COPD impairs chest (Table 10.2). Bronchial thickening seen on CT also
wall and respiratory muscle mechanics, increases has a strong correlation with symptoms [188].
breathlessness, impairs weaning from mechani- Visually identified bronchiectasis may be associ-
cal ventilation, decreases exercise performance, ated with an increased risk of death compared to
and increases mortality [175]. This led to the COPD patients without bronchiectasis [189].

Table 10.2  Summary of studies examining the relationship between emphysema and mortality
Population Risk factor Hazard ratio Outcome
Zulueta et al., [186] Smokers Presence/absence visually 9.3a Death from COPD
n = 9046 scored emphysema 1.7a Death from lung cancer
Haruna et al., [187] COPD n = 251 Emphysema ≥32% 1.52b (p < 0.001) All-cause mortality
Upper lobe emphysema 1.55 (p < 0.001)
FEV1% 10–35% 1.25 (p = 0.08)
Johannessen et al., Ever-smokers Emphysema 3–10% 56 months less All-cause mortalityc,
[190] n = 974 (49% survival (p < 0.05) respiratoryc,
with COPD) Emphysema >10% 67 months less cardiovascularc, cancer
survival (p < 0.05) and lung cancer
Table from “Clinical correlations of CT Imaging in COPD” [191]
Adjusted for age and smoking history
In multivariate models including age, emphysema, BMI, FEV1, RV/TLC, and DLCO/VA, only age and %LAA (low
areas of attenuation) were significant predictors of mortality from respiratory disease. In models examining all-cause
mortality, BMI, age, LAA% were all predictive
Significant also in models adjusted for sex, interaction between sex and %LAA, COPD status, post-bronchodilator
FEV! %predicted, age, smoking status, and inflation level for emphysema <10% group only
10  Phenotypes of COPD 157

Different radiographic features have been uti- Biologically Defined Phenotypes

lized to define different physiologic parameters,
which are associated with a variety of clinical While the term phenotype encompasses the clini-
outcomes. It is the airways less than 2 mm in cally relevant properties of the disease, it does
diameter, that are thought to be the site of airflow not necessarily defined a distinct to disease etiol-
obstruction in COPD [192], which unfortunately ogy or pathophysiology. Alternatively, endotypes
is below the limit of resolution for CT imaging. describe subtypes of a disease defined by an
COPD’s effects on small airways can be quanti- intrinsically distinct pathogenetic mechanism
fied indirectly through the measurement of air [200]. Endotypes can then be utilized to identify
trapping using both inspiratory and expiratory biomarkers and potential novel therapeutic
images on chest CT [193–195]. One of these approaches [200]. Linking of endotypes to
methods, the ratio of expiratory to inspiratory ­clinical phenotypes and to endotype-specific bio-
mean lung density has been found to correlate markers will be crucial because phenotypes and
with spirometric measure of air trapping (ratio of biomarkers are more accessible to clinicians than
residual volume to total lung capacity) [196] and endotypes [201] (Fig. 10.2).
has been shown to correlate with clinical out- Endotype identification in COPD is still in its
comes including six-minute walk distance, dys- relative infancy, but several possible endotypes
pnea, and BODE score [197]. have been identified. As previously discussed, the
Pulmonary vascular disease has classically asthma-COPD overlap phenotype is a clinically
been described as a phenomenon of severe apparent entity. A specific airway epithelial gene
COPD. The total cross-sectional area of small expression pattern that has previously been noted
pulmonary vessels assessed with CT has signifi- in Th2 inflammation in asthma has also been
cant correlation with pulmonary artery pressures noted in COPD to correlate with decreased lung
assess via right heart catheterization [198]. function, increased airway eosinophil counts,
Pulmonary vascular abnormalities may also greater blood eosinophil percentage, bronchodi-
occur earlier in COPD and have clinical signifi- lator reversibility, and improvement in hyperin-
cance. In fact, CT detected pulmonary artery flation with corticosteroid treatment [203].
enlargement, defined as ratio of the diameter of Interestingly, the presence of Th2-associated sig-
the pulmonary artery to the diameter of the aorta, nature is not predicted by a clinical history of
has been shown to identify patients at risk of asthma.
COPD exacerbation [199].

Clinical phenotype Symptom

driven Tx

Endotype Biomarker-
directed Tx

Risk factors
Exposome Genome

Fig. 10.2 Interrelations between exposome, genome, networks that enable and restrict reactions), and the clini-
endotype, and clinical phenotype of COPD—Diagram of cal phenotype (final clinical expression of the disease,
the interrelations (thin black arrows) between the expo- e.g., symptoms, exacerbations, response to treatment, rate
some (the totality of human environmental exposures, of disease progression, or death). Large arrows represent
from conception onwards [202]), genome (the genetic different treatment strategies. COPD chronic obstructive
background of the individual), the endotype (biological pulmonary disease, Tx treatment [200]
158 J. Sheth and M. Han

Related to this concept, previous studies in in the airways [217]. Further studies are needed
patients with COPD have also shown that eosino- to validate and evaluate potential therapeutic
philic airway inflammation is associated with implications.
acute exacerbations of COPD [204, 205] and can The six inflammatory biomarkers most stud-
also predict a beneficial response to treatment ied in COPD include white blood cell (WBC)
with corticosteroids [206–209]. Eosinophilic air- count, C-reactive protein (CRP), interleukins 6
way inflammation can be seen in 10–20% of (IL-6) and 8 (IL-8), fibrinogen, and tumor necro-
patients with COPD during both stable periods sis factor-alpha (TNF-α) [218]. A subgroup of
and acute exacerbations [204, 205, 207, 210, COPD patients has been identified with persis-
211]. The use of corticosteroid therapy to reduce tently elevated inflammatory biomarker levels
concentrations of airway eosinophils decreases that, despite relatively similar lung function
frequency of severe acute exacerbation [205]. impairment, had significantly increased all-cause
Benralizumab, an anti-interleukin-5 receptor mortality and exacerbation frequency [219].
(alpha) monoclonal antibody, depletes blood and While the severity of airflow limitation has been
sputum eosinophils [212, 213] and has been largely used as the most important criteria to
investigated as a therapeutic to decrease the fre- guide therapy in COPD [146], identification of
quency of acute exacerbation in COPD. Similar this phenotype demonstrates that patients with
to roflumilast, a recent Phase II study of benrali- similar levels of airflow limitation may have dif-
zumab compared to placebo did not reduce the ferent outcomes depending on the presence or
rate of acute exacerbations; however, subgroup absence of persistent systemic inflammation. The
analysis support further investigation in patients treatment implications of these findings will
with COPD and eosinophilia, which may repre- require further investigation.
sent a future target for directed therapy [214].
These data suggest several biomarkers related to Conclusions
inflammatory patterns seen in asthma have the To date the majority of large trials investigat-
potential to identify individuals with COPD who ing COPD therapies have not targeted-specific
may be candidates for directed therapies. phenotypes. Nor have they been designed to
Alpha-1 antitrypsin deficiency, as described clearly identify either predictors of response
above also meets criteria for an endotype of or harm, but development of a “personalized
COPD. It has known genetic foundation, distinct medicine” approach is a critical goal for the
clinical characteristics, characteristic histopa- future of COPD treatment approaches. Future
thology, distinct epidemiology, and a mechanism-­ studies in COPD will be increasingly pheno-
directed treatment approach that is guided by type or ideally endotype driven. In order to
biomarkers, serum a1AT concentration, a1AT achieve that goal, identification and validation
protein phenotyping, and a1AT genotyping [200]. of phenotypes in COPD will require analysis
Other possible endotypes in COPD are also of longitudinal data in carefully characterized
under current investigation. COPD has been patient populations [1]. Multiple large obser-
associated with signs of local inflammation in the vational and cohort studies, some of which
airways involving neutrophils, macrophages, and have been mentioned above, have already
T-cells [215]. Additionally, signs of systemic begun this process by systematically gather-
inflammation involving neutrophils have been ing clinical, physiologic, radiologic, biologic,
linked to the clinical course of smoking-related and genetic data on a wide spectrum of COPD
COPD [215, 216]. Local cytokine signaling by T patients. Representative studies include
helper (Th)17 cells via Interleukin (IL)-17 is ECLIPSE, COPDGene, Subpopulations and
important for antibacterial host defense in the air- Intermediate Outcome Measures in COPD
ways. In smokers with COPD including chronic (SPIROMICS), Canadian cohort obstructive
bronchitis, systemic cytokine signaling via IL-17 lung disease (CanCOLD), and the Korean
appears to be impaired, and this alteration may be Obstructive Lung Disease (KOLD) cohort.
linked to colonization by opportunistic pathogens Advanced statistical techniques incorporating
10  Phenotypes of COPD 159

systems biology and network medicine may BJ, Crapo JD, Bowler RP, Regan EA, Hokanson
also provide additional frameworks for under- JE, Kinney GL, Han MK, Soler X, Ramsdell JW,
Barr RG, Foreman M, van Beek E, Casaburi R,
standing this disease’s complexities [220– Criner GJ, Lutz SM, Rennard SI, Santorico S,
222] and aid in the identification of clinically Sciurba FC, DeMeo DL, Hersh CP, Silverman EK,
meaningful groups [223–226]. Cho MH. Cluster analysis in the COPDGene study
identifies subtypes of smokers with distinct pat-
terns of airway disease and emphysema. Thorax.
10. Eisner MD, Anthonisen N, Coultas D, Kuenzli N,
References Perez-Padilla R, Postma D, Romieu I, Silverman
EK, Balmes JR, Committee on Nonsmoking Copd
1. Han MK, Agusti A, Calverley PM, Celli BR, Criner E, Occupational Health A. An official American
G, Curtis JL, Fabbri LM, Goldin JG, Jones PW, Thoracic Society public policy statement: novel risk
Macnee W, Make BJ, Rabe KF, Rennard SI, Sciurba factors and the global burden of chronic obstructive
FC, Silverman EK, Vestbo J, Washko GR, Wouters pulmonary disease. Am J Respir Crit Care Med.
EF, Martinez FJ. Chronic obstructive pulmonary dis- 2010;182:693–718.
ease phenotypes: the future of COPD. Am J Respir 11. Han MK, Postma D, Mannino DM, Giardino ND,
Crit Care Med. 2010;182:598–604. Buist S, Curtis JL, Martinez FJ. Gender and chronic
2. Martinez CH, Han MK. Contribution of the environ- obstructive pulmonary disease: why it matters. Am
ment and comorbidities to chronic obstructive pul- J Respir Crit Care Med. 2007;176:1179–84.
monary disease phenotypes. Med Clin North Am. 12. Camp PG, Ramirez-Venegas A, Sansores RH, Alva
2012;96:713–27. LF, McDougall JE, Sin DD, Pare PD, Muller NL,
3. Vestbo J, Hurd SS, Agusti AG, Jones PW, Silva CI, Rojas CE, Coxson HO. COPD phenotypes
Vogelmeier C, Anzueto A, Barnes PJ, Fabbri LM, in biomass smoke- versus tobacco smoke-exposed
Martinez FJ, Nishimura M, Stockley RA, Sin DD, Mexican women. Eur Respir J. 2014;43:725–34.
Rodriguez-­Roisin R. Global strategy for the diagno- 13. Gonzalez-Garcia M, Maldonado Gomez D, Torres-­
sis, management, and prevention of chronic obstruc- Duque CA, Barrero M, Jaramillo Villegas C, Perez
tive pulmonary disease: GOLD executive summary. JM, Varon H. Tomographic and functional findings
Am J Respir Crit Care Med. 2013;187:347–65. in severe COPD: comparison between the wood
4. Lokke A, Lange P, Scharling H, Fabricius P, Vestbo smoke-related and smoking-related disease. J Bras
J. Developing COPD: a 25 year follow up study of Pneumol. 2013;39:147–54.
the general population. Thorax. 2006;61:935–9. 14. Moreira MA, Barbosa MA, Queiroz MC, Teixeira
5. McDonough JE, Yuan R, Suzuki M, Seyednejad N, KI, Torres PP, Santana Junior PJ, Montadon Junior
Elliott WM, Sanchez PG, Wright AC, Gefter WB, ME, Jardim JR. Pulmonary changes on HRCT scans
Litzky L, Coxson HO, Pare PD, Sin DD, Pierce RA, in nonsmoking females with COPD due to wood
Woods JC, McWilliams AM, Mayo JR, Lam SC, smoke exposure. J Bras Pneumol. 2013;39:155–63.
Cooper JD, Hogg JC. Small-airway obstruction and 15. Lieberman J, Winter B, Sastre A. Alpha
emphysema in chronic obstructive pulmonary dis- 1-­antitrypsin Pi-types in 965 COPD patients. Chest.
ease. N Engl J Med. 2011;365:1567–75. 1986;89:370–3.
6. Anthonisen NR, Connett JE, Kiley JP, Altose MD, 16. Tobin MJ, Cook PJ, Hutchison DC. Alpha 1 antitryp-
Bailey WC, Buist AS, Conway WA Jr, Enright PL, sin deficiency: the clinical and physiological features
Kanner RE, O’Hara P, et al. Effects of smoking inter- of pulmonary emphysema in subjects homozygous
vention and the use of an inhaled anticholinergic for Pi type Z. A survey by the British Thoracic
bronchodilator on the rate of decline of FEV1. The Association. Br J Dis Chest. 1983;77:14–27.
Lung Health Study. JAMA. 1994;272:1497–505. 17. Larsson C. Natural history and life expectancy in
7. Shavelle RM, Paculdo DR, Kush SJ, Mannino DM, severe alpha1-antitrypsin deficiency, Pi Z. Acta Med
Strauss DJ. Life expectancy and years of life lost Scand. 1978;204:345–51.
in chronic obstructive pulmonary disease: findings 18. Janus ED, Phillips NT, Carrell RW. Smoking, lung
from the NHANES III follow-up study. Int J Chron function, and alpha 1-antitrypsin deficiency. Lancet.
Obstruct Pulmon Dis. 2009;4:137–48. 1985;1:152–4.
8. Tan WC, Sin DD, Bourbeau J, Hernandez P, 19. Black LF, Kueppers F. Alpha1-antitrypsin deficiency
Chapman KR, Cowie R, FitzGerald JM, Marciniuk in nonsmokers. Am Rev Respir Dis. 1978;117:421–8.
DD, Maltais F, Buist AS, Road J, Hogg JC, Kirby 20. Green CE, Vayalapra S, Hampson JA, Mukherjee
M, Coxson H, Hague C, Leipsic J, O’Donnell DE, D, Stockley RA, Turner AM. PiSZ alpha-1 anti-
Aaron SD, Can CCRG. Characteristics of COPD in trypsin deficiency (AATD): pulmonary phenotype
never-smokers and ever-smokers in the general pop- and prognosis relative to PiZZ AATD and PiMM
ulation: results from the CanCOLD study. Thorax. COPD. Thorax. 2015;70(10):939–45.
2015;70(9):822–9. 21. Stockley RA, Turner AM. Alpha-1-antitrypsin defi-
9. Castaldi PJ, Dy J, Ross J, Chang Y, Washko GR, ciency: clinical variability, assessment, and treat-
Curran-Everett D, Williams A, Lynch DA, Make ment. Trends Mol Med. 2014;20:105–15.
160 J. Sheth and M. Han

22. Ferris BG. Epidemiology standardization project 32. Hatzelmann A, Schudt C. Anti-inflammatory and
(American Thoracic Society). Am Rev Respir Dis. immunomodulatory potential of the novel PDE4
1978;118:1–120. inhibitor roflumilast in vitro. J Pharmacol Exp Ther.
23. Forey BA, Thornton AJ, Lee PN. Systematic review 2001;297:267–79.
with meta-analysis of the epidemiological evidence 33. Miravitlles M. Cough and sputum production as
relating smoking to COPD, chronic bronchitis and risk factors for poor outcomes in patients with
emphysema. BMC Pulm Med. 2011;11:36. COPD. Respir Med. 2011;105:1118–28.
24. Dijkstra AE, Smolonska J, van den Berge M, 34. Lange P, Groth S, Nyboe J, Appleyard M, Mortensen
Wijmenga C, Zanen P, Luinge MA, Platteel M, J, Jensen G, Schnohr P. Chronic obstructive lung dis-
Lammers JW, Dahlback M, Tosh K, Hiemstra PS, ease in Copenhagen: cross-sectional epidemiologi-
Sterk PJ, Spira A, Vestbo J, Nordestgaard BG, cal aspects. J Intern Med. 1989;226:25–32.
Benn M, Nielsen SF, Dahl M, Verschuren WM, 35. Sobradillo V, Miravitlles M, Jimenez CA, Gabriel
Picavet HS, Smit HA, Owsijewitsch M, Kauczor R, Viejo JL, Masa JF, Fernandez-Fau L, Villasante
HU, de Koning HJ, Nizankowska-Mogilnicka E, C. Epidemiological study of chronic obstructive pul-
Mejza F, Nastalek P, van Diemen CC, Cho MH, monary disease in Spain (IBERPOC): prevalence of
Silverman EK, Crapo JD, Beaty TH, Lomas DA, chronic respiratory symptoms and airflow limitation.
Bakke P, Gulsvik A, Bosse Y, Obeidat MA, Loth Arch Bronconeumol. 1999;35:159–66.
DW, Lahousse L, Rivadeneira F, Uitterlinden AG, 36. Pallasaho P, Lundback B, Laspa SL, Jonsson E,
Hofman A, Stricker BH, Brusselle GG, van Duijn Kotaniemi J, Sovijarvi AR, Laitinen LA. Increasing
CM, Brouwer U, Koppelman GH, Vonk JM, Nawijn prevalence of asthma but not of chronic bronchitis
MC, Groen HJ, Timens W, Boezen HM, Postma DS, in Finland? Report from the FinEsS-Helsinki Study.
LifeLines Cohort s. Susceptibility to chronic mucus Respir Med. 1999;93:798–809.
hypersecretion, a genome wide association study. 37. von Hertzen L, Reunanen A, Impivaara O, Malkia
PLoS One. 2014;9:e91621. E, Aromaa A. Airway obstruction in relation to
25. Orlandi I, Moroni C, Camiciottoli G, Bartolucci symptoms in chronic respiratory disease—a nation-
M, Pistolesi M, Villari N, Mascalchi M. Chronic ally representative population study. Respir Med.
obstructive pulmonary disease: thin-section CT 2000;94:356–63.
measurement of airway wall thickness and lung 38. Cerveri I, Accordini S, Verlato G, Corsico A, Zoia
attenuation. Radiology. 2005;234:604–10. MC, Casali L, Burney P, de Marco R, European
26. Kim V, Davey A, Comellas AP, Han MK, Washko Community Respiratory Health Survey Study
G, Martinez CH, Lynch D, Lee JH, Silverman G. Variations in the prevalence across countries of
EK, Crapo JD, Make BJ, Criner GJ, Investigators chronic bronchitis and smoking habits in young
CO. Clinical and computed tomographic predic- adults. Eur Respir J. 2001;18:85–92.
tors of chronic bronchitis in COPD: a cross sec- 39. Janson C, Chinn S, Jarvis D, Burney P. Determinants
tional analysis of the COPDGene study. Respir Res. of cough in young adults participating in the
2014;15:52. European Community Respiratory Health Survey.
27. Calverley PM, Rabe KF, Goehring UM, Kristiansen Eur Respir J. 2001;18:647–54.
S, Fabbri LM, Martinez FJ, M, groups Ms. 40. Huchon GJ, Vergnenegre A, Neukirch F, Brami G,
Roflumilast in symptomatic chronic obstructive Roche N, Preux PM. Chronic bronchitis among
pulmonary disease: two randomised clinical trials. French adults: high prevalence and underdiagnosis.
Lancet. 2009;374:685–94. Eur Respir J. 2002;20:806–12.
28. Rennard SI, Calverley PM, Goehring UM, 41. Lundback B, Lindberg A, Lindstrom M, Ronmark
Bredenbroker D, Martinez FJ. Reduction of exac- E, Jonsson AC, Jonsson E, Larsson LG, Andersson
erbations by the PDE4 inhibitor roflumilast—the S, Sandstrom T, Larsson K, Obstructive lung disease
importance of defining different subsets of patients in Northern Sweden S. Not 15 but 50% of smokers
with COPD. Respir Res. 2011;12:18. develop COPD?—report from the obstructive lung
29. Grootendorst DC, Gauw SA, Verhoosel RM, disease in Northern Sweden studies. Respir Med.
Sterk PJ, Hospers JJ, Bredenbroker D, Bethke TD, 2003;97:115–22.
Hiemstra PS, Rabe KF. Reduction in sputum neu- 42. Miravitlles M, de la Roza C, Morera J, Montemayor
trophil and eosinophil numbers by the PDE4 inhibi- T, Gobartt E, Martin A, Alvarez-Sala JL. Chronic
tor roflumilast in patients with COPD. Thorax. respiratory symptoms, spirometry and knowledge
2007;62:1081–7. of COPD among general population. Respir Med.
30. Spina D. PDE4 inhibitors: current status. Br 2006;100:1973–80.
J Pharmacol. 2008;155:308–15. 43. Pelkonen M, Notkola IL, Nissinen A, Tukiainen
31. Bundschuh DS, Eltze M, Barsig J, Wollin L, H, Koskela H. Thirty-year cumulative incidence
Hatzelmann A, Beume R. In vivo efficacy in air- of chronic bronchitis and COPD in relation to
way disease models of roflumilast, a novel orally 30-year pulmonary function and 40-year mortal-
active PDE4 inhibitor. J Pharmacol Exp Ther. ity: a follow-up in middle-aged rural men. Chest.
2001;297:280–90. 2006;130:1129–37.
10  Phenotypes of COPD 161

44. de Marco R, Accordini S, Cerveri I, Corsico A, Anto 55. Menezes AM, Montes de Oca M, Perez-Padilla R,
JM, Kunzli N, Janson C, Sunyer J, Jarvis D, Chinn Nadeau G, Wehrmeister FC, Lopez-Varela MV,
S, Vermeire P, Svanes C, Ackermann-Liebrich Muino A, Jardim JR, Valdivia G, Talamo C, Team
U, Gislason T, Heinrich J, Leynaert B, Neukirch P. Increased risk of exacerbation and hospitaliza-
F, Schouten JP, Wjst M, Burney P. Incidence of tion in subjects with an overlap phenotype: COPD-­
chronic obstructive pulmonary disease in a cohort of asthma. Chest. 2014;145:297–304.
young adults according to the presence of chronic 56. Kitaguchi Y, Fujimoto K, Kubo K, Honda
cough and phlegm. Am J Respir Crit Care Med. T. Characteristics of COPD phenotypes classified
2007;175:32–9. according to the findings of HRCT. Respir Med.
45. Miravitlles M, Soriano JB, Garcia-Rio F, Munoz 2006;100:1742–52.
L, Duran-Tauleria E, Sanchez G, Sobradillo V, 57. Chaudhuri R, Livingston E, McMahon AD,
Ancochea J. Prevalence of COPD in Spain: impact Thomson L, Borland W, Thomson NC. Cigarette
of undiagnosed COPD on quality of life and daily smoking impairs the therapeutic response to oral
life activities. Thorax. 2009;64:863–8. corticosteroids in chronic asthma. Am J Respir Crit
46. Harmsen L, Thomsen SF, Ingebrigtsen T, Steffensen Care Med. 2003;168:1308–11.
IE, Skadhauge LR, Kyvik KO, Backer V. Chronic 58. Chalmers GW, MacLeod KJ, Thomson L, Little SA,
mucus hypersecretion: prevalence and risk fac- McSharry C, Thomson NC. Smoking and airway
tors in younger individuals. Int J Tuberc Lung Dis. inflammation in patients with mild asthma. Chest.
2010;14:1052–8. 2001;120:1917–22.
47. Kim V, Han MK, Vance GB, Make BJ, Newell JD, 59. Boulet LP, Lemiere C, Archambault F, Carrier G,
Hokanson JE, Hersh CP, Stinson D, Silverman EK, Descary MC, Deschesnes F. Smoking and asthma:
Criner GJ, Investigators CO. The chronic bronchitic clinical and radiologic features, lung function, and
phenotype of COPD: an analysis of the COPDGene airway inflammation. Chest. 2006;129:661–8.
study. Chest. 2011;140:626–33. 60. Chapman KR. Chronic obstructive pulmonary dis-
48. Martinez CH, Kim V, Chen Y, Kazerooni EA, ease: are women more susceptible than men? Clin
Murray S, Criner GJ, Curtis JL, Regan EA, Wan E, Chest Med. 2004;25:331–41.
Hersh CP, Silverman EK, Crapo JD, Martinez FJ, 61. Birring SS, Brightling CE, Bradding P, Entwisle JJ,
Han MK, Investigators CO. The clinical impact of Vara DD, Grigg J, Wardlaw AJ, Pavord ID. Clinical,
non-obstructive chronic bronchitis in current and radiologic, and induced sputum features of chronic
former smokers. Respir Med. 2014;108:491–9. obstructive pulmonary disease in nonsmokers:
49. Miravitlles M, Calle M, Soler-Cataluna JJ. Clinical a descriptive study. Am J Respir Crit Care Med.
phenotypes of COPD: identification, definition and 2002;166:1078–83.
implications for guidelines. Arch Bronconeumol. 62. Watson L, Vestbo J, Postma DS, Decramer
2012;48:86–98. M, Rennard S, Kiri VA, Vermeire PA, Soriano
50. Soler-Cataluna JJ, Cosio B, Izquierdo JL, Lopez-­ JB. Gender differences in the management and expe-
Campos JL, Marin JM, Aguero R, Baloira A, rience of chronic obstructive pulmonary disease.
Carrizo S, Esteban C, Galdiz JB, Gonzalez MC, Respir Med. 2004;98:1207–13.
Miravitlles M, Monso E, Montemayor T, Morera J, 63. de Torres JP, Casanova C, Hernandez C, Abreu J,
Ortega F, Peces-Barba G, Puente L, Rodriguez JM, Aguirre-Jaime A, Celli BR. Gender and COPD
Sala E, Sauleda J, Soriano JB, Viejo JL. Consensus in patients attending a pulmonary clinic. Chest.
document on the overlap phenotype COPD-asthma 2005;128:2012–6.
in COPD. Arch Bronconeumol. 2012;48:331–7. 64. Martinez FJ, Curtis JL, Sciurba F, Mumford J,
51. Bujarski S, Parulekar AD, Sharafkhaneh A, Hanania Giardino ND, Weinmann G, Kazerooni E, Murray
NA. The asthma COPD overlap syndrome (ACOS). S, Criner GJ, Sin DD, Hogg J, Ries AL, Han M,
Curr Allergy Asthma Rep. 2015;15:509. Fishman AP, Make B, Hoffman EA, Mohsenifar
52. Hardin M, Silverman EK, Barr RG, Hansel NN, Z, Wise R, National Emphysema Treatment Trial
Schroeder JD, Make BJ, Crapo JD, Hersh CP, Research G. Sex differences in severe pulmo-
Investigators CO. The clinical features of the nary emphysema. Am J Respir Crit Care Med.
overlap between COPD and asthma. Respir Res. 2007;176:243–52.
2011;12:127. 65. Jennings JH, Digiovine B, Obeid D, Frank C. The
53. Zeki AA, Schivo M, Chan A, Albertson TE, Louie association between depressive symptoms and acute
S. The Asthma-COPD overlap syndrome: a com- exacerbations of COPD. Lung. 2009;187:128–35.
mon clinical problem in the elderly. J Allergy. 66. Hurst JR, Vestbo J, Anzueto A, Locantore N,
2011;2011:861926. Mullerova H, Tal-Singer R, Miller B, Lomas
54. Kitaguchi Y, Komatsu Y, Fujimoto K, Hanaoka M, DA, Agusti A, Macnee W, Calverley P, Rennard
Kubo K. Sputum eosinophilia can predict responsive- S, Wouters EF, Wedzicha JA, Evaluation of
ness to inhaled corticosteroid treatment in patients CLtIPSEI. Susceptibility to exacerbation in chronic
with overlap syndrome of COPD and asthma. Int obstructive pulmonary disease. N Engl J Med.
J Chron Obstruct Pulmon Dis. 2012;7:283–9. 2010;363:1128–38.
162 J. Sheth and M. Han

67. Tashkin D, Celli B, Kesten S, Lystig T, Decramer 83. Gologanu D, Ionita D, Gartonea T, Stanescu C,
M. Effect of tiotropium in men and women with Bogdan MA. Body composition in patients with
COPD: results of the 4-year UPLIFT trial. Respir chronic obstructive pulmonary disease. Maedica.
Med. 2010;104:1495–504. 2014;9:25–32.
68. Varkey AB. Chronic obstructive pulmonary disease 84. Landbo C, Prescott E, Lange P, Vestbo J, Almdal
in women: exploring gender differences. Curr Opin TP. Prognostic value of nutritional status in chronic
Pulm Med. 2004;10:98–103. obstructive pulmonary disease. Am J Respir Crit
69. Hansell AL, Walk JA, Soriano JB. What do chronic Care Med. 1999;160:1856–61.
obstructive pulmonary disease patients die from? 85. Nishimura K, Izumi T, Tsukino M, Oga T. Dyspnea
A multiple cause coding analysis. Eur Respir is a better predictor of 5-year survival than air-
J. 2003;22:809–14. way obstruction in patients with COPD. Chest.
70. Sin DD, Wu L, Man SF. The relationship between 2002;121:1434–40.
reduced lung function and cardiovascular mortality: 86. Marquez-Martin E, Ramos PC, Lopez-Campos JL,
a population-based study and a systematic review of Serrano Gotarredona Mdel P, Herrero SN, Aguilar
the literature. Chest. 2005;127:1952–9. RT, Cortes EB, Ruiz FO. Components of physi-
71. Beaty TH, Newill CA, Cohen BH, Tockman MS, cal capacity in patients with chronic obstructive
Bryant SH, Spurgeon HA. Effects of pulmonary func- pulmonary disease: relationship with phenotypic
tion on mortality. J Chronic Dis. 1985;38:703–10. expression. Int J Chron Obstruct Pulmon Dis.
72. Speizer FE, Fay ME, Dockery DW, Ferris BG Jr. 2011;6:105–12.
Chronic obstructive pulmonary disease mortality in 87. Schols AM, Slangen J, Volovics L, Wouters
six U.S. cities. Am Rev Respir Dis. 1989;140:S49–55. EF. Weight loss is a reversible factor in the progno-
73. Hospers JJ, Postma DS, Rijcken B, Weiss ST, sis of chronic obstructive pulmonary disease. Am
Schouten JP. Histamine airway hyper-responsiveness J Respir Crit Care Med. 1998;157:1791–7.
and mortality from chronic obstructive pulmonary 88. Vestbo J, Prescott E, Almdal T, Dahl M, Nordestgaard
disease: a cohort study. Lancet. 2000;356:1313–7. BG, Andersen T, Sorensen TI, Lange P. Body mass, fat-
74. Anthonisen NR, Connett JE, Enright PL, Manfreda free body mass, and prognosis in patients with chronic
J, Lung Health Study Research G. Hospitalizations obstructive pulmonary disease from a random popula-
and mortality in the lung health study. Am J Respir tion sample: findings from the Copenhagen City heart
Crit Care Med. 2002;166:333–9. study. Am J Respir Crit Care Med. 2006;173:79–83.
75. Ross R. Atherosclerosis—an inflammatory disease. 89. Schols AM. Pulmonary cachexia. Int J Cardiol.
N Engl J Med. 1999;340:115–26. 2002;85:101–10.
76. Han MK, McLaughlin VV, Criner GJ, Martinez 90. Engelen MP, Schols AM, Does JD, Wouters
FJ. Pulmonary diseases and the heart. Circulation. EF. Skeletal muscle weakness is associated with
2007;116:2992–3005. wasting of extremity fat-free mass but not with air-
77. Short PM, Lipworth SI, Elder DH, Schembri S, flow obstruction in patients with chronic obstructive
Lipworth BJ. Effect of beta blockers in treatment of pulmonary disease. Am J Clin Nutr. 2000;71:733–8.
chronic obstructive pulmonary disease: a retrospec- 91. Mostert R, Goris A, Weling-Scheepers C, Wouters
tive cohort study. BMJ. 2011;342:d2549. EF, Schols AM. Tissue depletion and health related
78. Dransfield MT, Rowe SM, Johnson JE, Bailey WC, quality of life in patients with chronic obstructive
Gerald LB. Use of beta blockers and the risk of death pulmonary disease. Respir Med. 2000;94:859–67.
in hospitalised patients with acute exacerbations of 92. Baarends EM, Schols AM, Mostert R, Wouters
COPD. Thorax. 2008;63:301–5. EF. Peak exercise response in relation to tissue
79. Rutten FH, Zuithoff NP, Hak E, Grobbee DE, depletion in patients with chronic obstructive pul-
Hoes AW. Beta-blockers may reduce mortality monary disease. Eur Respir J. 1997;10:2807–13.
and risk of exacerbations in patients with chronic 93. Schols AM, Mostert R, Soeters PB, Wouters
obstructive pulmonary disease. Arch Intern Med. EF. Body composition and exercise performance in
2010;170:880–7. patients with chronic obstructive pulmonary disease.
80. Soyseth V, Brekke PH, Smith P, Omland T. Statin use Thorax. 1991;46:695–9.
is associated with reduced mortality in COPD. Eur 94. Sabino PG, Silva BM, Brunetto AF. Nutritional
Respir J. 2007;29:279–83. status is related to fat-free mass, exercise capacity
81. Mancini GB, Etminan M, Zhang B, Levesque LE, and inspiratory strength in severe chronic obstruc-
FitzGerald JM, Brophy JM. Reduction of morbid- tive pulmonary disease patients. Clinics (Sao Paulo).
ity and mortality by statins, angiotensin-converting 2010;65:599–605.
enzyme inhibitors, and angiotensin receptor block- 95. Budweiser S, Meyer K, Jorres RA, Heinemann F,
ers in patients with chronic obstructive pulmonary Wild PJ, Pfeifer M. Nutritional depletion and its
disease. J Am Coll Cardiol. 2006;47:2554–60. relationship to respiratory impairment in patients
82. von Haehling S, Anker SD. Cachexia as a major with chronic respiratory failure due to COPD
underestimated and unmet medical need: facts and or restrictive thoracic diseases. Eur J Clin Nutr.
numbers. J Cachex Sarcopenia Muscle. 2010;1:1–5. 2008;62:436–43.
10  Phenotypes of COPD 163

96. Ischaki E, Papatheodorou G, Gaki E, Papa I, JE, Crapo JD, Silverman EK, Martinez FJ, Washko
Koulouris N, Loukides S. Body mass and fat-free GR, Investigators CO. Chronic obstructive pul-
mass indices in COPD: relation with variables monary disease exacerbations in the COPDGene
expressing disease severity. Chest. 2007;132:164–9. study: associated radiologic phenotypes. Radiology.
97. Franssen FM, Broekhuizen R, Janssen PP, Wouters 2011;261:274–82.
EF, Schols AM. Effects of whole-body exercise 109. Mokhlesi B, Morris AL, Huang CF, Curcio AJ,
training on body composition and functional capac- Barrett TA, Kamp DW. Increased prevalence of
ity in normal-weight patients with COPD. Chest. gastroesophageal reflux symptoms in patients with
2004;125:2021–8. COPD. Chest. 2001;119:1043–8.
98. Bon J, Fuhrman CR, Weissfeld JL, Duncan SR, 110. Terada K, Muro S, Sato S, Ohara T, Haruna A,
Branch RA, Chang CC, Zhang Y, Leader JK, Marumo S, Kinose D, Ogawa E, Hoshino Y, Niimi A,
Gur D, Greenspan SL, Sciurba FC. Radiographic Terada T, Mishima M. Impact of gastro-­oesophageal
emphysema predicts low bone mineral density in reflux disease symptoms on COPD exacerbation.
a tobacco-exposed cohort. Am J Respir Crit Care Thorax. 2008;63:951–5.
Med. 2011;183:885–90. 111. Rascon-Aguilar IE, Pamer M, Wludyka P, Cury J,
99. Bolton CE, Ionescu AA, Shiels KM, Pettit RJ, Vega KJ. Poorly treated or unrecognized GERD
Edwards PH, Stone MD, Nixon LS, Evans WD, reduces quality of life in patients with COPD. Dig
Griffiths TL, Shale DJ. Associated loss of fat-free Dis Sci. 2011;56:1976–80.
mass and bone mineral density in chronic obstruc- 112. Rascon-Aguilar IE, Pamer M, Wludyka P, Cury
tive pulmonary disease. Am J Respir Crit Care Med. J, Coultas D, Lambiase LR, Nahman NS, Vega
2004;170:1286–93. KJ. Role of gastroesophageal reflux symptoms in
100. Nuti R, Siviero P, Maggi S, Guglielmi G, Caffarelli exacerbations of COPD. Chest. 2006;130:1096–101.
C, Crepaldi G, Gonnelli S. Vertebral fractures in 113. Martinez CH, Okajima Y, Murray S, Washko
patients with chronic obstructive pulmonary disease: GR, Martinez FJ, Silverman EK, Lee JH, Regan
the EOLO study. Osteoporos Int. 2009;20:989–98. EA, Crapo JD, Curtis JL, Hatabu H, Han MK,
101. Dam TT, Harrison S, Fink HA, Ramsdell J, Barrett-­ Investigators CO. Impact of self-reported gas-
Connor E. Osteoporotic fractures in men research troesophageal reflux disease in subjects from
G. Bone mineral density and fractures in older COPDGene cohort. Respir Res. 2014;15:62.
men with chronic obstructive pulmonary disease or 114. Sasaki T, Nakayama K, Yasuda H, Yoshida M,
asthma. Osteoporos Int. 2010;21:1341–9. Asamura T, Ohrui T, Arai H, Araya J, Kuwano K,
102. Beauchamp MK, O’Hoski S, Goldstein RS, Brooks Yamaya M. A randomized, single-blind study of
D. Effect of pulmonary rehabilitation on balance in lansoprazole for the prevention of exacerbations
persons with chronic obstructive pulmonary disease. of chronic obstructive pulmonary disease in older
Arch Phys Med Rehabil. 2010;91:1460–5. patients. J Am Geriatr Soc. 2009;57:1453–7.
103. Leone N, Courbon D, Thomas F, Bean K, Jego 115. Simon-Tuval T, Scharf SM, Maimon N, Bernhard-­
B, Leynaert B, Guize L, Zureik M. Lung function Scharf BJ, Reuveni H, Tarasiuk A. Determinants
impairment and metabolic syndrome: the critical of elevated healthcare utilization in patients with
role of abdominal obesity. Am J Respir Crit Care COPD. Respir Res. 2011;12:7.
Med. 2009;179:509–16. 116. Young T, Palta M, Dempsey J, Skatrud J, Weber S,
104. Song Y, Klevak A, Manson JE, Buring JE, Liu Badr S. The occurrence of sleep-disordered breath-
S. Asthma, chronic obstructive pulmonary disease, ing among middle-aged adults. N Engl J Med.
and type 2 diabetes in the Women’s health study. 1993;328:1230–5.
Diabetes Res Clin Pract. 2010;90:365–71. 117. Kinsman RA, Yaroush RA, Fernandez E, Dirks JF,
105. Lazarus R, Sparrow D, Weiss ST. Baseline venti- Schocket M, Fukuhara J. Symptoms and experi-
latory function predicts the development of higher ences in chronic bronchitis and emphysema. Chest.
levels of fasting insulin and fasting insulin resis- 1983;83:755–61.
tance index: the normative aging study. Eur Respir 118. Weitzenblum E, Chaouat A, Kessler R, Canuet
J. 1998;12:641–5. M. Overlap syndrome: obstructive sleep apnea in
106. Suissa S, Kezouh A, Ernst P. Inhaled corticosteroids patients with chronic obstructive pulmonary disease.
and the risks of diabetes onset and progression. Am Proc Am Thorac Soc. 2008;5:237–41.
J Med. 2010;123:1001–6. 119. Marin JM, Soriano JB, Carrizo SJ, Boldova A, Celli
107. O’Byrne PM, Rennard S, Gerstein H, Radner F, BR. Outcomes in patients with chronic obstructive
Peterson S, Lindberg B, Carlsson LG, Sin DD. Risk pulmonary disease and obstructive sleep apnea:
of new onset diabetes mellitus in patients with the overlap syndrome. Am J Respir Crit Care Med.
asthma or COPD taking inhaled corticosteroids. 2010;182:325–31.
Respir Med. 2012;106:1487–93. 120. Mermigkis C, Kopanakis A, Foldvary-Schaefer N,
108. Han MK, Kazerooni EA, Lynch DA, Liu LX, Golish J, Polychronopoulos V, Schiza S, Amfilochiou
Murray S, Curtis JL, Criner GJ, Kim V, Bowler RP, A, Siafakas N, Bouros D. Health-related quality of
Hanania NA, Anzueto AR, Make BJ, Hokanson life in patients with obstructive sleep apnoea and
164 J. Sheth and M. Han

chronic obstructive pulmonary disease (overlap syn- 133. Wise RA. The value of forced expiratory volume in 1
drome). Int J Clin Pract. 2007;61:207–11. second decline in the assessment of chronic obstruc-
121. Kunik ME, Roundy K, Veazey C, Souchek J, tive pulmonary disease progression. Am J Med.
Richardson P, Wray NP, Stanley MA. Surprisingly 2006;119:4–11.
high prevalence of anxiety and depression in chronic 134. Devanarayan V, Scholand MB, Hoidal J, Leppert
breathing disorders. Chest. 2005;127:1205–11. MF, Crackower MA, O’Neill GP, Gervais
122. Yohannes AM, Willgoss TG, Baldwin RC, Connolly FG. Identification of distinct plasma biomarker sig-
MJ. Depression and anxiety in chronic heart fail- natures in patients with rapid and slow declining
ure and chronic obstructive pulmonary disease: forms of COPD. COPD. 2010;7:51–8.
prevalence, relevance, clinical implications and 135. Tashkin DP, Altose MD, Connett JE, Kanner RE,
management principles. Int J Geriatr Psychiatry. Lee WW, Wise RA. Methacholine reactivity predicts
2010;25:1209–21. changes in lung function over time in smokers with
123. Solano JP, Gomes B, Higginson IJ. A comparison early chronic obstructive pulmonary disease. The
of symptom prevalence in far advanced cancer, Lung Health Study Research Group. Am J Respir
AIDS, heart disease, chronic obstructive pulmonary Crit Care Med. 1996;153:1802–11.
disease and renal disease. J Pain Symptom Manag. 136. Donaldson GC, Seemungal TA, Bhowmik A,
2006;31:58–69. Wedzicha JA. Relationship between exacerbation fre-
124. Laurin C, Lavoie KL, Bacon SL, Dupuis G, Lacoste quency and lung function decline in chronic obstruc-
G, Cartier A, Labrecque M. Sex differences in the tive pulmonary disease. Thorax. 2002;57:847–52.
prevalence of psychiatric disorders and psycho- 137. Kanner RE, Anthonisen NR, Connett JE, Lung
logical distress in patients with COPD. Chest. Health Study Research G. Lower respiratory ill-
2007;132:148–55. nesses promote FEV(1) decline in current smokers
125. Hanania NA, Mullerova H, Locantore NW, Vestbo J, but not ex-smokers with mild chronic obstructive
Watkins ML, Wouters EF, Rennard SI, Sharafkhaneh pulmonary disease: results from the lung health
A, Evaluation of CLtIPSEsi. Determinants of study. Am J Respir Crit Care Med. 2001;164:358–64.
depression in the ECLIPSE chronic obstructive pul- 138. Kanazawa H, Hirata K, Yoshikawa J. Accelerated
monary disease cohort. Am J Respir Crit Care Med. decline of lung function in COPD patients with
2011;183:604–11. chronic hepatitis C virus infection: a preliminary
126. Eisner MD, Blanc PD, Yelin EH, Katz PP, Sanchez study based on small numbers of patients. Chest.
G, Iribarren C, Omachi TA. Influence of anxiety on 2003;123:596–9.
health outcomes in COPD. Thorax. 2010;65:229–34. 139. Vergnenegre A, Pugnere N, Antonini MT, Arnaud
127. Fan VS, Ramsey SD, Giardino ND, Make BJ, Emery M, Melloni B, Treves R, Bonnaud F. Airway
CF, Diaz PT, Benditt JO, Mosenifar Z, McKenna R obstruction and rheumatoid arthritis. Eur Respir
Jr, Curtis JL, Fishman AP, Martinez FJ, National J. 1997;10:1072–8.
Emphysema Treatment Trial Research G. Sex, 140. Highland KB, Strange C, Heffner JE. Long-term
depression, and risk of hospitalization and mortal- effects of inhaled corticosteroids on FEV1 in patients
ity in chronic obstructive pulmonary disease. Arch with chronic obstructive pulmonary disease. A meta-­
Intern Med. 2007;167:2345–53. analysis. Ann Intern Med. 2003;138:969–73.
128. de Voogd JN, Wempe JB, Koeter GH, Postema K, 141. Sutherland ER, Allmers H, Ayas NT, Venn AJ,
van Sonderen E, Ranchor AV, Coyne JC, Sanderman Martin RJ. Inhaled corticosteroids reduce the pro-
R. Depressive symptoms as predictors of mortality gression of airflow limitation in chronic obstruc-
in patients with COPD. Chest. 2009;135:619–25. tive pulmonary disease: a meta-analysis. Thorax.
129. von Leupoldt A, Taube K, Lehmann K, Fritzsche A, 2003;58:937–41.
Magnussen H. The impact of anxiety and depression 142. Sandford AJ, Silverman EK. Chronic obstruc-
on outcomes of pulmonary rehabilitation in patients tive pulmonary disease. 1: susceptibility factors
with COPD. Chest. 2011;140:730–6. for COPD the genotype-environment interaction.
130. Mahler DA, Ward J, Fierro-Carrion G, Waterman LA, Thorax. 2002;57:736–41.
Lentine TF, Mejia-Alfaro R, Baird JC. Development 143. van Diemen CC, Postma DS, Vonk JM, Bruinenberg
of self-administered versions of modified baseline M, Schouten JP, Boezen HM. A disintegrin and
and transition dyspnea indexes in COPD. COPD. metalloprotease 33 polymorphisms and lung
2004;1:165–72. function decline in the general population. Am
131. Jones PW. Health status measurement in J Respir Crit Care Med. 2005;172:329–33.
chronic obstructive pulmonary disease. Thorax. 144. Tashkin DP, Celli B, Senn S, Burkhart D, Kesten S,
2001;56:880–7. Menjoge S, Decramer M, Investigators US. A 4-year
132. Brown CD, Benditt JO, Sciurba FC, Lee SM, Criner trial of tiotropium in chronic obstructive pulmonary
GJ, Mosenifar Z, Shade DM, Slivka WA, Wise RA, disease. N Engl J Med. 2008;359:1543–54.
National emphysema Treatment Trial Research 145. Celli BR, Thomas NE, Anderson JA, Ferguson GT,
G. Exercise testing in severe emphysema: associa- Jenkins CR, Jones PW, Vestbo J, Knobil K, Yates JC,
tion with quality of life and lung function. COPD. Calverley PM. Effect of pharmacotherapy on rate of
2008;5:117–24. decline of lung function in chronic obstructive pul-
10  Phenotypes of COPD 165

monary disease: results from the TORCH study. Am Emphysema Treatment Trial Research G. A random-
J Respir Crit Care Med. 2008;178:332–8. ized trial comparing lung-volume-reduction surgery
146. Rabe KF, Hurd S, Anzueto A, Barnes PJ, Buist with medical therapy for severe emphysema. N Engl
SA, Calverley P, Fukuchi Y, Jenkins C, Rodriguez-­ J Med. 2003;348:2059–73.
Roisin R, van Weel C, Zielinski J, Global Initiative 159. Anthonisen NR, Lindgren PG, Tashkin DP,
for Chronic Obstructive Lung D. Global strategy Kanner RE, Scanlon PD, Connett JE, Lung Health
for the diagnosis, management, and prevention Study Research G. Bronchodilator response in
of chronic obstructive pulmonary disease: GOLD the lung health study over 11 yrs. Eur Respir
executive summary. Am J Respir Crit Care Med. J. 2005;26:45–51.
2007;176:532–55. 160. Dowson LJ, Guest PJ, Stockley RA. Longitudinal
147. Han MK, Wise R, Mumford J, Sciurba F, Criner changes in physiological, radiological, and health
GJ, Curtis JL, Murray S, Sternberg A, Weinman G, status measurements in alpha(1)-antitrypsin defi-
Kazerooni E, Fishman AP, Make B, Hoffman EA, ciency and factors associated with decline. Am
Mosenifar Z, Martinez FJ, Group NR. Prevalence J Respir Crit Care Med. 2001;164:1805–9.
and clinical correlates of bronchoreversibility in 161. Kueppers F, Miller RD, Gordon H, Hepper NG,
severe emphysema. Eur Respir J. 2010;35:1048–56. Offord K. Familial prevalence of chronic obstruc-
148. Soler-Cataluna JJ, Martinez-Garcia MA, Roman tive pulmonary disease in a matched pair study. Am
Sanchez P, Salcedo E, Navarro M, Ochando J Med. 1977;63:336–42.
R. Severe acute exacerbations and mortality in 162. Beaty TH, Liang KY, Seerey S, Cohen BH. Robust
patients with chronic obstructive pulmonary disease. inference for variance components models in families
Thorax. 2005;60:925–31. ascertained through probands: II. Analysis of spiro-
149. Donaldson GC, Wilkinson TM, Hurst JR, Perera metric measures. Genet Epidemiol. 1987;4:211–21.
WR, Wedzicha JA. Exacerbations and time spent 163. Cohen BH, Ball WC Jr, Brashears S, Diamond
outdoors in chronic obstructive pulmonary disease. EL, Kreiss P, Levy DA, Menkes HA, Permutt S,
Am J Respir Crit Care Med. 2005;171:446–52. Tockman MS. Risk factors in chronic obstruc-
150. Seemungal TA, Donaldson GC, Paul EA, Bestall tive pulmonary disease (COPD). Am J Epidemiol.
JC, Jeffries DJ, Wedzicha JA. Effect of exacerbation 1977;105:223–32.
on quality of life in patients with chronic obstruc- 164. Cohen BH. Chronic obstructive pulmonary disease: a
tive pulmonary disease. Am J Respir Crit Care Med. challenge in genetic epidemiology. Am J Epidemiol.
1998;157:1418–22. 1980;112:274–88.
151. Sullivan SD, Ramsey SD, Lee TA. The economic 165. Celedon JC, Speizer FE, Drazen JM, Weiss ST,
burden of COPD. Chest. 2000;117:5S–9S. Campbell EJ, Carey VJ, Reilly JJ, Ginns L, Silverman
152. Calverley PM, Anderson JA, Celli B, Ferguson GT, EK. Bronchodilator responsiveness and serum total
Jenkins C, Jones PW, Yates JC, Vestbo J, investiga- IgE levels in families of probands with severe early-­
tors T. Salmeterol and fluticasone propionate and onset COPD. Eur Respir J. 1999;14:1009–14.
survival in chronic obstructive pulmonary disease. N 166. Silverman EK, Chapman HA, Drazen JM, Weiss ST,
Engl J Med. 2007;356:775–89. Rosner B, Campbell EJ, O’Donnell WJ, Reilly JJ,
153. Celli BR, MacNee W, Force AET. Standards for the Ginns L, Mentzer S, Wain J, Speizer FE. Genetic
diagnosis and treatment of patients with COPD: a epidemiology of severe, early-onset chronic obstruc-
summary of the ATS/ERS position paper. Eur Respir tive pulmonary disease. Risk to relatives for airflow
J. 2004;23:932–46. obstruction and chronic bronchitis. Am J Respir Crit
154. Calverley PM, Burge PS, Spencer S, Anderson JA, Care Med. 1998;157:1770–8.
Jones PW. Bronchodilator reversibility testing in 167. Palmer LJ, Celedon JC, Chapman HA, Speizer FE,
chronic obstructive pulmonary disease. Thorax. Weiss ST, Silverman EK. Genome-wide linkage
2003;58:659–64. analysis of bronchodilator responsiveness and post-­
155. Tashkin DP, Celli B, Decramer M, Liu D, Burkhart bronchodilator spirometric phenotypes in chronic
D, Cassino C, Kesten S. Bronchodilator respon- obstructive pulmonary disease. Hum Mol Genet.
siveness in patients with COPD. Eur Respir 2003;12:1199–210.
J. 2008;31:742–50. 168. Casanova C, Cote C, de Torres JP, Aguirre-Jaime A,
156. Anthonisen NR, Wright EC. Bronchodilator Marin JM, Pinto-Plata V, Celli BR. Inspiratory-to-­
response in chronic obstructive pulmonary disease. total lung capacity ratio predicts mortality in patients
Am Rev Respir Dis. 1986;133:814–9. with chronic obstructive pulmonary disease. Am
157. Pellegrino R, Viegi G, Brusasco V, Crapo RO, J Respir Crit Care Med. 2005;171:591–7.
Burgos F, Casaburi R, Coates A, van der Grinten 169. O’Donnell DE, Lam M, Webb KA. Spirometric
CP, Gustafsson P, Hankinson J, Jensen R, Johnson correlates of improvement in exercise performance
DC, MacIntyre N, McKay R, Miller MR, Navajas D, after anticholinergic therapy in chronic obstructive
Pedersen OF, Wanger J. Interpretative strategies for pulmonary disease. Am J Respir Crit Care Med.
lung function tests. Eur Respir J. 2005;26:948–68. 1999;160:542–9.
158. Fishman A, Martinez F, Naunheim K, Piantadosi S, 170. Gelb AF, Hogg JC, Muller NL, Schein MJ, Kuei
Wise R, Ries A, Weinmann G, Wood DE, National J, Tashkin DP, Epstein JD, Kollin J, Green RH,
166 J. Sheth and M. Han

Zamel N, Elliott WM, Hadjiaghai L. Contribution National Emphysema Treatment Trial Research
of emphysema and small airways in COPD. Chest. G. Long-term follow-up of patients receiving lung-­
1996;109:353–9. volume-­ reduction surgery versus medical therapy
171. Owens GR, Rogers RM, Pennock BE, Levin D. The for severe emphysema by the National Emphysema
diffusing capacity as a predictor of arterial oxygen Treatment Trial Research Group. Ann Thorac Surg.
desaturation during exercise in patients with chronic 2006;82:431–43.
obstructive pulmonary disease. N Engl J Med. 184. Vestbo J, Edwards LD, Scanlon PD, Yates JC,
1984;310:1218–21. Agusti A, Bakke P, Calverley PM, Celli B, Coxson
172. Dillard TA, Piantadosi S, Rajagopal KR. Determinants HO, Crim C, Lomas DA, MacNee W, Miller BE,
of maximum exercise capacity in patients with Silverman EK, Tal-Singer R, Wouters E, Rennard SI,
chronic airflow obstruction. Chest. 1989;96:267–71. Investigators E. Changes in forced expiratory vol-
173. Pinto-Plata V, Toso J, Lee K, Park D, Bilello J, Mullerova ume in 1 second over time in COPD. N Engl J Med.
H, De Souza MM, Vessey R, Celli B. Profiling serum 2011;365:1184–92.
biomarkers in patients with COPD: associations with 185. Nishimura M, Makita H, Nagai K, Konno S,
clinical parameters. Thorax. 2007;62:595–601. Nasuhara Y, Hasegawa M, Shimizu K, Betsuyaku
174. Hersh CP, Demeo DL, Lazarus R, Celedon JC, Raby T, Ito YM, Fuke S, Igarashi T, Akiyama Y, Ogura
BA, Benditt JO, Criner G, Make B, Martinez FJ, S, Hokkaido CCSI. Annual change in pulmonary
Scanlon PD, Sciurba FC, Utz JP, Reilly JJ, Silverman function and clinical phenotype in chronic obstruc-
EK. Genetic association analysis of functional tive pulmonary disease. Am J Respir Crit Care Med.
impairment in chronic obstructive pulmonary dis- 2012;185:44–52.
ease. Am J Respir Crit Care Med. 2006;173:977–84. 186. Zulueta JJ, Wisnivesky JP, Henschke CI, Yip
175. Criner GJ, Cordova F, Sternberg AL, Martinez R, Farooqi AO, McCauley DI, Chen M, Libby
FJ. The National Emphysema Treatment Trial DM, Smith JP, Pasmantier MW, Yankelevitz
(NETT): part I: lessons learned about emphysema. DF. Emphysema scores predict death from COPD
Am J Respir Crit Care Med. 2011;184:763–70. and lung cancer. Chest. 2012;141:1216–23.
176. Cooper JD, Trulock EP, Triantafillou AN, Patterson 187. Haruna A, Muro S, Nakano Y, Ohara T, Hoshino Y,
GA, Pohl MS, Deloney PA, Sundaresan RS, Roper Ogawa E, Hirai T, Niimi A, Nishimura K, Chin K,
CL. Bilateral pneumectomy (volume reduction) for Mishima M. CT scan findings of emphysema predict
chronic obstructive pulmonary disease. J Thorac mortality in COPD. Chest. 2010;138:635–40.
Cardiovasc Surg. 1995;109:106–16. discussion 188. Grydeland TB, Dirksen A, Coxson HO, Eagan TM,
116-109 Thorsen E, Pillai SG, Sharma S, Eide GE, Gulsvik
177. Keenan RJ, Landreneau RJ, Sciurba FC, Ferson A, Bakke PS. Quantitative computed tomography
PF, Holbert JM, Brown ML, Fetterman LS, Bowers measures of emphysema and airway wall thickness
CM. Unilateral thoracoscopic surgical approach are related to respiratory symptoms. Am J Respir
for diffuse emphysema. J Thorac Cardiovasc Surg. Crit Care Med. 2010;181:353–9.
1996;111:308–15. discussion 315-306 189. Martinez-Garcia MA, de la Rosa CD, Soler-­
178. McKenna RJ Jr, Brenner M, Gelb AF, Mullin M, Cataluna JJ, Donat-Sanz Y, Serra PC, Lerma MA,
Singh N, Peters H, Panzera J, Calmese J, Schein Ballestin J, Sanchez IV, Selma Ferrer MJ, Dalfo AR,
MJ. A randomized, prospective trial of stapled lung Valdecillos MB. Prognostic value of bronchiectasis
reduction versus laser bullectomy for diffuse emphy- in patients with moderate-to-severe chronic obstruc-
sema. J Thorac Cardiovasc Surg. 1996;111:317–21. tive pulmonary disease. Am J Respir Crit Care Med.
discussion 322 2013;187:823–31.
179. Sciurba FC, Rogers RM, Keenan RJ, Slivka WA, 190. Johannessen A, Skorge TD, Bottai M, Grydeland
Gorcsan J 3rd, Ferson PF, Holbert JM, Brown ML, TB, Nilsen RM, Coxson H, Dirksen A, Omenaas E,
Landreneau RJ. Improvement in pulmonary function Gulsvik A, Bakke P. Mortality by level of emphy-
and elastic recoil after lung-reduction surgery for dif- sema and airway wall thickness. Am J Respir Crit
fuse emphysema. N Engl J Med. 1996;334:1095–9. Care Med. 2013;187:602–8.
180. Gelb AF, Zamel N, McKenna RJ Jr, Brenner 191. Han MK. Clinical correlations of computed tomog-
M. Mechanism of short-term improvement in lung raphy imaging in chronic obstructive pulmonary dis-
function after emphysema resection. Am J Respir ease. Ann Am Thorac Soc. 2013;10(Suppl):S131–7.
Crit Care Med. 1996;154:945–51. 192. Hogg JC. Pathophysiology of airflow limitation
181. Dantzker DR, Scharf SM. Surgery to reduce lung in chronic obstructive pulmonary disease. Lancet.
volume. N Engl J Med. 1996;334:1128–9. 2004;364:709–21.
182. Weinmann GG, Chiang YP, Sheingold S. The 193. Jain N, Covar RA, Gleason MC, Newell JD Jr,
National Emphysema Treatment Trial (NETT): a Gelfand EW, Spahn JD. Quantitative computed
study in agency collaboration. Proc Am Thorac Soc. tomography detects peripheral airway disease in asth-
2008;5:381–4. matic children. Pediatr Pulmonol. 2005;40:211–8.
183. Naunheim KS, Wood DE, Mohsenifar Z, Sternberg 194. Matsuoka S, Kurihara Y, Yagihashi K, Hoshino M,
AL, Criner GJ, DeCamp MM, Deschamps CC, Watanabe N, Nakajima Y. Quantitative assessment
Martinez FJ, Sciurba FC, Tonascia J, Fishman AP, of air trapping in chronic obstructive pulmonary
10  Phenotypes of COPD 167

disease using inspiratory and expiratory volumetric chronic obstructive pulmonary disease: identifica-
MDCT. AJR Am J Roentgenol. 2008;190:762–9. tion of biologic clusters and their biomarkers. Am
195. O’Donnell RA, Peebles C, Ward JA, Daraker A, J Respir Crit Care Med. 2011;184:662–71.
Angco G, Broberg P, Pierrou S, Lund J, Holgate 205. Siva R, Green RH, Brightling CE, Shelley M,
ST, Davies DE, Delany DJ, Wilson SJ, Djukanovic Hargadon B, McKenna S, Monteiro W, Berry M,
R. Relationship between peripheral airway dysfunc- Parker D, Wardlaw AJ, Pavord ID. Eosinophilic
tion, airway obstruction, and neutrophilic inflamma- airway inflammation and exacerbations of
tion in COPD. Thorax. 2004;59:837–42. COPD: a randomised controlled trial. Eur Respir
196. Mets OM, Zanen P, Lammers JW, Isgum I, Gietema J. 2007;29:906–13.
HA, van Ginneken B, Prokop M, de Jong PA. Early 206. Brightling CE, McKenna S, Hargadon B, Birring
identification of small airways disease on lung can- S, Green R, Siva R, Berry M, Parker D, Monteiro
cer screening CT: comparison of current air trapping W, Pavord ID, Bradding P. Sputum eosinophilia and
measures. Lung. 2012;190:629–33. the short term response to inhaled mometasone in
197. Lee YK, Oh YM, Lee JH, Kim EK, Lee JH, Kim N, chronic obstructive pulmonary disease. Thorax.
Seo JB, Lee SD, Group KS. Quantitative assessment 2005;60:193–8.
of emphysema, air trapping, and airway thickening 207. Brightling CE, Monteiro W, Ward R, Parker D,
on computed tomography. Lung. 2008;186:157–65. Morgan MD, Wardlaw AJ, Pavord ID. Sputum
198. Matsuoka S, Washko GR, Yamashiro T, Estepar RS, eosinophilia and short-term response to predniso-
Diaz A, Silverman EK, Hoffman E, Fessler HE, lone in chronic obstructive pulmonary disease: a ran-
Criner GJ, Marchetti N, Scharf SM, Martinez FJ, domised controlled trial. Lancet. 2000;356:1480–5.
Reilly JJ, Hatabu H, National Emphysema Treatment 208. Leigh R, Pizzichini MM, Morris MM, Maltais F,
Trial Research G. Pulmonary hypertension and com- Hargreave FE, Pizzichini E. Stable COPD: predict-
puted tomography measurement of small pulmonary ing benefit from high-dose inhaled corticosteroid
vessels in severe emphysema. Am J Respir Crit Care treatment. Eur Respir J. 2006;27:964–71.
Med. 2010;181:218–25. 209. Pizzichini E, Pizzichini MM, Gibson P,
199. Wells JM, Washko GR, Han MK, Abbas N, Nath Parameswaran K, Gleich GJ, Berman L, Dolovich
H, Mamary AJ, Regan E, Bailey WC, Martinez J, Hargreave FE. Sputum eosinophilia predicts
FJ, Westfall E, Beaty TH, Curran-Everett D, Curtis benefit from prednisone in smokers with chronic
JL, Hokanson JE, Lynch DA, Make BJ, Crapo obstructive bronchitis. Am J Respir Crit Care Med.
JD, Silverman EK, Bowler RP, Dransfield MT, 1998;158:1511–7.
Investigators CO, Investigators ES. Pulmonary 210. Barker BL, Brightling CE. Phenotyping the hetero-
arterial enlargement and acute exacerbations of geneity of chronic obstructive pulmonary disease.
COPD. N Engl J Med. 2012;367:913–21. Clin Sci (Lond). 2013;124:371–87.
200. Woodruff PG, Agusti A, Roche N, Singh D, Martinez 211. Singh D, Kolsum U, Brightling CE, Locantore N,
FJ. Current concepts in targeting chronic obstruc- Agusti A, Tal-Singer R, investigators E. Eosinophilic
tive pulmonary disease pharmacotherapy: making inflammation in COPD: prevalence and clinical
progress towards personalised management. Lancet. characteristics. Eur Respir J. 2014;44:1697–700.
2015;385:1789–98. 212. Busse WW, Katial R, Gossage D, Sari S, Wang B,
201. Lotvall J, Akdis CA, Bacharier LB, Bjermer L, Kolbeck R, Coyle AJ, Koike M, Spitalny GL, Kiener
Casale TB, Custovic A, Lemanske RF Jr, Wardlaw PA, Geba GP, Molfino NA. Safety profile, phar-
AJ, Wenzel SE, Greenberger PA. Asthma endo- macokinetics, and biologic activity of MEDI-563,
types: a new approach to classification of disease an anti-IL-5 receptor alpha antibody, in a phase I
entities within the asthma syndrome. J Allergy Clin study of subjects with mild asthma. J Allergy Clin
Immunol. 2011;127:355–60. Immunol. 2010;125:1237–44. e1232
202. Vrijheid M. The exposome: a new paradigm to 213. Laviolette M, Gossage DL, Gauvreau G, Leigh
study the impact of environment on health. Thorax. R, Olivenstein R, Katial R, Busse WW, Wenzel S,
2014;69:876–8. Wu Y, Datta V, Kolbeck R, Molfino NA. Effects of
203. Christenson SA, Steiling K, van den Berge M, Hijazi benralizumab on airway eosinophils in asthmatic
K, Hiemstra PS, Postma DS, Lenburg ME, Spira A, patients with sputum eosinophilia. J Allergy Clin
Woodruff PG. Asthma-COPD overlap. Clinical rel- Immunol. 2013;132:1086–96. e1085
evance of genomic signatures of type 2 inflamma- 214. Brightling CE, Bleecker ER, Panettieri RA Jr,
tion in chronic obstructive pulmonary disease. Am Bafadhel M, She D, Ward CK, Xu X, Birrell C, van
J Respir Crit Care Med. 2015;191:758–66. der Merwe R. Benralizumab for chronic obstructive
204. Bafadhel M, McKenna S, Terry S, Mistry V, Reid pulmonary disease and sputum eosinophilia: a ran-
C, Haldar P, McCormick M, Haldar K, Kebadze T, domised, double-blind, placebo-controlled, phase 2a
Duvoix A, Lindblad K, Patel H, Rugman P, Dodson study. Lancet Respir Med. 2014;2:891–901.
P, Jenkins M, Saunders M, Newbold P, Green RH, 215. Barbu C, Iordache M, Man MG. Inflammation in
Venge P, Lomas DA, Barer MR, Johnston SL, COPD: pathogenesis, local and systemic effects.
Pavord ID, Brightling CE. Acute exacerbations of Rom J Morphol Embryol. 2011;52:21–7.
168 J. Sheth and M. Han

216. Nussbaumer-Ochsner Y, Rabe KF. Systemic mani- Fabbri LM, Pavord ID, Barnes N, Wouters EF,
festations of COPD. Chest. 2011;139:165–73. Agusti A. Moving from the Oslerian paradigm to the
217. Andelid K, Tengvall S, Andersson A, Levanen post-genomic era: are asthma and COPD outdated
B, Christenson K, Jirholt P, Ahren C, Qvarfordt I, terms? Thorax. 2014;69:72–9.
Ekberg-Jansson A, Linden A. Systemic cytokine 222. Loscalzo J, Kohane I, Barabasi AL. Human disease
signaling via IL-17 in smokers with obstructive pul- classification in the postgenomic era: a complex
monary disease: a link to bacterial colonization? Int systems approach to human pathobiology. Mol Syst
J Chron Obstruct Pulmon Dis. 2015;10:689–702. Biol. 2007;3:124.
218. Gan WQ, Man SF, Senthilselvan A, Sin 223. Burgel PR, Paillasseur JL, Caillaud D, Tillie-­
DD. Association between chronic obstructive pulmo- Leblond I, Chanez P, Escamilla R, Court-Fortune I,
nary disease and systemic inflammation: a systematic Perez T, Carre P, Roche N, Initiatives BSC. Clinical
review and a meta-analysis. Thorax. 2004;59:574–80. COPD phenotypes: a novel approach using prin-
219. Agusti A, Edwards LD, Rennard SI, MacNee W, Tal-­ cipal component and cluster analyses. Eur Respir
Singer R, Miller BE, Vestbo J, Lomas DA, Calverley J. 2010;36:531–9.
PM, Wouters E, Crim C, Yates JC, Silverman EK, 224. Weatherall M, Travers J, Shirtcliffe PM, Marsh
Coxson HO, Bakke P, Mayer RJ, Celli B, Evaluation SE, Williams MV, Nowitz MR, Aldington S,
of CLtIPSEI. Persistent systemic inflammation is Beasley R. Distinct clinical phenotypes of airways
associated with poor clinical outcomes in COPD: a disease defined by cluster analysis. Eur Respir
novel phenotype. PLoS One. 2012;7:e37483. J. 2009;34:812–8.
220. Agusti A, Anto JM, Auffray C, Barbe F, Barreiro E, 225. Paoletti M, Camiciottoli G, Meoni E, Bigazzi F,
Dorca J, Escarrabill J, Faner R, Furlong LI, Garcia-­ Cestelli L, Pistolesi M, Marchesi C. Explorative
Aymerich J, Gea J, Lindmark B, Monso E, Plaza V, data analysis techniques and unsupervised clustering
Puhan MA, Roca J, Ruiz-Manzano J, Sampietro-­ methods to support clinical assessment of Chronic
Colom L, Sanz F, Serrano L, Sharpe J, Sibila O, Obstructive Pulmonary Disease (COPD) pheno-
Silverman EK, Sterk PJ, Sznajder JI. Personalized types. J Biomed Inform. 2009;42:1013–21.
respiratory medicine: exploring the horizon, 226. Cho MH, Washko GR, Hoffmann TJ, Criner GJ,
addressing the issues. Summary of a BRN-AJRCCM Hoffman EA, Martinez FJ, Laird N, Reilly JJ,
workshop held in Barcelona on June 12, 2014. Am Silverman EK. Cluster analysis in severe emphy-
J Respir Crit Care Med. 2015;191:391–401. sema subjects using phenotype and genotype
221. Vanfleteren LE, Kocks JW, Stone IS, Breyer-­ data: an exploratory investigation. Respir Res.
Kohansal R, Greulich T, Lacedonia D, Buhl R, 2010;11:30.
Genetics of COPD
Woo Jin Kim

Introduction History of Genetic Studies for COPD

Although environmental factors, including ciga- Family studies have supported genetic factors to
rette smoking and biomass smoke exposure, are play an important role in the development of
major risk factors of COPD, genetic risk factors COPD [8]. Twin studies have reported heritabil-
are also important [1]. In addition, an interaction ity of lung function between 30 and 50% [9].
between genetics and environment is believed to Recently, heritability of COPD was estimated
drive the development of COPD. 35–40% in population-based study [10].
Pathways that play a role in COPD pathogen- Genome-wide linkage analysis using Boston
esis include the response to oxidative stress, the early onset COPD identified several loci that
protease–antiprotease imbalance, cell death, and were associated with lung function that is the
inflammation [2–5]. Genetic studies have been most important phenotype of COPD [11].
performed to identify genetic risk factors and to Candidate gene strategies were used to test
understand the pathogenesis of COPD. Family-­ hypothesis of genetic associations with
based studies and candidate gene association COPD. However, there were few genetic associa-
studies have found associations for many genes tions that were consistently significant, and this
and loci. However, alpha-1 antitrypsin deficiency strategy has limitation in identifying novel mech-
caused by mutations in SERPINA1 is the only anisms of COPD.
established genetically driven cause of COPD
that has a potential intervention so far [6].
Future research is needed to characterize the Genome-Wide Association Study
effect of genetic variants, validate gene function
in humans and model systems, and elucidate the Although whole-genome and exome sequencing
genes’ transcriptional and post-transcriptional may be the next tools used for the genetic study of
regulatory mechanisms [7]. COPD, genome-wide association study (GWAS) is
currently the most widely used method for the dis-
covery of candidate genes [12]. Several GWASs
have discovered novel genes and pathways that are
associated with COPD susceptibility. Even more
W.J. Kim genes have been found to be significantly associated
Department of Internal Medicine and Environmental with lung function in the general population. Some
Health Center, Kangwon National University, of these lung function genes are also associated with
Chuncheon, South Korea
e-mail: COPD ­susceptibility. The genetic basis of ­different

© Springer-Verlag Berlin Heidelberg 2017 169

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_11
170 W.J. Kim

COPD-related phenotypes, including emphysema CHRNA3/5 and CHRNB4 are subunits of the
and chronic bronchitis, also overlaps with that of nicotine cholinergic receptor, and the cholinergic
COPD susceptibility. After being implicated in dis- system is active not only in cholinergic neuronal
ease pathogenesis, these genes can be used as poten- cells, but also in bronchial epithelial cells and air-
tial drug targets or as biomarkers that can influence way inflammatory cells. The proteins are respon-
diagnosis and personalized treatment. sive to nicotine and are upregulated during
Currently, the most well-known candidate chronic tobacco exposure. A recent study inte-
genes for COPD are CHRNA3/5 (cholinergic nic- grating GWAS results with expression quantita-
otine receptor alpha 3/5), IREB2 (iron regulatory tive trait loci (eQTL) study results found that
binding protein 2), HHIP (hedgehog-interacting SNPs in the 15q25 region were associated with
protein), FAM13A (family with sequence similar- the expression of IREB2 and CHRNA3 in blood
ity 13, member A), and AGER (advanced glyco- and sputum samples [20]. CHRNA3/5 and IREB2
sylation end product-specific receptor). They may play different roles in the pathogenesis of
have been replicated in multiple populations. COPD.
None of them are targeted by treatments for IREB2 was first identified by characterizing
COPD yet, and the mechanisms by which they the differential gene expression in lung tissue
alter COPD risk are still largely unknown. There between COPD patients and controls, and geno-
is some emerging evidence that they may be good typing the SNPs within the candidate regions
targets for treatments or useful as biomarkers. [21]. IREB2 is a protein that binds iron-­responsive
However, more study is required to understand elements (IREs), maintains cellular iron metabo-
the functional roles of these candidate genes. lism, and is regulated in response to oxygen and
iron supply. IREB2 expression is higher in the
lung tissue of COPD cases. The Ireb2 knockout
CHRNA3, CHRNA5, and IREB2 mouse has abnormal iron metabolism in the
brain, which causes cellular dysfunction [22].
There are several genes at chromosome 15q25 However, the role of IREB2 in COPD pathogen-
that have been identified by GWAS for affecting esis is still not known. A GWAS of the pulmo-
COPD risk, including CHRNA3, CHRNA5, and nary artery measurement obtained by computed
IREB2 [13–15]. The COPD cohorts investigated tomography (CT) in cohorts from the COPDGene
were the Norway case/control cohort (GenKOLS), Study and the Evaluation of COPD Longitudinally
the family-based ICGN cohort, the NETT to Identify Predictive Surrogate Endpoints
(National Emphysema Treatment Trial)/NAS (ECLIPSE) study found a genome-wide signifi-
(normative aging study) cohorts, the Boston early cant association to IREB2 [23]. This suggests a
onset COPD cohort, and the COPDGene study role for IREB2 in the pathogenesis of pulmonary
cohort. The association between CHRNA3/5 and hypertension in COPD, particularly in the vascu-
COPD has been replicated in multiple ethnic lar subtype.
populations by direct genotyping [16–18]. The
CHRNA3/5 region is also associated with lung
cancer and nicotine addiction. It has been debated HHIP
whether this common susceptibility region is the
result of a common pathogenic pathway for lung Many recently identified COPD-associated vari-
cancer and COPD, or if it is simply associated ants are located at chromosome 4q31, upstream of
with nicotine addiction, a risk factor for both dis- HHIP. This intergenic region has associated with
eases. In addition, the causal variant within the COPD susceptibility in several GWASs [13, 14,
CHRNA3/5 locus may be different in lung cancer 24], and has consistently replicated in multiple
than in COPD. There is some evidence that this ethnicities [25–28]. This region is also a­ ssociated
locus has independent roles in the pathogenesis with lung function in the general population [29–
of COPD and smoking behavior [19]. 31]. HHIP is also associated with adult height in
11  Genetics of COPD 171

the general population [32]. Considering that the Hedgehog pathway is a critical mediator of ciga-
FEV1 prediction is determined by height, there rette smoke-induced lung cancer, and it may act
may be genetic factors that control both pheno- as a common pathway for the development of
types. Using the candidate gene strategy, it was COPD and lung cancer [40].
found that the HHIP gene interacted with envi-
ronmental tobacco smoke in utero, suggesting
that this gene is involved in the lung response to FAM13A
smoke exposure in early life [33].
HHIP encodes a membrane glycoprotein that A GWAS using three COPD cohorts, GenKOLS,
is an endogenous antagonist for the Hedgehog NETT/NAS, and the ECLIPSE study, identified
pathway. Hedgehog signaling is important for the variants at chromosome 4q22 in the gene FAM13A
morphogenesis of the lung and other organs [34]. [41]. These are some of the most highly associated
Although the role of HHIP in COPD is not fully SNPs in COPD and are located in an intron. These
understood, several studies have validated the associations have replicated in a subset of the
function of this gene in COPD pathogenesis. The patients in the COPDGene Study and the cohort of
associated SNPs are located upstream of HHIP, the International COPD Genetics Network. They
suggesting that they may affect promoter activity. also replicated in Asian populations, assayed using
A lung eQTL study revealed that SNPs associ- the candidate gene strategy [42, 43]. FAM13A was
ated with COPD affect the expression of HHIP, first found to associate with lung function in a
and the risk allele of rs1828591 decreases expres- GWAS using the general population [29], and it is
sion [35]. Zhou et al. reported that HHIP expres- associated with lung function in asthmatic subjects
sion is reduced in COPD lung tissues and the [44]. Of note, FAM13A is also associated with
genomic region upstream of HHIP interacts with idiopathic pulmonary disease (IPF) [45], but the
the HHIP promoter. The risk allele of a variant in expression of FAM13A in lung tissues does not dif-
the HHIP enhancer region reduces promoter fer by case/control status or by genotype.
activity via a differential binding affinity to tran- FAM13A was originally identified in cattle near
scription factors [36]. a quantitative trait locus affecting milk production,
These studies suggest that the genetic varia- and is expressed in the kidney, pancreas, lung, and
tion of the HHIP region affects the risk of COPD thymus [46]. Although the function of FAM13A
by affecting HHIP expression in lung tissues. has not been extensively studied, its RhoGAP
HHIP silencing in an airway epithelial cell line domain may be related to COPD. Rho GTPases
leads to a change in gene expression, and these are key regulators of cytoskeletal dynamics, are
differentially expressed genes are enriched in involved in the pulmonary endothelial barrier, and
pathways related to the extracellular matrix and are dysregulated in several lung diseases [47]. A
cell growth, which are processes relevant to lung eQTL study suggested that the expression of
COPD pathogenesis [37]. Recently, Lao et al. FAM13A may be associated with particular SNPs
found that Hhip-haploinsufficient mice have [35]. In the case of COPD, the FAM13A risk allele
increased airspace size after cigarette smoke is associated with increased FAM13A expression
exposure, increased lung compliance, and in the lung although expression does not differ in
increased numbers of lymphoid aggregates. The lung tissues between COPD cases and controls
functions of the genes with altered expression in [42]. A recent study by Jin et al. found that
Hhip+/− mice exposed to cigarette smoke were FAM13A activates Wnt signaling by increasing
enriched in the pathway of lymphocyte activation the stability of β-catenin [48]. Although depletion
[38]. They used haploinsufficient mice because of FAM13A in a lung cancer cell line reduces Wnt
Hhip−/− mice die shortly after birth due to lung signaling activity, FAM13A knockout mice are
branching morphogenesis failure. viable and FAM13A-mutant lungs are morpho­
HHIP was also found to be associated with logically indistinguishable from wild-type
lung cancer by a candidate gene study [39]. The lungs, and Wnt signaling remains normal in
172 W.J. Kim

Fam13a-­knockout lungs. They also found that Akt the airway and alveolar walls of COPD lungs
regulates the phosphorylation of FAM13A, which [57]. RAGE expression in mice increases after
can lead to cytoplasmic sequestration of FAM13A. cigarette smoke exposure, and cigarette smoking-­
Considering that Akt has a role in the pathogenesisinduced inflammatory responses by alveolar
of COPD [49], FAM13A may contribute to lung macrophages are diminished in RAGE knockout
disease through aberrant Akt signaling. Further mice [58]. Transgenic mice with upregulated
work is needed to validate the functional role of RAGE have impaired alveolar morphogenesis
FAM13A in the pathogenesis of COPD. during lung development, distal airspace enlarge-
ment, and increased alveolar cell apoptosis [59].
Another study using RAGE transgenic mice
AGER found incremental dilation of alveolar spaces, as
well as pronounced inflammation in the periph-
GWASs of lung function in the general popula- eral lung and alveolar destabilization [60]. A pro-
tion have found that chromosome 6p21 is associ- moter variant of AGER in cystic fibrosis patients
ated with FEV1/FVC and FEV1, which are is associated with poor lung function, and it
important physiologic parameters of COPD [31, increases expression in airway epithelial cell
29, 50]. This association was investigated in lines, suggesting that it is a modifier of lung dis-
COPD patients identified from the population ease severity [61].
cohort using spirometry criteria, and the study The soluble isoform of RAGE (sRAGE) con-
found a suggestive association between COPD tains the RAGE extracellular domain and can
risk and AGER, although it was not statistically bind to circulating proinflammatory ligands,
significant [51]. A candidate gene study in preventing RAGE activation. Mice that are
NETT/NAS, GenKOLS, ECLIPSE, and a subset exposed to chronic hypoxia have down-regu-
of the COPDGene Study cohort found that it is lated pulmonary RAGE protein and increased
associated with COPD susceptibility although a levels of sRAGE, which might be adaptive to
subsequent GWAS did not find a significant and protective against chronic hypoxia [62].
association [52]. On the other hand, an associa- Circulatory levels of sRAGE are reduced in
tion has been found in multiple ethnic popula- COPD patients [63]. Reduced sRAGE levels are
tions [53]. associated with increased emphysema in two
Chromosome 6p21 region that showed signifi- COPD cohorts [64]. Decreased plasma sRAGE
cant association with COPD includes many levels are also associated with the progression
genes: TNXB, PPT2, AGER, and NOTCH4. of airflow limitation over time [65]. In patients
However, AGER has a potential functional vari- of the Treatment of Emphysema with a Selective
ant, rs2070600, and has been studied the most in Retinoid Agonist (TESRA) and ECLIPSE stud-
the pathogenesis of COPD. A GWAS of percent ies, sRAGE is associated with diffusing capac-
emphysema determined by CT using the Multi-­ ity, emphysema, and COPD disease status, and
Ethnic Study of Atherosclerosis cohort identified the variant rs2070600 is associated with circu-
a significant association with the AGER/PPT lating sRAGE levels [66]. The significant asso-
region [54]. This region did associate with ciation between rs2070600 and plasma sRAGE
emphysema severity and gas trapping in a GWAS levels was also found in Dutch diabetes mellitus
using cohorts from the COPDGene, ECLIPSE, and control subjects [67]. RAGE has been stud-
NETT, and GenKOLS studies [55]. ied in metabolic diseases, and decreased levels
The protein product of AGER, the receptor for of sRAGE are linked to ­vascular complications.
advanced glycan end-products (RAGE), is a RAGE contributes to the pathogenesis of COPD
multi-ligand receptor of the immunoglobulin in the lung probably via the regulation of inflam-
superfamily and interacts with molecules impli- mation and apoptosis, and further study of the
cated in homeostatic function, inflammation, and functions of this gene may lead to it being iden-
development [56]. RAGE levels are increased in tified as a potential therapeutic target.
11  Genetics of COPD 173

Other Candidate Genes The CHRNA3/5 locus is associated with emphy-

sema and smoking intensity in COPD [76, 77].
There have been several more regions identified HHIP is associated with various CT pheno-
in GWASs of COPD. A GWAS using subjects types in COPD including distinct patterns of
from the ECLIPSE, NETT/NAS, GenKOLS, and emphysema [77] and the severity of emphysema
COPDGene studies identified chromosome [55]. HHIP is more associated with emphysema
19q13 as being associated with COPD, along measurements than with airway phenotypes and
with the previously identified HHIP, FAM13A, has a more significant association in emphysema
and 15q25 regions [14]. Chromosome 19q13 subgroups [78]. This difference may reflect a dif-
contains CYP2A6, RAB4B, MIA, and EGLN, ferent pathogenic process driven by HHIP, or
which could potentially be involved in COPD may be driven by correlations between COPD
pathogenesis, and EGLN2 was found to be dys- status and imaging measurements.
regulated in the airway epithelium of smokers Genome-wide association studies using COPD
[68]. A GWAS using the full COPDGene cohort cases with chronic bronchitis in the COPDGene
identified additional associations with TGFB2, Study, GenKOLS, and ECLIPSE cohorts identi-
MMP12, and RIN3 [24]. TGFB2 and MMP12 fied a significant association with FAM13A [79],
have been previously studied in COPD or related whereas several GWASs for emphysema did not
phenotypes [69, 70], whereas RIN3 has not been identify a genome-wide association. The odds
studied in COPD and needs to be investigated ratios of FAM13A SNPs for COPD with chronic
further. SERPINE2 was identified using a linkage bronchitis were significantly higher than those for
analysis of gene expression changes in lung tis- non-chronic bronchitis COPD, suggesting that
sue [71]. A recent GWAS of airway thickness FAM13A is more related to the pathogenesis of
identified rs734556 on chromosome 2q, which is the chronic bronchitis subtype.
associated with SERPINE2 expression [72]. GWAS in the presence of emphysema identi-
These associations require more replications and fied BICD as a susceptibility gene for emphy-
further fine-mapping studies are needed to find sema [80]. GWAS of percent emphysema in the
the causal variants of COPD, as well as studies to general population identified SNRPF and PPT2
functionally validate the identified genes. [54]. GWAS on airway wall thickness MAGI2
and NT5C3B were associated with airway wall
thickness [72].
GWAS for Heterogeneity As pulmonary hypertension is a well-­
established complication and an important factor
CT phenotypes including emphysema severity and of prognosis, GWAS of pulmonary artery
airway thickness quantitatively measured using enlargement have found IREB2 and GALC asso-
standardized methods are useful in understanding ciated with pulmonary artery enlargement defined
heterogeneity of COPD by characterizing lung as PA/A ratio more than 1 in COPD subjects [23].
parenchyma and airways. Previous study using BMI is important in prognosis of COPD. The
candidate gene approach reported that associations HHIP locus is associated with fat-free mass and
between SERPINE2 and upper lobe dominance exacerbations in COPD subjects [76]. GWAS on
[73], ADRB2 and airway lumen area [74]. Another BMI in COPD identified FTO was associated
study reported that EPHX1, SERPINE2, and with BMI and fat-free mass index [81].
GSTP1 were associated with emphysema severity
were associated with airway phenotypes [75]. Pharmacogenetics
After GWAS identified several COPD-­
associated genes, those identified genes were Recently, genotype variation can be used to
tested for CT phenotypes, and also GWAS was individualized therapy. In COPD, the most
performed on CT phenotypes. studied subject is ADRB2 polymorphisms of
174 W.J. Kim

β2-adrenergic receptor on β2 agonist therapy 4. Kirkham PA, Barnes PJ. Oxidative stress in

COPD. Chest. 2013;144:266–73.
[82, 83]. Another gene included CRHR1 poly-
5. Bagdonas E, Raudoniute J, Bruzauskaite I, et al.
morphism [84]. Warfarin is a good example of Novel aspects of pathogenesis and regeneration
individual variation in pharmacokinetics; how- mechanisms in COPD. Int J Chron Obstruct Pulmon
ever, drug used in COPD are not known for Dis. 2015;10:995–1013.
6. Silverman EK, Sandhaus RA. Alpha1-antitrypsin
gene influencing drug metabolism.
deficiency. N Engl J Med. 2009;360:2749–57.
Previous study reported that COPD candi- 7. Ober C, Butte AJ, Elias JA, et al. Getting from genes
date genes may influence bronchodilator to function in lung disease. Am J Respir Crit Care
responsiveness [85]. Considering that treat- Med. 2010;182:732–7.
8. Silverman EK, Chapman HA, Drazen JM, et al.
ment with PDE4 inhibitor is effective only for
Genetic epidemiology of severe, early-onset chronic
the chronic bronchitis subtype, there may be a obstructive pulmonary disease. Am J Respir Crit Care
mechanism that is unique to this subtype. Med. 1998;157:1770–8.
Candidate genes can be used to determine per- 9. Ingebrigtsen T, Thomsen S, van der Sluis S, et al.
Genetic influences on pulmonary function: a large
sonalized treatment because they may help
sample twin study. Lung. 2011;189:323–30.
identify a subtype-unique pathogenesis, as 10. Zhou JJ, Cho MH, Castaldi PJ, et al. Heritability of
well as variation in a drug-action site, or vari- chronic obstructive pulmonary disease and related
able drug metabolism [86]. phenotypes in smokers. Am J Respir Crit Care Med.
11. Silverman EK, Palmer LJ, Mosley JD, et al.

Conclusion Genomewide linkage analysis of quantitative
Recently, several candidate genes associated Spirometric phenotypes in severe early-onset chronic
with COPD risk have been identified using obstructive pulmonary disease. Am J Hum Genet.
GWAS. Replication and functional validation
12. Todd JL, Goldstein DB, Ge D, et al. The state of
studies may lead to clinical applications for genome-wide association studies in pulmonary dis-
these genes such as novel therapeutics, sub- ease. Am J Respir Crit Care Med. 2011;184:873–80.
typing, and risk prediction for COPD. Also, 13. Pillai SG, Ge D, Zhu G, et al. A genome-wide associ-
ation study in chronic obstructive pulmonary disease
phenotype heterogeneity can be investigated
(COPD): identification of two major susceptibility
using association studies on various COPD- loci. PLoS Genet. 2009;5:e1000421.
related phenotypes. More regions have been 14. Cho MH, Castaldi PJ, Wan ES, et al. A genome-wide
identified in GWASs on FEV1 and FEV1/FVC association study of COPD identifies a susceptibil-
ity locus on chromosome 19q13. Hum Mol Genet.
in the general population, probably because of
the larger sample sizes than COPD case/con- 15. Wilk JB, Shrine NR, Loehr LR, et al. Genome-wide
trol subjects. GWASs using a greater sample association studies identify CHRNA5/3 and HTR4 in
size of COPD subjects may find more candi- the development of airflow obstruction. Am J Respir
Crit Care Med. 2012;186:622–32.
date genes [87]. Another approach for finding
16. Hardin M, Zielinski J, Wan ES, et al. CHRNA3/5,
more candidate genes is to identify rare varia- IREB2, and ADCY2 are associated with severe
tion using exome sequencing or arrays. chronic obstructive pulmonary disease in Poland. Am
J Respir Cell Mol Biol. 2012;47:203–8.
17. Kim WJ, Oh YM, Kim TH, et al. CHRNA3 variant
for lung cancer is associated with chronic obstruc-
tive pulmonary disease in Korea. Respiration.
References 2013;86:117–22.
18. Yang L, Qiu F, Lu X, et al. Functional polymorphisms
1. Postma DS, Bush A, van den Berge M. Risk factors of CHRNA3 predict risks of chronic obstructive pul-
and early origins of chronic obstructive pulmonary monary disease and lung cancer in Chinese. PLoS
disease. Lancet. 2015;385:899–909. One. 2012;7:e46071.
2. Cosio MG, Saetta M, Agusti A. Immunologic aspects 19. Siedlinski M, Tingley D, Lipman PJ, et al. Dissecting
of chronic obstructive pulmonary disease. N Engl direct and indirect genetic effects on chronic obstruc-
J Med. 2009;360:2445–54. tive pulmonary disease (COPD) susceptibility. Hum
3. Tuder R, Petrache I. Pathogenesis of chronic obstruc- Genet. 2013;132:431–41.
tive pulmonary disease. J Clin Invest. 2012;122: 20. Castaldi PJ, Cho MH, Zhou X, et al. Genetic control
2749–55. of gene expression at novel and established chronic
11  Genetics of COPD 175

obstructive pulmonary disease loci. Hum Mol Genet. 36. Zhou X, Baron RM, Hardin M, et al. Identification
2015;24:1200–10. of a chronic obstructive pulmonary disease genetic
21. DeMeo DL, Mariani T, Bhattacharya S, et al.
determinant that regulates HHIP. Hum Mol Genet.
Integration of genomic and genetic approaches impli- 2012;21:1325–35.
cates IREB2 as a COPD susceptibility Gene. Am 37. Zhou X, Qiu W, Sathirapongsasuti JF, et al. Gene
J Hum Genet. 2009;85:493–502. expression analysis uncovers novel hedgehog inter-
22. Zumbrennen-Bullough KB, Becker L, Garrett L, et al. acting protein (HHIP) effects in human bronchial epi-
Abnormal brain iron metabolism in Irp2 deficient thelial cells. Genomics. 2013;101:263–72.
mice is associated with mild neurological and behav- 38. Lao T, Glass K, Qiu W, et al. Haploinsufficiency
ioral impairments. PLoS One. 2014;9:e98072. of hedgehog interacting protein causes increased
23. Lee JH, Cho MH, Hersh CP, et al. IREB2 and GALC emphysema induced by cigarette smoke through net-
are associated with pulmonary artery enlargement in work rewiring. Genome Med. 2015;7:12.
chronic obstructive pulmonary disease. Am J Respir 39. Young RP, Whittington CF, Hopkins RJ, et al.

Cell Mol Biol. 2015;52:365–76. Chromosome 4q31 locus in COPD is also asso-
24. Cho MH, McDonald ML, Zhou X, et al. Risk loci for ciated with lung cancer. Eur Respir J. 2010;36:
chronic obstructive pulmonary disease: a genome-­ 1375–82.
wide association study and meta-analysis. Lancet 40. Lemjabbar-Alaoui H, Dasari V, Sidhu SS, et al. Wnt
Respir Med. 2014;2:214–25. and hedgehog are critical mediators of cigarette
25. Van Durme YM, Eijgelsheim M, Joos GF, et al.
smoke-induced lung cancer. PLoS One. 2006;1:e93.
Hedgehog-interacting protein is a COPD susceptibility 41. Cho MH, Boutaoui N, Klanderman BJ, et al. Variants
gene: the Rotterdam study. Eur Respir J. 2010;36:89–95. in FAM13A are associated with chronic obstructive
26. Kim WJ, Oh YM, Lee JH, et al. Genetic variants pulmonary disease. Nat Genet. 2010;42:200–2.
in HHIP are associated with FEV1 in subjects with 42. Kim WJ, Lim MN, Hong Y, et al. Association of lung
chronic obstructive pulmonary disease. Respirology. function genes with chronic obstructive pulmonary
2013;18:1202–9. disease. Lung. 2014;192:473–80.
27. Xie J, Wu H, Xu Y, et al. Gene susceptibility identi- 43. Guo Y, Lin H, Gao K, et al. Genetic analysis of IREB2,
fication in a longitudinal study confirms new loci in FAM13A and XRCC5 variants in Chinese Han
the development of chronic obstructive pulmonary patients with chronic obstructive pulmonary disease.
disease and influences lung function decline. Respir Biochem Biophys Res Commun. 2011;415:284–7.
Res. 2015;16:49. 44. Li X, Hawkins GA, Ampleford EJ, et al. Genome-­
28. Wang B, Zhou H, Yang J, et al. Association of HHIP wide association study identifies TH1 pathway genes
polymorphisms with COPD and COPD-related associated with lung function in asthmatic patients.
phenotypes in a Chinese Han population. Gene. J Allergy Clin Immunol. 2013;132:313–20, e315.
2013;531:101–5. 45. Fingerlin TE, Murphy E, Zhang W, et al. Genome-­
29. Hancock DB, Eijgelsheim M, Wilk JB, et al. Meta-­ wide association study identifies multiple suscep-
analyses of genome-wide association studies identify tibility loci for pulmonary fibrosis. Nat Genet.
multiple loci associated with pulmonary function. Nat 2013;45:613–20.
Genet. 2010;42:45–52. 46. Cohen M, Reichenstein M, Everts-van der Wind A,
30. Wilk JB, Chen TH, Gottlieb DJ, et al. A genome-­ et al. Cloning and characterization of FAM13A1—a
wide association study of pulmonary function mea- gene near a milk protein QTL on BTA6: evidence
sures in the Framingham heart study. PLoS Genet. for population-wide linkage disequilibrium in Israeli
2009;5:e1000429. Holsteins. Genomics. 2004;84:374–83.
31. Repapi E, Sayers I, Wain LV, et al. Genome-wide 47. Corvol H, Hodges CA, Drumm ML, et al. Moving
association study identifies five loci associated with beyond genetics: is FAM13A a major biological con-
lung function. Nat Genet. 2010;42:36–44. tributor in lung physiology and chronic lung diseases?
32. van der Valk RJP, Kreiner-Møller E, Kooijman MN, J Med Genet. 2014;51:646–9.
et al. A novel common variant in DCST2 is associated 48. Jin Z, Chung JW, Mei W, et al. Regulation of nuclear-­
with length in early life and height in adulthood. Hum cytoplasmic shuttling and function of family with
Mol Genet. 2015;24:1155–68. sequence similarity 13, member a (Fam13a), by
33. Kerkhof M, Boezen HM, Granell R, et al. Transient B56-containing PP2As and Akt. Mol Biol Cell.
early wheeze and lung function in early childhood 2015;26:1160–73.
associated with chronic obstructive pulmonary disease 49. Bozinovski S, Vlahos R, Hansen M, et al. Akt in the
genes. J Allergy Clin Immunol. 2014;133:68–76.e64. pathogenesis of COPD. Int J Chron Obstruct Pulmon
34. Chuang PT, Kawcak T, McMahon AP. Feedback
Dis. 2006;1:31–8.
control of mammalian hedgehog signaling by the 50. Kim WJ, Lee MK, Shin C, et al. Genome-wide asso-
hedgehog-­binding protein, Hip1, modulates Fgf sig- ciation studies identify locus on 6p21 influencing
naling during branching morphogenesis of the lung. lung function in the Korean population. Respirology.
Genes Dev. 2003;17:342–7. 2014;19:360–8.
35. Lamontagne M, Couture C, Postma DS, et al. Refining 51. Soler Artigas M, Wain LV, Repapi E, et al. Effect of
susceptibility loci of chronic obstructive pulmonary five genetic variants associated with lung function on
disease with lung eqtls. PLoS One. 2013;8:e70220. the risk of chronic obstructive lung disease, and their
176 W.J. Kim

joint effects on lung function. Am J Respir Crit Care 66. Cheng DT, Kim DK, Cockayne DA, et al. Systemic
Med. 2011;184:786–95. soluble receptor for advanced glycation endproducts is
52. Castaldi PJ, Cho MH, Litonjua AA, et al. The associa- a biomarker of emphysema and associated with AGER
tion of genome-wide significant spirometric loci with genetic variants in patients with chronic obstructive pul-
chronic obstructive pulmonary disease susceptibility. monary disease. Am J Respir Crit Care Med. 2013;188:
Am J Respir Cell Mol Biol. 2011;45:1147–53. 948–57.
53. Li Y, Yang C, Ma G, et al. Association of
67. Gaens KHJ, Ferreira I, van der Kallen CJH, et al.
Polymorphisms of the receptor for advanced glyca- Association of polymorphism in the receptor for
tion end products Gene with COPD in the Chinese advanced glycation end products (RAGE) Gene with
population. DNA Cell Biol. 2014;33:251–8. circulating RAGE levels. J Clin Endocrinol Metabol.
54. Manichaikul A, Hoffman EA, Smolonska J, et al. 2009;94:5174–80.
Genome-wide study of percent emphysema on com- 68. Ryan DM, Vincent TL, Salit J, et al. Smoking dys-
puted tomography in the general population. The regulates the human airway basal cell transcriptome at
multi-ethnic study of atherosclerosis lung/SNP health COPD risk locus 19q13.2. PLoS One. 2014;9:e88051.
association resource study. Am J Respir Crit Care 69. Hunninghake GM, Cho MH, Tesfaigzi Y, et al.

Med. 2014;189:408–18. MMP12, lung function, and COPD in high-risk popu-
55. Cho MH, Castaldi PJ, Hersh CP, et al. A genome-wide lations. N Engl J Med. 2009;361:2599–608.
association study of emphysema and airway quantita- 70. Artigas MS, Loth DW, Wain LV, et al. Genome-­

tive imaging phenotypes. Am J Respir Crit Care Med. wide association and large-scale follow up identifies
2015;192(5):559–69. 16 new loci influencing lung function. Nat Genet.
56. Sorci G, Riuzzi F, Giambanco I, et al. RAGE in tis- 2011;43:1082–90.
sue homeostasis, repair and regeneration. Biochim 71. DeMeo DL, Mariani TJ, Lange C, et al. The SERPINE2
Biophys Acta. 2013;1833:101–9. gene is associated with chronic obstructive pulmonary
57. Wu L, Ma L, Nicholson LFB, et al. Advanced gly- disease. Am J Hum Genet. 2006;78:253–64.
cation end products and its receptor (RAGE) are 72. Dijkstra AE, Postma DS, van Ginneken B, et al.
increased in patients with COPD. Respir Med. Novel genes for airway wall thickness identified
2011;105:329–36. with combined genome-wide association and expres-
58. Robinson AB, Johnson KD, Bennion BG, et al.
sion analyses. Am J Respir Crit Care Med. 2014;
RAGE signaling by alveolar macrophages influences 191:547–56.
tobacco smoke-induced inflammation. Am J Physiol 73. DeMeo DL, Hersh CP, Hoffman EA, et al. Genetic
Lung Cell Mol Physiol. 2012;302(11):L1192–9. determinants of emphysema distribution in the
59. Fineschi S, De Cunto G, Facchinetti F, et al. Receptor National Emphysema Treatment Trial. Am J Respir
for advanced glycation end products contributes to Crit Care Med. 2007;176:42–8.
postnatal pulmonary development and adult lung 74. Kim WJ, Oh Y-M, Sung J, et al. CT scanning-based
maintenance program in mice. Am J Respir Cell Mol phenotypes vary with ADRB2 polymorphisms in
Biol. 2013;48:164–71. chronic obstructive pulmonary disease. Respir Med.
60. Stogsdill MP, Stogsdill JA, Bodine BG, et al.
Conditional overexpression of receptors for advanced 75. Kim WJ, Hoffman E, Reilly J, et al. Association of
glycation end-products in the adult murine lung COPD candidate genes with computed ­tomography
causes airspace enlargement and induces inflamma- emphysema and airway phenotypes in severe
tion. Am J Respir Cell Mol Biol. 2013;49:128–34. COPD. Eur Respir J. 2011;37:39–43.
61. Beucher J, Boëlle P-Y, Busson P-F, et al. AGER-­ 76. Pillai SG, Kong X, Edwards LD, et al. Loci identified
429T/C is associated with an increased lung disease by genome-wide association studies influence differ-
severity in cystic fibrosis. PLoS One. 2012;7:e41913. ent disease-related phenotypes in chronic obstruc-
62. Gopal P, Gosker HR, de Theije CC, et al. Effect of tive pulmonary disease. Am J Respir Crit Care Med.
chronic hypoxia on RAGE and its soluble forms in 2010;182:1498–505.
lungs and plasma of mice. Biochim Biophys Acta 77. Castaldi PJ, Cho MH, San Jose Estepar R, et al. Genome-
2015;1852:992–1000 wide association identifies regulatory loci associated
63. Miniati M, Monti S, Basta G, et al. Soluble receptor with distinct local histogram emphysema patterns. Am
for advanced glycation end products in COPD: rela- J Respir Crit Care Med. 2014;190:399–409.
tionship with emphysema and chronic cor pulmonale: 78. Castaldi PJ, Dy J, Ross J, et al. Cluster analysis in the
a case-control study. Respir Res. 2011;12:37. COPDGene study identifies subtypes of smokers with
64. Carolan B, Hughes G, Morrow J, et al. The asso- distinct patterns of airway disease and emphysema.
ciation of plasma biomarkers with computed Thorax. 2014;69:416–23.
tomography-­assessed emphysema phenotypes. Respir 79. Lee JH, Cho MH, Hersh CP, et al. Genetic suscep-
Res. 2014;15:127. tibility for chronic bronchitis in chronic obstructive
65. Iwamoto H, Gao J, Pulkkinen V, et al. Soluble recep- pulmonary disease. Respir Res. 2014;15:113.
tor for advanced glycation end-products and progres- 80. Kong X, Cho MH, Anderson W, et al. Genome-wide
sion of airway disease. BMC Pulm Med. 2014;14:68. association study identifies BICD1 as a susceptibil-
11  Genetics of COPD 177

ity gene for emphysema. Am J Respir Crit Care Med. 84. Kim WJ, Sheen SS, Kim T-H, et al. Association

2011;183:43–9. between CRHR1 polymorphism and improved lung
81. Wan ES, Cho MH, Boutaoui N, et al. Genome-wide function in response to inhaled corticosteroid in
association analysis of body mass in chronic obstruc- patients with COPD. Respirology. 2009;14:260–3.
tive pulmonary disease. Am J Respir Cell Mol Biol. 85. Kim WJ, Hersh CP, DeMeo DL, et al. Genetic associa-
2011;45:304–10. tion analysis of COPD candidate genes with broncho-
82. Kim W, Oh Y-M, Sung J, et al. Lung function response dilator responsiveness. Respir Med. 2009;103:552–7.
to 12-week treatment with combined inhalation of 86. Hersh CP. Pharmacogenetics of chronic obstructive
long-acting β2 agonist and glucocorticoid according pulmonary disease: challenges and opportunities.
to ADRB2 polymorphism in patients with chronic Pharmacogenomics. 2010;11:237–47.
obstructive pulmonary disease. Lung. 2008;186:381–6. 87. Silverman EK, Vestbo J, Agusti A, et al. Opportunities
83. Bleecker ER, Meyers DA, Bailey WC, et al. ADrb2 and challenges in the genetics of COPD 2010: an
polymorphisms and budesonide/formoterol responses international COPD genetics conference report.
in COPD. Chest. 2012;142:320–8. COPD. 2011;8:121–35.
Imaging Heterogeneity of COPD
Sang Min Lee and Joon Beom Seo

Overview In this chapter, we reviewed previous research on

imaging in COPD patients briefly and addressed
Although chronic obstructive pulmonary disease the current concept and future direction of imag-
(COPD) is defined as persistent airflow limitation ing phenotyping.
which is usually progressive and associated with
an enhanced chronic inflammatory response [1],
heterogeneity of COPD has been realized while  T: Airway vs. Emphysema
our understanding of this agonizing disease has Predominance
grown over the past two decades [2]. Patients
with same or similar pulmonary function impair- The concept that COPD phenotype can be divided
ment show different symptoms, disease progres- according to varying combination and severity of
sion, and prognosis. As a result, the concept of emphysema and small airway disease on CT was
COPD has been changing from an airflow firstly suggested by Nakano et al. [5]. Initially,
limitation-­centric view to a complex and hetero- Nakano et al. showed that CT could quantify air-
geneous condition, preferring multidimensional way abnormalities in 114 smokers [6]. They
approaches and finding phenotypes. In this con- demonstrated the accuracy and reproducibility of
text, imaging features, especially quantitative quantitative airway measurement on CT and
analysis of CT, have garnered attention as they revealed that both quantitative analyses of airway
have been demonstrated to be of help in evaluat- and emphysema on CT were useful and comple-
ing patients’ status as well as predicting acute mentary in the evaluation of patients with COPD
exacerbation and prognosis of patients. The two [6]. This technical advance allows to evaluate
main components of COPD, emphysema and structural change due to airway inflammation and
small airway disease, can be accurately and reli- remodeling in vivo. Nakano et al. finally sug-
ably assessed by quantitative CT analysis [3, 4]. gested that COPD patients can be divided into
groups who had predominant emphysema or
thickening and narrowing of the apical segmental
bronchus using quantitative assessment of rela-
S.M. Lee, M.D. • J.B. Seo, M.D., Ph.D. (*) tive area of low parenchymal attenuation and per-
Division of Cardiothoracic Radiology, Department of cent airway wall area. In Korean Obstructive
Radiology, Asan Medical Center, University of Ulsan Lung Disease (KOLD) Cohort study [7], 530
College of Medicine, Ulsan, South Korea
patients also demonstrate a similar distribution
e-mail:; (Fig. 12.1).

© Springer-Verlag Berlin Heidelberg 2017 179

S.-D. Lee (ed.), COPD, DOI 10.1007/978-3-662-47178-4_12
180 S.M. Lee and J.B. Seo

Emphysema Index (%)





50 60 70 80
Wall Area (WA%)

Fig. 12.1  Emphysema index and airway wall thickening indicates the mean + 2 standard deviation of wall area of
in Korean Obstructive Lung Disease (KOLD) Cohort. the normal individuals (73.0%). COPD patients can be
This graph demonstrates the distribution of 497 COPD categorized as different groups using these thresholds:
patients and 33 normal individuals according to emphy- airway-predominant, emphysema-predominant, and a
sema index and airway wall thickening. Horizontal line mixed group. In KOLD cohort, the emphysema-­
indicates the mean + 2 standard deviation of emphysema predominant group comprises a large proportion of COPD
index of the normal individuals (12.8%). Vertical line patients

This concept of dividing patients based on CT be defined as pixels less than 6% of −950 HU at
imaging characteristics evolves and expands in quantitative CT. Although this classification is
recent article by Fleischner society [8]. They sug- tentative and much area remains for future
gested seven different subtypes of COPD research, the key concept that two components of
patients: five different patterns of emphysema-­ emphysema and airway disease mainly consist of
predominant subtype according to severity and COPD will be effective.
location of emphysema (mild centrilobular In addition, technical advances in imaging
emphysema, moderate centrilobular emphysema, processing enable novel and detailed quantifica-
confluent emphysema, advanced destructive tion of COPD and provide imaging-biomarkers
emphysema, panlobular emphysema, and para- for diagnosis of COPD phenotypes and disease
septal emphysema) and two patterns of airway-­ progression. Galban et al. [3] demonstrated more
predominant subtype according to level of clear separation of emphysema and functional
involved airways (bronchial disease and small small airway disease using non-rigid registration
airway disease). The most important factor dif- between inspiratory and expiratory CT images.
ferentiating subtypes of COPD is quantitative Furthermore, Kim et al. [4] showed a new
amount of emphysema, that is, emphysema-­ approach to quantifying air-trapping using a co-­
predominant subgroup can be defined as more registration method which defined air-trapping as
than 6% of pixels less than −950 HU at quantita- a volume with attenuation increase less than
tive CT and airway-predominant subgroup can 50 HU between inspiration and expiration CT
12  Imaging Heterogeneity of COPD 181

a b

c d

Fig. 12.2  Co-registration of inspiration and expiration tion between inspiration and expiration CT was performed
CT scans for analysis on air-trapping. (a) Inspiration CT in co-registered images (c, d). These images allow to eas-
and (b) expiration CT were obtained in a 77-year-old male ily detect the areas of air-trapping with small attenuation
patient. Expiration CT was deformed and registered to the change on respiration. Red color represents voxels with a
inspiration CT using a non-rigid registration method. difference of voxel attenuation between inspiration and
Color mapping according to differences of voxel attenua- expiration CT below a threshold of 50

(Fig. 12.2). Using this approach, respective con- study by Kitaguchi et al. [9], they subjectively
tributions of different densities seen on ­inspiration classified into three phenotypes of 85 COPD
CT to air-trapping could be assessed in detail. patients according to components of emphysema
The previous two studies quantified air-­trapping and bronchial wall thickening on CT; airway-­
area rather than airway wall for evaluation of predominant, emphysema-dominant, and mixed
small airway disease as it is difficult to quantify airway and emphysema. They assessed area of
small airway disease because there are limita- emphysema and airway wall thickness visually.
tions in the direct visualization of small airways Interestingly, three CT phenotypes showed differ-
(diameter < 2 mm) on CT. ent characteristics in terms of body mass index,
onset of dyspnea, proportion of never-­smoker, and
dependency of long-term oxygen therapy. This
Prediction of Treatment Response implies that CT phenotyping can explain clinical
features. More importantly, CT phenotyping was
One of important roles of CT phenotyping is to significantly related to treatment response with
predict treatment response in COPD patients. In a inhaled ß2-agonist and corticosteroid. Airway-
182 S.M. Lee and J.B. Seo

predominant phenotype showed significantly investigate a relationship between quantitative

greater reversibility with inhaled ß2-agonist CT measures and COPD exacerbation frequency.
(change in FEV1: airway-­predominant, 253.3 mL; According to their study, greater extent of emphy-
emphysema-­predominant, 94.0 mL; mixed pheno- sema and airway wall thickness was associated
type, 133.7 mL). Airway-predominant phenotype with COPD exacerbations, irrespective of the
and mixed phenotype were significantly associ- severity of airflow obstruction. Importantly, mean
ated with improvement in FEV1 when using segmental bronchial wall thickness showed the
inhaled corticosteroid (change in FEV1: airway-­ highest odds ratio (1.84) among significant vari-
predominant, 313.9 mL; emphysema-­predominant, ables on multivariate analysis. This result corre-
116.2 mL; mixed phenotype, 247.9 mL). This sponds well with essential role of airway
result suggests that bronchial wall thickening on inflammation and remodeling in development
CT may be an indicator for predicting good and progression of COPD [14, 15]. Based on Han
response to treatment. However, semiquantitative et al.’s study, we can identify subgroups of
evaluation has limitation due to requirement for patients who experience exacerbations more fre-
expert radiologists and interobserver variation. quently and subsequently provide more personal-
Similar study was performed using a quantita- ized therapy.
tive method in KOLD cohort by Lee et al. [10].
They objectively categorized 165 patients into
four subtypes using extent of emphysema on CT Regional Heterogeneity
and FEV1: emphysema-dominant, obstruction-­ of Emphysema
dominant, mild-mixed subtype, and severe-mixed
subtype. They also reported that obstruction-­ Emphysema can vary according to subtypes (cen-
dominant and mixed subtypes showed signifi- trilobular, paraseptal, and panlobular) and
cantly greater improvement in FEV1 than regional distribution. It is known that regional
emphysema-dominant group after 3 months of heterogeneity of emphysema in anterior-­posterior
combined inhalation of long-acting beta-agonist and upper-lower direction was independent
and corticosteroid. Furthermore, obstruction-­ determinants of FEV1 and FEV1/FVC and the
dominant subtype patients showed marked lower and posterior regional dominant emphy-
improvement of dyspnea compared with sema is associated with a decrease in FEV1 and
emphysema-­dominant patients. Given the results FEV1/FVC [16]. Basal distribution of emphy-
of two studies, we can safely tell that bronchial sema is also associated with greater impairment
wall thickening on CT should be assessed to pre- of FEV1 [17].
dict treatment response before treatment. Such regional heterogeneity of emphysema
has clinical relevance to the treatment of
COPD. According to results of a randomized trial
Prediction of Acute Exacerbation for lung volume reduction surgery (LVRS) [18],
it has a survival gain for patients with both pre-
Acute exacerbation of COPD is defined as acute dominantly upper-lobe emphysema and low
event characterized by a worsening of patients’ base-line exercise capacity even though upper-­
respiratory symptoms that is beyond normal day-­ lobe predominance of emphysema was visually
to-­day variations and leads to a change in medi- determined by each center’s radiologist. Thus,
cation [1]. It is known that all-cause mortality assessment of regional heterogeneity of emphy-
3 years after hospitalization due to acute exacer- sema using CT is important and useful for select-
bation is as high as 49% [11]. Therefore, early ing candidates for LVRS. In addition, even in
detection and prompt treatment of acute exacer- endobronchial valves for advanced emphysema,
bations as well as prevention are crucial to reduce heterogeneity of emphysema on CT is the criteria
the burden of COPD [12]. Regarding acute exac- for selecting patients [19]. In Sciurba et al.’s
erbation, Han et al. [13] performed a study to study [19], the percentage of heterogeneity was
12  Imaging Heterogeneity of COPD 183

defined as the difference in the quantitative morbidity and mortality in COPD patients [23].
emphysema score (the proportion of pixels of The mechanisms for this process likely include
less than −910 Hounsfield units) between the tar- inflammation or hypoxic vasoconstriction due
geted lobe and the ipsilateral adjacent nontar- to emphys