Accepted Manuscript

Smoking-related Lung Disease

Brett M. Elicker MD , Kimberly G. Kallianos MD ,

Kirk D. Jones MD , Travis S. Henry MD

PII: S0887-2171(18)30126-4
DOI: https://doi.org/10.1053/j.sult.2018.11.010
Reference: YSULT 856

To appear in: Seminars in Ultrasound CT and MRI

Please cite this article as: Brett M. Elicker MD , Kimberly G. Kallianos MD , Kirk D. Jones MD ,
Travis S. Henry MD , Smoking-related Lung Disease, Seminars in Ultrasound CT and MRI (2018),
doi: https://doi.org/10.1053/j.sult.2018.11.010

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service
to our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and
all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Smoking-related Lung Disease

Brett M. Elicker, MD*; Kimberly G. Kallianos*, MD; Kirk D. Jones, MD§; Travis S.

Henry, MD*.

T
IP
*Department of Radiology & Biomedical Imaging, University of California, San

CR
Francisco

§Department of Pathology, University of California, San Francisco

US
Correspondence to: Brett M. Elicker, MD, Department of Radiology
AN
and Biomedical Imaging, University of California San Francisco,

505 Parnassus Avenue, Box 0628, San Francisco, CA 94143 (e-mail:

M

brett.elicker@ucsf.edu).
ED

Abstract
PT

Smoke from cigarettes and other sources may induce a variety of patterns of
CE

lung injury. While smoking related lung diseases, in general, have a better prognosis

than many other diffuse lung diseases, they may be a cause of significant symptoms
AC

and, in some cases, may even require lung transplantation. On histology the

manifestations of these patterns range from reversible inflammation to irreversible

emphysema or fibrosis. High-resolution chest CT plays a critical role in the

diagnosis of smoking related lung diseases. It has several roles including 1) helping
 ACCEPTED MANUSCRIPT

determine diagnosis, 2) assessing the pattern of injury that is present, 3) evaluating

the extent and severity of disease, and 4) determining the response to treatment.

The practicing radiologist must have a knowledge of the clinical, pathologic and

imaging features of the differnent patterns of lung injury associated with smoke

inhalation.

T
IP
Introduction

CR
Cigarette smoke is estimated to be responsible for approximately 1 in 5

US
deaths in the United States(1). Much of this mortality is attributable to an increased

risk of heart disease, stroke, and lung cancer; however, smoking-related lung
AN
diseases (SRLDs) are also a contributor. While smoking is best known for its ability

to cause emphysema and chronic bronchitis, it is also associated with a variety of

M

other patterns of diffuse lung disease (DLD) that may also be a source of mortality.

In a study by Wasko and colleagues(2) interstitial lung abnormalities (ILAs) were

ED

visualized in 8% of smokers on HRCT, and these ILAs have been associated with
PT

reduced survival(3). SRLDs are also important to recognize given their ability to

mimic DLDs that are unrelated to smoking. Making an accurate diagnosis in these
CE

cases often has a significant impact on treatment and prognosis. The goal of this

paper is to describe the clinical, radiologic, and pathologic manifestations of the

AC

spectrum of SRLDs.
 ACCEPTED MANUSCRIPT

Clinical Approach

The key components of the clinical history in any patient with suspected DLD

include: 1) inhaled exposures, 2) drug exposures, and 3) manifestations of

T
connective tissue disease. Inhaled exposures that are most closely associated with

IP
DLD include organic antigens (e.g. birds), inorganic antigens (e.g. silica), and smoke.

CR
The type and volume of smoke inhaled, including changes over time, are important

in determining the likelihood of SRLD as a potential cause of pulmonary symptoms.

US
While injury due to cigarette smoke inhalation has been studied most

extensively, other materials, such as cannabis and electronic cigarettes, are potential
AN
perpetrators. Their effects on the lungs, however, have been poorly studied. In a

small pathologic series of cannabis smokers, intra-alveolar macrophages and

M

interstitial fibrosis resembling desquamative interstitial pneumonia was seen(4).

ED

Case reports have identified an association between cannabis and bullous

emphysema on imaging (5). Synthetic cannabis has been shown to manifest as

PT

diffuse centrilobular nodules and tree-in-bud opacities on CT, corresponding

pathologically to organizing pneumonia with or without diffuse alveolar damage(6).

CE

Patients who stop smoking cigarettes and switch to electronic cigarettes show
AC

improvement in the degree of airflow obstruction(7), however the pathologic and

radiographic manifestations of electronic cigarettes have not yet been described.

The effects of secondhand smoke inhalation are also poorly studied. In one study of

asbestos workers(8), secondhand smoke was associated with ground glass opacity

on HRCT, possibly indicative of early RB or DIP.

 ACCEPTED MANUSCRIPT

It is important to note that while many patients with SRLD have pulmonary

symptoms, an asymptomatic presentation is also not uncommon. Respiratory

bronchiolitis and smoking related interstitial fibrosis are both well documented in

asymptomatic patients(9). The mainstay of treatment of SRLDs is smoking

cessation. When inflammation is the predominant finding, smoking cessation may

T
be effective in reversing the injury. Rarely findings may progress even with

IP
adherence to smoking cessation. Corticosteroids may also be used in cases resistant

CR
to improvement with smoking cessation alone, however their efficacy is poorly

validated. In advanced cases of SRLD lung transplantation may be required.

US
AN
Pathologic Approach

From a pathologic perspective SRLDs result in three general lung

M

manifestations: 1) inflammation and cellular infiltration, 2) emphysema, and 3)

fibrosis. There are multiple different patterns of injury associated with smoke
ED

inhalation each of which may show one or more of these end pathologic results.
PT

While these patterns are often described as separate entities, in practice an overlap

is common(10). Many demonstrate inflammation as their earliest manifestation.

CE

Inflammation, in turn, results in the induction of repair mechanisms that may

eventually induce emphysema and/or fibrosis.

AC

The role of pathology in the multidisciplinary approach of DLD is to

determine diagnosis when a combination of clinical and radiographic findings is

nonspecific. Given that DLDs can show significant variability in their findings

throughout the lungs, surgical lung biopsy is preferred over transbronchial or

 ACCEPTED MANUSCRIPT

transthoracic biopsies. The specific patterns of injury that may be seen in

association with smoking are discussed below, however several features on

pathology are shared by many of these patterns. Lightly pigmented alveolar

macrophages are commonly present in both asymptomatic patients and those with

pulmonary symptoms. A bronchiolar accentuation of findings is also a typical

T
feature of many smoking related lung diseases, although this distribution is seen

IP
with other inhaled diseases such as hypersensitivity pneumonitis. Emphysema on

CR
pathology manifests as airspace enlargement and alveolar septal fragmentation.

Fibrosis due to smoking has a distinctive appearance that is different from that of
US
many other fibrotic lung diseases, such as usual interstitial pneumonia (UIP). In
AN
UIP, the fibrosis is irregular and heterogeneous with marked architectural

distortion. Fibrosis due to smoking, on the other hand, manifests as paucicellular

M

uniform thickening of the alveolar septal walls closely resembling nonspecific

interstitial pneumonia.
ED
PT

Radiologic Approach

High-resolution chest computed tomography (HRCT) is a critical part of the

CE

multi-disciplinary approach to diagnosis and should be obtained in all cases of

suspected DLD. When typical HRCT findings are seen in association with a smoking
AC

history, an accurate diagnosis of SRLD can often be made without pathology. There

is a wide spectrum of HRCT findings that may be seen with SRLD. A knowledge of

the typical HRCT manifestations of each pattern, in addition to the non-smoking

 ACCEPTED MANUSCRIPT

related lung diseases whose findings overlap with those patterns, is critical in

making accurate diagnoses.

The key findings of SRLD on HRCT include ground glass opacity, nodules

(either ground glass attenuation or solid), emphysema, and fibrosis. In many cases,

several of these findings co-exist. Ground glass opacity most commonly

T
corresponds to interstitial inflammation and partial alveolar filling by macrophages.

IP
Alternatively, glass opacity may correspond to interstitial fibrosis and, in fact, is a

CR
more common manifestation of fibrosis in SRLDs compared to other fibrotic lung

diseases. Irregular reticulation, traction bronchiectasis, and honeycombing may be

US
present, but are generally less pronounced in SRLDs. Nodules are typically
AN
centrilobular in distribution and reflect the bronchiolar distribution of the immune

reaction. Emphysema corresponds to areas of lung destruction.

M
ED

Patterns of Injury in Smoking-related Lung Disease

PT

The different patterns of SRLD reflect the heterogeneous relationship

between the type of smoke inhaled and the nature of the immune reaction that
CE

occurs as a response. The insult due to smoke inhalation results in three separate

pathways of lung response: 1) inflammation and cellular infiltration, 2) emphysema

AC

(lung destruction), and 3) fibrosis. Regardless of the specific pattern present, the

consequence will be one or more of these end results (Figure 1)

The specific patterns of injury associated with smoke inhalation (Table 1)

will be discussed below, with those most closely associated with inflammatory
 ACCEPTED MANUSCRIPT

changes presented first followed by those with predominantly irreversible findings.

The discussion of each pattern will focus on the following: 1) clinical manifestations,

2) radiology-pathology correlation, 3) differential diagnosis, and 4) natural history

and prognosis.

T
 Acute Eosinophilic Pneumonia

IP
Clinical

CR
Greater than 95% of cases of acute eosinophilic pneumonia (AEP) are due to

cigarette smoke exposure(11). The majority of patients have a history of a change

US
in their smoking habits (e.g. increased number of cigarettes smoked per day) in the
AN
month leading up to the development of symptoms(12). In contradistinction to

other patterns of SRLD, AEP presents with rapidly progressive acute symptoms,
M

resembling acute respiratory distress syndrome. Admission to the intensive care

unit is required in most. As opposed to other eosinophilic lung diseases, a

ED

peripheral eosinophilia is only seen in a minority of patients (approximately 30%).

PT

Radiology-Pathology Correlation

Pathology in AEP is essentially diffuse alveolar damage (DAD) with airspace

CE

consolidation by macrophages, fibrin, and eosinophils(13). The HRCT findings

resemble DAD from other causes with a combination of extensive bilateral ground
AC

glass opacity and smooth interlobular septal thickening (the crazy-paving” pattern)

or consolidation (Figure 2). Fibrosis and emphysema are not characteristic features

of AEP.

Differential Diagnosis
 ACCEPTED MANUSCRIPT

The clinical and radiographic features of AEP are relatively nonspecific and

overlap with other causes of acute pulmonary symptoms. Diagnosis is further

confounded by the relatively low rate of peripheral eosinophilia. Making a diagnosis

of AEP necessitates a high suspicion based upon the smoking history.

Bronchoalveolar lavage, in search for tissue eosinophils, may be helpful in patients

T
with a history of smoking and diffuse alveolar damage of unknown etiology. From a

IP
radiologic perspective the findings of AEP on HRCT are indistinguishable from

CR
pulmonary edema, certain infections (e.g. viral), diffuse alveolar damage, and diffuse

alveolar hemorrhage.

Natural History and Treatment US

AN
In distinction to other causes of diffuse alveolar damage, AEP is a pattern that

demonstrates rapid improvement after corticosteroid treatment, thus a diagnosis of

M

AEP has a significant impact on treatment. Resolution of chest radiographic

abnormalities is seen in about 85% of patients within 7 days(11).

ED
PT

 Langerhan’s Cell Histiocytosis

Clinical
CE

Langerhan's Cell Histiocytosis (LCH) is a rare manifestation of SRLD with

distinctive pathologic and imaging findings. In children, LCH is a systemic disorder

AC

unrelated to smoking, however in adults it is predominantly a SRLD. In a study of

102 adults with Langerhans cell histiocytosis 68% were current smokers and 27%

were former smokers. Extrapulmonary involvement in this study was uncommon,

present in only 17%. Patients with LCH may present with symptoms of dyspnea or
 ACCEPTED MANUSCRIPT

pneumothorax. In a study of 122 adults with LCH, 16% developed a

pneumothorax(14). However, in about 15% of patients, LCH is discovered as an

incidental finding(15).

Radiology-pathology Correlation

Pathologically, LCH is characterized by bronchiolocentric stellate nodules

T
composed of Langerhans cell histiocytes and eosinophils. On HRCT (Figure 3) this

IP
corresponds to centrilobular nodules, often of soft tissue attenuation. Over time,

CR
these collections of histiocytes cause destruction of the centrilobular bronchiolar

wall and focal dilation of the airway(16). On HRCT this will appear as air-
US
attenuation lucencies at the center of the solid nodules, resembling cavitation.
AN
Progressive bronchiolar dilation and lung destruction eventually results in

emphysema surround by fibrosis, manifesting as coalescing irregularly shaped cysts

M

on HRCT(17). These findings predominate in the mid to upper lungs, sparing the

costophrenic angles.
ED

Differential diagnosis
PT

LCH, RB and DIP often co-exist, thus overlapping clinical, pathologic, and

imaging findings may be present. The differential diagnosis of LCH is somewhat

CE

different than other SRLDs because of its unique imaging findings. The

predominantly solid nodules of LCH may be confused with metastatic malignancy,

AC

certain infections (e.g. fungal or mycobacterial organisms) or tracheobronchial

papillomatosis. When cysts are the predominant finding,

lymphangioleiomyomatosis (LAM) is often considered. LAM is distinguished from

LCH by a lack of nodules, and cysts that are round and involve the lung bases.
 ACCEPTED MANUSCRIPT

Natural history and Treatment

The treatment for LCH is smoking cessation and, if necessary, corticosteroid

treatment. The prognosis is generally quite good compared to other DLDs. In a

study of 102 adults with LCH, the median survival was 12.5 years(15). Paradoxical

recurrence after smoking cessation and regression while continuing smoking have

T
been reported(18). In a study of serial HRCTs in 21 patients with LCH, there was a

IP
reduction in the percentage of CTs with micronodules (initial: 86%, follow-up: 43%)

CR
and thick-walled cysts (initial: 48%, follow-up: 24%) over time (19). Findings that

worsened on interval imaging included: thinned walled cysts (initial: 62%, follow-
US
up: 67%) and emphysema (initial: 24%, follow-up: 43%).
AN

 Respiratory Bronchiolitis and Desquamative Interstitial Pneumonia

M

Clinical

Respiratory bronchiolitis (RB) and desquamative interstitial pneumonia

ED

(DIP) are thought to represent a spectrum of disease with significant clinical,

PT

radiologic, and pathologic overlap(20). They are both closely associated with a

history of smoking, although in one study(21) 40% of patients with DIP were non-
CE

smokers. Both may also be seen as a minor secondary pattern in association with

another DLD, such as idiopathic pulmonary fibrosis in which case they may simply
AC

be a marker of smoke exposure. Also, RB and DIP may be detected as an incidental

radiographic or pathologic finding in asymptomatic patients. For this reason, a

diagnosis of RB or DIP as a cause of pulmonary symptoms should only be made

when other causes of DLD are excluded.

 ACCEPTED MANUSCRIPT

Radiology-pathology Correlation

RB and DIP are characterized by alveolar filling by macrophages in either a

peribronchiolar or diffuse distribution respectively. These may be associated with

inflammation or fibrosis of the alveolar septal walls. While the typical HRCT

findings of RB and DIP will be discussed separately, it is important to remember that

T
they exist on a spectrum and thus overlap of radiologic findings is common (20).

IP
In a study of 21 smokers(22) with respiratory bronchiolitis, the most

CR
common HRCT features included central airway thickening in 90%, centrilobular

nodules in 71%, and ground glass opacity in 67% of patients. As both airway
US
thickening and ground glass opacity are nonspecific, the most suggestive feature of
AN
RB is centrilobular nodules of ground glass attenuation (Figure 4). This finding

corresponds pathologically to the patchy alveolar filling by smokers macrophages,

M

interstitial inflammation, and cellular infiltration adjacent to the centrilobular

bronchiole.
ED

DIP classically manifests on HRCT as ground glass opacities that have a

PT

basilar predominance(23,24). These ground glass opacities correspond

pathologically to uniform thickening of the interstitium by inflammation or fibrosis,

CE

and patchy filling of the alveolar spaces by pigmented macrophages. While ground

glass opacity is the most common finding of DIP, seen in nearly 100% of patients,
AC

fibrosis and cystic lucencies may also be present (Figure 5). The cystic lucencies

have a different morphology than honeycombing and may be representative of

emphysema or dilated alveolar ducts and bronchioles(24). On pathology, fibrosis in

DIP closely resembles that of NSIP with uniform thickening of the alveolar septal
 ACCEPTED MANUSCRIPT

walls by fibrosis and relatively preserved lung architecture. This explains the

relatively high occurrence of ground glass opacity as a predominant finding in both

fibrotic DIP and NSIP. Irregular reticulation, traction bronchiectasis and

honeycombing may be seen with DIP, however they are typically less prominent

than with other fibrotic lung diseases.

T
Differential diagnosis

IP
The diagnosis of RB or DIP is often made by the presence of typical HRCT

CR
findings and an appropriate smoking history. The differential diagnosis of these

patterns depends upon the predominant features present at the time of detection.
US
In cases of RB presenting with centrilobular ground glass nodules, hypersensitivity
AN
pneumonitis (HP) is often the main diagnostic consideration(25). A history of

exposure to bird or mold antigens in conjunction with a lymphocytosis on

M

bronchoalveolar lavage are suggestive of HP. Smoking is relatively protective

against the development of HP.

ED

As discussed above, DIP and NSIP show significant overlap in their pathologic
PT

and HRCT findings. On HRCT both patterns present with basilar predominant

ground glass opacity. The presence of subpleural sparing on imaging strongly

CE

suggests NSIP. The cystic lucencies of fibrotic DIP may resemble honeycombing in

which case usual interstitial pneumonia (UIP) is often also considered. The
AC

morphology of these cysts differs from honeycombing in that they have thinner

walls, irregular shapes (i.e. not round) and often spare the subpleural lung. The

presence of ground glass opacity and the relative paucity of honeycombing also

helps distinguish DIP from UIP(26).

 ACCEPTED MANUSCRIPT

Natural history and Treatment

While RB has been traditionally thought of as having a 5 year survival

approaching 100%, a more recent investigation by Portnoy and colleagues(27)

estimated that survival in symptomatic RB was at approximately 75% after 7 years.

Interestingly, only 28% of patients in this study showed clinical improvement over

T
time suggesting that in most patients the disease was predominantly irreversible. In

IP
the study by Ryu and colleagues(28) the 3 year survival of DIP was 74%.

CR
Symptomatic improvement and a decrease in HRCT abnormalities were also only

seen in a minority of patients (24% and 33% respectively). While patients with RB
US
and DIP who adhere to smoking cessation may show a decrease in the nodules and
AN
ground glass opacity on HRCT (29), residual ground glass opacity often persists,

likely reflecting microscopic areas of fibrosis. In a study of the serial HRCT follow-
M

up of DIP patients, Hartman and colleagues(30) discovered that >90% of patients

with DIP had residual ground glass opacity on interval follow-up. Additionally, 20%
ED

of DIP patients developed new reticulation and/or honeycombing over time. In

PT

patients who continue to smoke, the severity of ground glass opacity and

emphysema on HRCT may increase over time(31), and patients may develop cystic
CE

lucencies or honeycombing (32). The cystic lucencies, in these cases, may represent

areas of emphysema surrounded by fibrosis.

AC
 ACCEPTED MANUSCRIPT

 Smoking-related Interstitial Fibrosis and Airspace Enlargement with

Fibrosis

Clinical

Smoking may induce a reaction that results in lung fibrosis without a clear

T
preceding inflammatory stage. Smoking-related interstitial fibrosis (SRIF)(9),

IP
airspace enlargement with fibrosis(33), and respiratory bronchiolitis with fibrosis

CR
(34) are pathologic terms given to this entity. These patterns have primarily been

described as incidental findings on histology obtained in patients in whom there

US
was not a suspicion for DLD. In the study by Kawabata et al(33), 18% of moderate
AN
smokers had this histologic finding on lobectomy specimens.

Radiology-pathology Correlation
M

Typical histology includes centrilobular and subpleural alveolar septal

thickening by fibrosis associated with respiratory bronchiolitis and emphysema. It

ED

is likely that these patterns are simply the sequela of prior RB or DIP in which case
PT

the inflammatory precursor was missed. The HRCTs (Figure 6) in these patients

may be normal or the findings may overlap with RB and DIP. In the series by
CE

Yousem et al., 8 of 9 patients with SRIF demonstrated bilateral micronodules on

CT(34). Chae et al. (35) compared the HRCT findings of SRIF and usual interstitial
AC

pneumonia (UIP). They found that findings associated with SRIF included

“honeycombing” that spared the subpleural lung, lack of honeycombing in the upper

lobes, and emphysema adjacent to areas of honeycombing. These descriptions, in

some respects, resemble cases of fibrotic DIP. Reddy et al(36) found that the HRCT
 ACCEPTED MANUSCRIPT

findings of SRIF included mid to upper lung emphysema associated with mild

reticulation and/or ground glass opacity.

Differential Diagnosis

The differential diagnosis of SRIF is more of a pathological problem than a

clinical or radiologic one. A pathologist must distinguish SRIF from more ominous

T
patterns such as usual interstitial pneumonia. From a clinical perspective, most

IP
patients with SRIF are asymptomatic, although studies evaluating the long-term

CR
progression of SRIF into a clinically significant disease have not been performed. On

imaging, the HRCT findings of SRIF overlap with RB and DIP. Since these patterns
US
may all represent a spectrum of the same process, their distinction may not be
AN
important.

Natural history and Treatment

M

While the long-term prognosis of patients with SRIF is unknown, the fact that

many patients are asymptomatic suggests that this may be a relatively indolent
ED

pattern. In the study by Yousem et al.(34) 78% of patients were stable over time
PT

and 12% were slowly progressive over a median 3.2 year period. Follow-up of the

patients in the study by Katzenstein et al.(9) also demonstrated an indolent course

CE

in most patients.
AC

 Idiopathic Pulmonary Fibrosis

Clinical

Approximately 60-70% of patients with idiopathic pulmonary fibrosis (IPF)

are current or former smokers(37,38). The exact mechanism of this interaction is

 ACCEPTED MANUSCRIPT

not clear. Smoke may, in part, be directly responsible for the injury associated with

lung fibrosis or it may sensitize the lung to injury. Regardless, IPF differs from other

smoking related lung diseases in that it is a process characterized entirely by

fibrosis without inflammation.

Given the high percentage of smokers amongst IPF patients, lung fibrosis and

T
emphysema may coexist. This combination of findings is termed combined

IP
pulmonary fibrosis and emphysema (CPFE). CPFE is not a diagnosis per se, but is

CR
simply a combination of emphysema and a fibrotic lung disease. That fibrotic lung

disease is often IPF, but not always. In a study of 10 patients with CPFE(39), 80%
US
had usual interstitial pneumonia whereas 20% had DIP. Given the different
AN
prognosis and treatment of DIP compared to IPF, a confident diagnosis of the

underlying fibrotic lung disease in patients with CPFE is important.

M

Radiology-pathology Correlation

The imaging findings of IPF are usually easily distinguishable from other
ED

patterns of injury in SRLD with the exception of DIP. Both IPF and DIP may show a
PT

basilar and subpleural distribution of fibrosis. The emphysema or cystic lucencies

of DIP may resemble the honeycombing of IPF. DIP has a greater preponderance of
CE

ground glass opacity, and less traction bronchiectasis and irregular reticulation.

The cystic lucencies of DIP differ from honeycombing in that they have irregular
AC

shapes (i.e. are not round), have thin walls, and do not extend to the subpleural lung.

From a clinical perspective, DIP tends to progress much more slowly than IPF.

Natural history and Treatment

 ACCEPTED MANUSCRIPT

In distinction to most other SRLDs, the mortality of IPF is high with a mean

survival of approximately 3-4 years after diagnosis(40). Serial HRCTs in IPF

patients demonstrate worsening of the extent of fibrosis in about 90% of patients

over a 4 year follow-up period(41). Acute exacerbation, manifesting as rapidly

worsening clinical symptoms and new ground glass opacity on HRCT, is also more

T
common with IPF than other SRLDs. Pirfenidone and nintedanib are unique

IP
treatments for IPF that may slow the progression of fibrosis(42,43).

CR
 Emphysema

Clinical US
AN
A comprehensive discussion of emphysema is beyond the scope of this paper;

however, it is important to mention given that it is the end result of many of the
M

other patterns of SRLD. Emphysema reflects an irreversible consequence of smoke

inhalation characterized by lung destruction from chronic inflammation and cellular

ED

infiltration. Clinically it may result in profound airway obstruction and dyspnea, or

PT

it may be detected at an early, subclinical stage. It is often seen in association with

the other patterns described above.

CE

Radiology-pathology Correlation

On pathology, emphysema manifests as airspace enlargement and

AC

destruction of the alveolar septal walls, resulting in areas of lung with a simplified

architecture. A centrilobular distribution of emphysema is most common, reflecting

the airway-centered nature of the injury. On HRCT this manifests as focal, air-

attenuation lucencies without a wall predominating in the central portions of the

 ACCEPTED MANUSCRIPT

upper lungs. When emphysema is surrounded by fibrosis, as seen in DIP, it may

appear to have a wall at its periphery.

It is important to note that uniform distribution of emphysema throughout

the lungs may result in clinically significant abnormalities that are not detectable on

HRCT. Quantitative lung imaging is a technique that may improve the HRCT

T
sensitivity for emphysema and may also provide an assessment of the extent and

IP
severity of disease. Additionally, quantitative lung imaging may be useful in the

CR
serial evaluation of emphysema over time. However, the clinical utility of these

techniques is still to be determined.

Differential Diagnosis US
AN
Centrilobular emphysema has a characteristic appearance on HRCT.

Distinction should be made between emphysema and other air attenuation

M

abnormalities. Cysts are air-attenuation lucencies that have a thin, but discreet wall

at their periphery, although emphysema surrounded by fibrosis may appear to have

ED

a wall. Bronchiectasis has a tubular configuration corresponding to dilated airways.

PT

Honeycombing cysts are clustered, have thick walls, and involve the subpleural lung.

Natural history and Treatment

CE

As emphysema is an irreversible finding, treatment focuses on preventing

further progression of disease and ameliorating symptoms. As with other patterns

AC

of SRLD, smoking cessation forms the cornerstone of treatment. Beta-agonists are

used to temporarily relieve airways obstruction and corticosteroids to reduce

airways inflammation. Even in patients who stop smoking, progression of

 ACCEPTED MANUSCRIPT

emphysema may occur. Quantitative lung imaging may aid in characterizing this

progression and may be more sensitive than other measures of lung function(44).

Conclusion

The effects of smoking on the lungs are varied and result in a variety of

T
IP
patterns of injury characterized by either potentially reversible inflammatory

changes or irreversible changes of emphysema or fibrosis. Most SRLDs have a good

CR
prognosis, and an asymptomatic presentation is not uncommon. HRCT is important

US
in the detection and characterization of SRLDs, and in their distinction from other

DLDs. This distinction is critical in the treatment of patients, particularly in terms of

AN
the need for exposure removal and immunosuppressive therapy.
M

References
ED

1. The Health Conequences of Smoking-50 Years Of Progress. A Report of the

Surgeon General. Retrieved 11/1/2018 from:

PT

httpswww.cdc.govtobaccodatastatisticssgrth-anniversaryindex.htm
CE

2. Washko GR, Hunninghake GM, Fernandez IE, et al. Lung volumes and
AC

emphysema in smokers with interstitial lung abnormalities. N Engl J Med.

2011;364:897–906.

3. Putman RK, Hatabu H, Araki T, et al. Association Between Interstitial Lung

Abnormalities and All-Cause Mortality. JAMA. 2016;315:672–681.

 ACCEPTED MANUSCRIPT

4. Gill A. Bong lung: regular smokers of cannabis show relatively distinctive

histologic changes that predispose to pneumothorax. Am J Surg Pathol.

2005;29:980–982.

5. Johnson MK, Smith RP, Morrison D, Laszlo G, White RJ. Large lung bullae in

T
marijuana smokers. Thorax. 2000;55:340–342.

IP
6. Berkowitz EA, Henry TS, Veeraraghavan S, Staton GW, Gal AA. Pulmonary

CR
effects of synthetic marijuana: chest radiography and CT findings. AJR

American journal of roentgenology. American Roentgen Ray Society.

2015;204:750–757. US
AN
7. Cibella F, Campagna D, Caponnetto P, et al. Lung function and respiratory

symptoms in a randomized smoking cessation trial of electronic cigarettes.

M

Clin Sci. Portland Press Limited. 2016;130:1929–1937.

ED

8. Vierikko T, Järvenpää R, Uitti J, et al. The effects of secondhand smoke

exposure on HRCT findings among asbestos-exposed workers. Respir Med.

PT

2008;102:658–664.
CE

9. Katzenstein A-LA, Mukhopadhyay S, Zanardi C, Dexter E. Clinically occult

interstitial fibrosis in smokers: classification and significance of a surprisingly

AC

common finding in lobectomy specimens. Hum Pathol. 2010;41:316–325.

10. Vassallo R, Jensen EA, Colby TV, et al. The overlap between respiratory

bronchiolitis and desquamative interstitial pneumonia in pulmonary

 ACCEPTED MANUSCRIPT

Langerhans cell histiocytosis: high-resolution CT, histologic, and functional

correlations. Chest. 2003;124:1199–1205.

11. Rhee CK, Min KH, Yim NY, et al. Clinical characteristics and corticosteroid

treatment of acute eosinophilic pneumonia. Eur Respir J. 2013;41:402–409.

T
12. Uchiyama H, Suda T, Nakamura Y, et al. Alterations in smoking habits are

IP
associated with acute eosinophilic pneumonia. Chest. 2008;133:1174–1180.

CR
13. Tazelaar HD, Linz LJ, Colby TV, Myers JL, Limper AH. Acute eosinophilic

US
pneumonia: histopathologic findings in nine patients. Am J Respir Crit Care

Med. 1997;155:296–302.
AN

14. Mendez JL, Nadrous HF, Vassallo R, Decker PA, Ryu JH. Pneumothorax in
M

pulmonary Langerhans cell histiocytosis. Chest. 2004;125:1028–1032.

15. Vassallo R, Ryu JH, Schroeder DR, Decker PA, Limper AH. Clinical outcomes of
ED

pulmonary Langerhans'-cell histiocytosis in adults. N Engl J Med.

PT

2002;346:484–490.
CE

16. Kambouchner M, Basset F, Marchal J, Uhl JF, Hance AJ, Soler P. Three-

dimensional characterization of pathologic lesions in pulmonary langerhans

AC

cell histiocytosis. Am J Respir Crit Care Med. 2002;166:1483–1490.

17. Kim HJ, Lee KS, Johkoh T, et al. Pulmonary Langerhans cell histiocytosis in

adults: high-resolution CT-pathology comparisons and evolutional changes at

CT. Eur Radiol. Springer-Verlag. 2011;21:1406–1415.

 ACCEPTED MANUSCRIPT

18. Westerlaan HE, van der Valk PDLPM. Clinical and radiological evolution in

patients with pulmonary Langerhans' cell histiocytosis. Neth J Med.

2002;60:320–326.

19. Brauner MW, Grenier P, Tijani K, Battesti JP, Valeyre D. Pulmonary

T
Langerhans cell histiocytosis: evolution of lesions on CT scans. Radiology.

IP
1997;204:497–502.

CR
20. Heyneman LE, Ward S, Lynch DA, Remy-Jardin M, Johkoh T, Müller NL.

Respiratory bronchiolitis, respiratory bronchiolitis-associated interstitial lung

US
disease, and desquamative interstitial pneumonia: different entities or part of
AN
the spectrum of the same disease process? AJR American journal of

roentgenology. American Public Health Association. 1999;173:1617–1622.

M

21. Craig PJ, Wells AU, Doffman S, et al. Desquamative interstitial pneumonia,
ED

respiratory bronchiolitis and their relationship to smoking. Histopathology.

Wiley/Blackwell (10.1111). 2004;45:275–282.

PT

22. Park JS, Brown KK, Tuder RM, Hale VAE, King TE, Lynch DA. Respiratory
CE

bronchiolitis-associated interstitial lung disease: radiologic features with

clinical and pathologic correlation. J Comput Assist Tomogr. 2002;26:13–20.

AC

23. Johkoh T, Müller NL, Cartier Y, et al. Idiopathic interstitial pneumonias:

diagnostic accuracy of thin-section CT in 129 patients. Radiology.

1999;211:555–560.
 ACCEPTED MANUSCRIPT

24. Akira M, Yamamoto S, Hara H, Sakatani M, Ueda E. Serial computed

tomographic evaluation in desquamative interstitial pneumonia. Thorax.

1997;52:333–337.

25. Okada F, Ando Y, Yoshitake S, et al. Clinical/pathologic correlations in 553

T
patients with primary centrilobular findings on high-resolution CT scan of the

IP
thorax. Chest. 2007;132:1939–1948.

CR
26. Hartman TE, Primack SL, Swensen SJ, Hansell D, McGuinness G, Müller NL.

Desquamative interstitial pneumonia: thin-section CT findings in 22 patients.

Radiology. 1993;187:787–790. US
AN
27. Portnoy J, Veraldi KL, Schwarz MI, et al. Respiratory bronchiolitis-interstitial

lung disease: long-term outcome. Chest. 2007;131:664–671.

M

28. Ryu JH, Myers JL, Capizzi SA, Douglas WW, Vassallo R, Decker PA.
ED

Desquamative interstitial pneumonia and respiratory bronchiolitis-associated

interstitial lung disease. Chest. 2005;127:178–184.

PT

29. Nakanishi M, Demura Y, Mizuno S, et al. Changes in HRCT findings in patients

CE

with respiratory bronchiolitis-associated interstitial lung disease after

smoking cessation. Eur Respir J. 2007;29:453–461.

AC

30. Hartman TE, Primack SL, Kang EY, et al. Disease progression in usual

interstitial pneumonia compared with desquamative interstitial pneumonia.

Assessment with serial CT. Chest. 1996;110:378–382.

 ACCEPTED MANUSCRIPT

31. Remy-Jardin M, Edme J-L, Boulenguez C, Remy J, Mastora I, Sobaszek A.

Longitudinal follow-up study of smoker's lung with thin-section CT in

correlation with pulmonary function tests. Radiology. 2002;222:261–270.

32. Kawabata Y, Takemura T, Hebisawa A, et al. Desquamative interstitial

T
pneumonia may progress to lung fibrosis as characterized radiologically.

IP
Respirology. Blackwell Publishing Asia. 2012;17:1214–1221.

CR
33. Kawabata Y, Hoshi E, Murai K, et al. Smoking-related changes in the

background lung of specimens resected for lung cancer: a semiquantitative

US
study with correlation to postoperative course. Histopathology. 2008;53:707–
AN
714.

34. Yousem SA. Respiratory bronchiolitis-associated interstitial lung disease with

M

fibrosis is a lesion distinct from fibrotic nonspecific interstitial pneumonia: a

ED

proposal. Mod Pathol. Nature Publishing Group. 2006;19:1474–1479.

35. Chae KJ, Jin GY, Jung HN, et al. Differentiating Smoking-Related Interstitial
PT

Fibrosis (SRIF) from Usual Interstitial Pneumonia (UIP) with Emphysema

CE

Using CT Features Based on Pathologically Proven Cases. PLoS ONE.

2016;11:e0162231.
AC

36. Reddy TL, Mayo J, Churg A. Respiratory bronchiolitis with fibrosis. High-

resolution computed tomography findings and correlation with pathology.

Ann Am Thorac Soc. American Thoracic Society. 2013;10:590–601.

 ACCEPTED MANUSCRIPT

37. Bjoraker JA, Ryu JH, Edwin MK, et al. Prognostic significance of

histopathologic subsets in idiopathic pulmonary fibrosis. Am J Respir Crit

Care Med. 1998;157:199–203.

38. Baumgartner KB, Samet JM, Stidley CA, Colby TV, Waldron JA. Cigarette

T
smoking: a risk factor for idiopathic pulmonary fibrosis. Am J Respir Crit Care

IP
Med. 1997;155:242–248.

CR
39. Jankowich MD, Polsky M, Klein M, Rounds S. Heterogeneity in combined

pulmonary fibrosis and emphysema. Respiration. Karger Publishers.

2008;75:411–417. US
AN
40. Flaherty KR, Toews GB, Travis WD, et al. Clinical significance of histological

classification of idiopathic interstitial pneumonia. Eur Respir J. 2002;19:275–

M

283.
ED

41. Nishiyama O, Taniguchi H, Kondoh Y, et al. Familial idiopathic pulmonary

fibrosis: serial high-resolution computed tomography findings in 9 patients. J

PT

Comput Assist Tomogr. 2004;28:443–448.

CE

42. King TE, Bradford WZ, Castro-Bernardini S, et al. A phase 3 trial of pirfenidone

in patients with idiopathic pulmonary fibrosis. N Engl J Med. 2014;370:2083–

AC

2092.

43. Richeldi L, Bois du RM, Raghu G, et al. Efficacy and safety of nintedanib in

idiopathic pulmonary fibrosis. N Engl J Med. 2014;370:2071–2082.

 ACCEPTED MANUSCRIPT

44. Stolk J, Putter H, Bakker EM, et al. Progression parameters for emphysema: a

clinical investigation. Respir Med. 2007;101:1924–1930.

Table 1. Patterns of injury due to smoke inhalation.

T
IP
Pattern Clinical Pathologic HRCT

CR
Acute eosinophilic -Acute onset -Diffuse alveolar -Ground glass

pneumonia damage opacity and

-Resembles acute
US
respiratory distress -Airspace
interlobular septal

thickening (e.g.
AN
syndrome consolidation by
“crazy paving”)
macrophages,
-Peripheral
M

and/or
fibrin, and
eosinophilia in consolidation
eosinophils
ED

only 30%
-Diffuse or

-Rapid response to
PT

symmetric

corticosteroids distribution
CE

Langerhans cell -Children: -Collections of -Soft tissue

AC

histiocytosis systemic disorder histiocytes around attenuation

unrelated to bronchioles centrilobular

smoking nodules
-Progressive

-Adults: smoking airway dilation -“Cavitation”

 ACCEPTED MANUSCRIPT

related lung within nodules within nodules

disease, usually
-Irregular -Irregular cysts
isolated to lungs
emphysema
-Sparing of lung
-May present with surrounded by scar
bases

T
dyspnea or

IP
pneumothorax;

CR
sometimes

asymptomatic

Respiratory -Spectrum with US - Intra-alveolar -Centrilobular

AN
bronchiolitis (RB) DIP macrophages ground glass

nodules
-Present in all - Alveolar septal
M

smokers on wall thickening by -Upper lobe

pathology inflammation or distribution

ED

fibrosis
-May be a cause of
PT

dyspnea but is -Peribronchiolar

CE

more commonly distribution

asymptomatic
AC

Desquamative -Spectrum with -Intra-alveolar -Basilar ground

interstitial RB but a more macrophages glass opacity

pneumonia (DIP) generalized

-Alveolar septal -Cystic lucencies
 ACCEPTED MANUSCRIPT

reaction wall thickening by -Fibrosis, although

inflammation or ground glass

-Symptoms often
fibrosis opacity is often a
more pronounced
manifestation of
than with RB -More diffuse
fibrosis in DIP

T
distribution than

IP
with RB

CR
Smoking-related -Often an -Centrilobular and -HRCT often

interstitial fibrosis incidental finding subpleural alveolar normal

US
seen on pathologic septal wall
-When findings
AN
specimens in thickening by
are present they
patients without fibrosis
overlap with RB
M

symptoms
-Associated and DIP

respiratory
ED

bronchiolitis and
PT

emphysema
CE

Idiopathic -60-70% of IPF -Microscopic -Fibrosis with

pulmonary fibrosis patient are honeycombing honeycombing

AC

(IPF) smokers
-Fibroblastic foci -Subpleural and

-Progressive basilar distribution

-Subpleural
disease with poor
fibrosis
 ACCEPTED MANUSCRIPT

prognosis

-Acute

exacerbation may

occur

T
Emphysema -Irreversible -Airspace -Focal air-

IP
consequence of enlargement attenuation

CR
chronic smoke lucencies without
-Alveolar septal
inhalation a wall

-Represents lung
US
fragmentation

-Simplification of
-May appear to
AN
destruction have a wall when
lung architecture
emphysema is
-Wide range of
M

surrounded by
severity
fibrosis
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Figures.

Figure 1a-c. Evolution of smoking related lung disease over time. Baseline CT (Figure

1a) demonstrates peripheral ground glass opacity with scattered cystic lucencies typical

T
of desquamative interstitial pneumonia. Four years later (Figure b) the cystic lucencies

IP
are more extensive and have more discreet walls. While the ground glass persists there is

CR
development of reticulation and mild traction bronchiectasis. Another five years later

(Figure c) the ground glass has disappeared, and the cystic lucencies and reticulation are

US
the main abnormality. These findings correspond to emphysema and fibrosis.
AN

Figure 2a-c. Acute eosinophilic pneumonia. Surgical biopsy (Figure 2a) shows alveolar
M

filling by macrophages and eosinophils, and mild alveolar septal thickening by edema

and early fibrosis. On HRCT (Figure 2b) ground glass opacity and interlobular septal
ED

thickening is present in a patient who recently increased the number of cigarettes smoked

per day. Three months after treatment with corticosteroids (Figure 2c) there is near
PT

resolution of the abnormalities.

CE

Figure 3a-d. Langerhans cell histiocytosis (LCH). Surgical biopsy (Figure 3a) shows a
AC

stellate nodule composed of Langerhans cell histiocytes. On high magnification, the

Langerhans cells show characteristic reniform nuclei with nuclear contour irregularities.

Scattered eosinophils are present. On HRCT in one patient (Figure 3b) soft tissue,

attenuation centrilobular nodules are present, many of which have air attenuation
 ACCEPTED MANUSCRIPT

lucencies at their centers. In another patient presenting with a pneumothorax (Figure 3c)

there are irregular cysts with thick and thin walls, in addition to a few nodules. In a third

patient (Figure 3d) extensive cysts that nearly replace the normal lung parenchyma are

present. These may be difficult to distinguish from extensive emphysema.

T
Figure 4a-b. Respiratory bronchiolitis (RB). Surgical biopsy (Figure 4a) shows

IP
peribronchiolar alveolar filling by lightly-pigmented pulmonary macrophages. Airspace

CR
enlargement of centriacinar emphysema is present. On HRCT (Figure 4b) these

pathologic findings correlate with centrilobular nodules of ground glass attenuation. Also

note the presence of emphysema. US

AN

Figure 5a-c. Desquamative interstitial pneumonia (DIP). Surgical biopsy (Figure 5a)
M

shows alveolar filling by macrophages and thickening of the alveolar septa by fibrosis.

On HRCT (Figure 5b) there is a basilar distribution of ground glass opacity and
ED

associated with cystic lucencies. In another patient with more advanced DIP (Figure 5c)

there are extensive subpleural air attenuation cystic lucencies without ground glass
PT

opacity. This appearance could be confused with honeycombing, however the thin walls,
CE

irregular shapes and sparing of the immediate subpleural interstitium in some regions are

typical of DIP.
AC

Figure 6a-c. Smoking related interstitial fibrosis (SRIF). Surgical biopsy (Figure 6a)

shows uniform subpleural alveolar septal thickening by dense hyalinized collagen. On

HRCT (Figure 6b) mild upper lung ground glass opacity, centrilobular ground glass
 ACCEPTED MANUSCRIPT

nodules and emphysema in SRIF resembles respiratory bronchiolitis. In another patient

with SRIF (Figure 6c) peripheral ground glass opacity, reticulation and cystic lucencies

resemble desquamative interstitial pneumonia.

T
Fig. 1A

IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 1B

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 1C

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 2A

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 2B

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 2C

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 3A

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 3B

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 3C

T
IP
CR
US
AN

Fig. 3D
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 4A

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 4B

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 5A

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 5B

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 5C

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 6A

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 6B

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Fig. 6C

T
IP
CR
US
AN
M
ED
PT
CE
AC