Note: This copy is for your personal non-commercial use only.

To order presentation-ready
copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.
䡲 THORACIC IMAGING

Lung Morphology in the Elderly:

Comparative CT Study of Subjects
over 75 Years Old versus Those under
55 Years Old1
ORIGINAL RESEARCH

Susan J. Copley, MD, MRCP, FRCR

Athol U. Wells, MD, FRACP, FRCR
Katherine E. Hawtin, MA, MBBS, MRCP, FRCR
Daren J. Gibson, BSc, MBBS, MRCS, FRCR
James M. Hodson, MA, MBBS, MRCP, FRCR
Audrey E. T. Jacques, BSc, MBBS, MRCP, FRCR
David M. Hansell, MD, FRCP, FRCR
Purpose: To describe thin-section pulmonary computed tomographic
(CT) features in asymptomatic elderly individuals.
Materials and Institutional review board approval was given, and in-
Methods: formed consent was obtained. Two study groups (older
group, over 75 years of age; younger group, under 55
years) were prospectively identified from outpatient re-
quests for CT of the abdomen or brain. Fifty-six consecu-
tive volunteers (older group: n ⫽ 40, 18 men, 22 women;
younger group: n ⫽ 16, eight men, eight women) with no
known respiratory disease were included. Prone thin-sec-
tion CT imaging was performed, and two observers inde-
pendently scored images for the presence and extent of CT
features (including reticular pattern, ground glass opacity,
and thin-walled cystic air spaces). Group comparisons
were made, and logistic regression analysis was used to
assess relationships between CT findings and age and
smoking history.

Results: A limited predominantly subpleural basal reticular pattern

was identified in the majority (24 of 40, 60%) of individuals
in the older group and was absent (zero of 16) in the
younger group (P ⬍ .001). Cysts were seen in 10 (25%) of
the 40 subjects in the older group but were seen in none of
the subjects in the younger group (P ⫽ .02). Bronchial
dilation and wall thickening were also seen significantly
more frequently (P ⬍ .001) in the older group (24 [60%]
and 22 [55%] of 40, respectively) than in the younger
group (both one [6%] of 16). All findings were indepen-
dent of pack-year smoking history with multiple logistic
regression analysis.

Conclusion: Thin-section CT findings usually associated with interstitial

lung disease are frequently seen in asymptomatic elderly
individuals and are absent in younger subjects. Therefore,
these findings may not necessarily represent clinically rel-
evant disease.
1
From the Department of Radiology, Hammersmith Hos-
pital, Imperial NHS Trust, Du Cane Rd, London W12 0HS, 娀 RSNA, 2009
England (S.J.C., K.E.H., D.J.G., J.M.H.); Interstitial Lung
Disease Unit (A.U.W.) and Department of Radiology
(D.M.H.), Royal Brompton Hospital, London, England; and
Department of Radiology, Guy’s and St. Thomas’ NHS
Trust, London, England (A.E.T.J.). Received July 17, 2008;
revision requested September 3; revision received Octo-
ber 23; accepted November 16; final version accepted
November 17. Supported by a Royal College of Radiolo-
gists (UK) Kodak Bursary. Address correspondence to
S.J.C. (e-mail: sue.copley@imperial.nhs.uk ).

姝 RSNA, 2009

566 radiology.rsnajnls.org ▪ Radiology: Volume 251: Number 2—May 2009

THORACIC IMAGING: CT Lung Morphology in the Elderly
P
hysiologic and morphologic changes
in the lung are a well-recognized
consequence of aging (1). The
documented physiologic changes in pul-
monary function that occur with ad-
vancing years are a result of loss of elas-
tic recoil in both alveoli and airways
(1,2). In animal models, histopathologic
studies have shown increased collagen
accumulation and progressive pulmo-
nary fibrosis associated with aging (3).
In humans, there is an increase in the
size of airspaces, dilation of alveoli, a
decrease in gas-exchange surface area,
and decreased supporting tissue for
small airways (1). Radiographic studies
(4,5) have found increased peripheral
lung markings with increasing age.
Thin-section computed tomography
(CT) provides greater anatomic detail
and allows increased sensitivity and
specificity in comparison to radiography
in a range of pulmonary conditions, in-
cluding interstitial lung disease (6–8).
However, to the best of our knowl-
edge, there has been no prospective
thin-section CT study that has examined
the morphology of the lung parenchyma
in asymptomatic older individuals. In an
increasingly aged population, establish-
ing the “normal” CT appearance of the
older lung is important in order to avoid
misdiagnosis of clinically important dis-
ease and the potentially harmful investi-
gations and treatment that might ensue.
The aim of our study was to describe the
thin-section pulmonary CT features in
elderly asymptomatic individuals who
were not known to have lung disease.
Materials and Methods
Institutional review board approval was
granted, and informed consent for a
limited CT examination and full pulmo-
Advance in Knowledge
䡲 Thin-section CT features usually
encountered in patients with in-
terstitial lung disease are present
in over half of asymptomatic el-
derly individuals who have no pul-
monary function deficit, and these
CT findings are absent in younger
individuals.
Radiology: Volume 251: Number 2—May 2009 ▪ radiology.rsnajnls.org
nary function tests (PFTs) was obtained
for all recruited individuals. Consecu-
tive individuals over 75 years of age or
under 55 years of age who had been
referred for CT of the abdomen or brain
on an outpatient basis between July
2003 and February 2007 were con-
tacted for participation in our study.
Individuals were initially mailed a
written explanation of the purpose of
our study and were subsequently called
by one of the investigators (S.J.C.,
J.M.H., A.E.T.J., D.J.G., or K.E.H.),
who completed a questionnaire. The
purpose of the questionnaire was to
confirm that the patient met the inclu-
sion criteria prior to CT or PFTs. Dys-
pnea was graded on a modified Medical
Research Council breathlessness score:
0 ⫽ none; 1 ⫽ shortness of breath on
strenuous activity, hurrying, or walking
uphill; 2 ⫽ walking on flat limited by
breathlessness; and 3 ⫽ breathless at
rest (9). Symptoms of a recent (within 6
weeks) respiratory infection or cough
or sputum production for greater than 3
months of the year were recorded. An
occupational and drug history was
taken. Exclusion criteria were current
smoking, history of smoking within
the past 5 years, any known chronic
pulmonary disease, any respiratory
symptoms (eg, over 3 months of cough
and sputum production per year and
symptoms of a chest infection within 2
months of the study) other than grade 1
dyspnea, sedentary lifestyle with re-
duced exercise tolerance (correspond-
ing to grade 2 or above on the modified
Medical Research Council question-
naire) owing to causes apart from
breathlessness, known cardiac failure,
malignancy, asbestos exposure or his-
tory of working in an occupation asso-
ciated with chronic lung disease (eg,
Implications for Patient Care
䡲 There is a spectrum of CT fea-
tures in asymptomatic older indi-
viduals that may not necessarily
represent clinically relevant dis-
ease.
䡲 Awareness of this phenomenon
may obviate unwarranted further
investigation or follow-up.
Copley et al
coal mining), history of drugs known to
be pneumotoxic (eg, chemotherapeutic
agents or amiodarone), known connec-
tive tissue disease (eg, rheumatoid ar-
thritis), and previous thoracotomy or
radiation therapy. Individuals who were
confused or unable to give written in-
formed consent and patients who were
unable to lie prone owing to musculo-
skeletal limitations (eg, severe kypho-
sis) were also excluded. No patients
with a history of working in farming or
keeping a pet bird were included.
Fifty-six consecutive asymptomatic
volunteers (40 ⬎ 75 years old and 16 ⬍
55 years old) were enrolled. Smoking
history in pack-years, dyspnea scores,
and the number of years of city dwell-
ing (within Greater London) were re-
corded. Thirty-eight individuals had
concurrent evaluation with PFTs.
CT Technique
Prone inspiratory thin-section CT imag-
ing of the thorax from the lung apices to
the bases (without intravenous contrast
agent) was performed in all subjects with
an eight-detector CT scanner (Lightspeed
Ultra; GE Healthcare, Waukesha, Wis)
with 1-mm section thickness, 20-mm sec-
tion interspacing, 120 kVp, and 400 mAs.
A large field of view was used, and images
were reconstructed with a high spatial
frequency algorithm.
Published online
10.1148/radiol.2512081242
Radiology 2009; 251:566 –573
Abbreviation:
PFT ⫽ pulmonary function test
Author contributions:
Guarantor of integrity of entire study, S.J.C.; study con-
cepts/study design or data acquisition or data analysis/
interpretation, all authors; manuscript drafting or manu-
script revision for important intellectual content, all au-
thors; approval of final version of submitted manuscript,
all authors; literature research, S.J.C., K.E.H., D.J.G.,
J.M.H., A.E.T.J.; clinical studies, S.J.C., K.E.H., D.J.G.,
J.M.H., A.E.T.J., D.M.H.; statistical analysis, S.J.C.,
A.U.W.; and manuscript editing, S.J.C., A.U.W., K.E.H.,
D.J.G., J.M.H., D.M.H.
Authors stated no financial relationship to disclose.
567
THORACIC IMAGING: CT Lung Morphology in the Elderly Copley et al

Assessment of CT Features defined as follows: 1 ⫽ trivial, 2 ⫽ less performed concurrently owing to pre-
All images were immediately reviewed than 50% of the diameter of the accom- booked clinical patients or equipment
for lung nodules or other potentially im- panying pulmonary artery, and 3 ⫽ be- failure or servicing. These individuals
portant unsuspected disease (eg, lung tween 50% and 100% of the diameter of were given appointments within a week
cancer), but none was detected in either the accompanying pulmonary artery. for PFTs but did not attend. We did not
group. The images were rendered anon- The extent of thickened interlobular have funding for the initial 10 individu-
ymous: All clinical and demographic septa was quantified as follows: 1 ⫽ als to have full PFT work-ups.
data, including date of birth, were re- more than four thickened septa that Forced expiratory volume in 1 sec-
moved. The images were reviewed on were limited to one lobe and 2 ⫽ wide- ond, forced vital capacity, and the ra-
soft copy on a workstation at a window spread bilateral distribution involving tio of the two were obtained by using
width (1500 HU) and level (⫺650 HU) two or more lobes. The presence of fea- flow volume loop spirometry (Morgan
appropriate for the assessment of lung tures was resolved by consensus. Benchmark; nSpire Health, Hertford,
parenchyma. Image sets were presented England). Total lung capacity and resid-
in random order. PFT Data ual volume were obtained by using
Two thoracic radiologists (S.J.C. Concurrent PFTs were obtained on the whole-body plethysmography (Sensor-
and D.M.H., with 10 and 20 years expe- same day as the CT imaging in 27 of the medics Autobox 6200; Cardinal Health,
rience with thoracic CT, respectively) 40 older (⬎75 years) individuals and 11 Warwick, England). Single-breath car-
who were blinded to the clinical data of the 16 younger (⬍55 years) individu- bon monoxide diffusing capacity cor-
independently scored all sections for the als. PFTs could be performed in the pul- rected for hemoglobin concentration
presence, extent (to the nearest 5% of a monary function laboratory at the time and adjusted for alveolar volume was
lobe), and distribution (ie, upper, mid- of CT for most subjects (n ⫽ 38). For acquired with gas transfer equipment
dle, lower, random, widespread, unilat- eight subjects, PFTs were not able to be by using helium and carbon monoxide
eral, bilateral, asymmetric, central, or
peripheral) of reticular pattern, ground Table 1
glass opacity, and centrilobular or para-
septal emphysema. A reticular pattern Demographic and PFT Data
adjacent to thoracic vertebral osteo- Parameter Older Group Younger Group P Value
phytes was not included in the reticular
Demographic datum
pattern score (10). The number and
No. of subjects 40 16
size of thin-walled cystic airspaces (dis-
Age (y) 80.6 ⫾ 4.2 39.4 ⫾ 7.5 ⬍.001*
tinct from centrilobular emphysema),
Man-to-woman ratio 18:22 8:8 NS†
the extent of bronchial dilation, bron-
Smoking history (pack-years)‡ 0 (0–50) 0 (0–10) NS§
chial wall thickening, and the presence Ex-smoker–to-nonsmoker ratio 17:23 3:13 NS†
of thickened interlobular septa were re- City dwelling history (y) 57.9 ⫾ 25.9 21.3 ⫾ 13.4 ⬍.001*
corded. These CT features were defined Body mass index (kg/m2) 29.1 ⫾ 5.6 30.7 ⫾ 6.2 NS*
according to the nomenclature recom- Dyspnea score㛳
mended by the Fleischner Society (11). 0 8 9
Reticular pattern coarseness was 1 32 7
categorized by using a three-point scale: PFT datum#
1 ⫽ very fine reticulation, 2 ⫽ pure re- No. of subjects 27 11
ticular pattern without honeycombing, FEV1 96.4 ⫾ 24.6 98.7 ⫾ 9.1
and 3 ⫽ reticular pattern with honey- Vital capacity 93 ⫾ 19.2 94.9 ⫾ 9.0
combing. The size of discrete pulmo- Total lung capacity 95.8 ⫾ 11.9 97.6 ⫾ 13.6
nary cysts was recorded in millimeters, Residual volume 109.9 ⫾ 22.2 103.5 ⫾ 23.6
and the profusion within each lobe was DLCO 86.9 ⫾ 17.9 102.4 ⫾ 13.6
recorded on a three-point scale: 1 ⫽ KCO 89.2 ⫾ 14.9 106.8 ⫾ 10.5
one cyst, 2 ⫽ two to four cysts, and 3 ⫽ Alveolar volume 86.9 ⫾ 12.7 98.7 ⫾ 11.9
five or more cysts. The extent of bron- Note.—Unless otherwise specified, data are means ⫾ standard deviations. DLCO ⫽ single-breath carbon monoxide diffusing
chial dilation was also evaluated by us- capacity corrected for hemoglobin concentration, FEV1 ⫽ forced expiratory volume in 1 sec, KCO ⫽ DLCO adjusted for alveolar
ing a three-point scale: 1 ⫽ trivial, 2 ⫽ volume, NS ⫽ not statistically significant.

internal diameter greater than but less * Paired t test.

than twice that of the accompanying

†
␹2 Test.
‡
pulmonary artery, and 3 ⫽ internal di- §
Data are medians, with ranges in parentheses.
Rank sum test.
ameter greater than twice that of the 㛳
Data are numbers of subjects.
accompanying pulmonary artery. The ex- #
Unless otherwise specified, PFT data are given as a percentage of the predicted value.
tent of bronchial wall thickening was

568 radiology.rsnajnls.org ▪ Radiology: Volume 251: Number 2—May 2009

THORACIC IMAGING: CT Lung Morphology in the Elderly Copley et al

(Morgan Benchmark; nSpire Health).
Digital pulse oximetry was used to mea-
sure oxygen saturation (Datex Ohmeda
3900; GE Healthcare, Amersham, En-
gland). Data were expressed as per-
centages of predicted values deter-
mined on the basis of the subject’s age,
sex, and weight (12). Body mass index
was calculated from the individual’s
weight and height.
Statistical Analysis
Statistical analysis was performed by us-
ing software (Stata; StataCorp, College
Station, Tex). The observers’ consensus
reading for the presence of features and
the mean extent of the observations were
used for analysis. Interobserver agree-
ment was evaluated by using the weighted
␬ statistic (13). Group comparisons were
made by using the ␹2, rank sum, or Fisher
exact test, as appropriate. Spearman
Table 2
PFT Results in the Older Group in Ex-Smokers versus Nonsmokers
PFT Result
FEV1
Vital capacity
Total lung capacity
Residual volume
DLCO
KCO
Alveolar volume
Note.—Data are means ⫾ standard deviations. None of the differences between subgroups were significant (paired t tests).
DLCO ⫽ single-breath carbon monoxide diffusing capacity corrected for hemoglobin concentration, FEV1 ⫽ forced expiratory
volume in 1 sec, KCO ⫽ DLCO adjusted for alveolar volume.
Table 3
Prevalence of CT Features in Older Group versus Younger Group
CT Feature
Reticular pattern
Cysts
Bronchial dilation
Bronchial wall thickening
Ground glass opacity
Interlobular septal thickening
Centrilobular emphysema
Note.—NS ⫽ not statistically significant.
* Data are numbers of subjects, with percentages in parentheses.
†
␹2 Test.
‡
Fisher exact test.
rank correlation was used to examine the Twenty-three of the subjects in the
correlation between CT features and pul- older group had never smoked, in com-
monary function data. Multiple regres- parison to 13 in the younger group. Of
sion analysis was performed, and inde- the individuals who underwent PFTs
pendent interrelationships among the CT (n ⫽ 38), 17 in the older group were
features and between CT features, pul- ex-smokers, in comparison to two in the
monary function, and smoking history younger group. For the older group, a
were examined by using stepwise for- comparison of PFT results between ex-
ward regression. Variables that were not smokers and nonsmokers is shown in
normally distributed were transformed Table 2.
prior to analysis (categorical or zero- The prevalence of CT findings is
skewness logarithmic transformations, as shown in Table 3. Interobserver agree-
appropriate). The assumptions of multi- ment (weighted ␬) for the presence of
ple linear regression were met for all CT features was moderate (cysts, weighted
analyses as determined by testing for het- ␬ ⫽ 0.56) to good (reticular pattern,
eroscedasticity and omitted variables. weighted ␬ ⫽ 0.61; bronchial dilation,
weighted ␬ ⫽ 0.68; bronchial wall thick-
ness, weighted ␬ ⫽ 0.67) (13). Ground
Results glass opacity, emphysema, and interlob-
Demographic data, including smoking ular septal thickening were recorded in
history in pack-years, dyspnea scores, too few individuals (Table 3) to allow
and PFT results, are shown in Table 1. observer agreement to be assessed.
There were no statistically signifi-
cant differences in the prevalence of CT
features between individuals with and
those without PFT data (Fisher exact
Ex-Smokers (n ⫽ 13) Nonsmokers (n ⫽ 14) test). In a subanalysis of the older group
wherein the results for individuals 80
87.7 ⫾ 25.1 102.9 ⫾ 22.6
years or older (n ⫽ 22) were compared
92.7 ⫾ 16.0 93.3 ⫾ 22.6
with those of subjects younger than 80
100.5 ⫾ 13.0 91.0 ⫾ 10.4
years (n ⫽ 18), there was a trend to-
118.7 ⫾ 25.4 101.2 ⫾ 16.0
ward increasing presence of the reticu-
85.9 ⫾ 16.9 87.7 ⫾ 19.3
87.7 ⫾ 15.2 90.5 ⫾ 15.0
lar pattern with increasing age, but it
86.6 ⫾ 12.5 87.2 ⫾ 13.3 was not statistically significant (P ⫽
.07). The extent of bronchial dilation in
the younger group (median, 0; range,
0 –2) was significantly lower (P ⬍ .001)
than that in the older group (median, 2;
range, 0 – 8). Similarly, the extent of
bronchial wall thickness in the younger
group (median, 0; range, 0 –2) was sig-
nificantly lower (P ⬍ .001) than that in
the older group (median, 1.5; range,
Older Group (n ⫽ 40)* Younger Group (n ⫽ 16)* P Value 0 – 6). Of the 32 older individuals with
airway abnormality, 14 had both airway
24 (60) 0 (0) ⬍.001†
dilation and wall thickening, whereas 18
10 (25) 0 (0) .02‡
had either dilation or wall thickening.
24 (60) 1 (6) ⬍.001†
22 (55) 1 (6) ⬍.001†
The one younger subject who had air-
0 (0) 1 (6) NS‡ way abnormality had both dilation and
7 (18) 0 (0) NS‡ thickening. By using both the Fisher ex-
2 (5) 0 (0) NS‡ act test and logistic regression, no link
was found between airway dilation and
wall thickening in the older group. In
addition, there were no significant asso-
ciations between cysts and the reticular
pattern, cysts and either airway abnor-

Radiology: Volume 251: Number 2—May 2009 ▪ radiology.rsnajnls.org 569

THORACIC IMAGING: CT Lung Morphology in the Elderly Copley et al

Table 4
mality, or the reticular pattern and ei-
ther airway abnormality. Prevalence of CT Features in the Older Group in Ex-Smokers versus Nonsmokers
The percentage of ex-smokers in the
CT Feature Ex-smokers (n ⫽ 17)* Nonsmokers (n ⫽ 23)* P Value
older group (42%) was higher than that
in the younger group (19%), but the Reticular pattern 11 (65) 13 (57) NS†
difference was not statistically signifi- Cysts 6 (35) 4 (17) NS‡
cant (P ⫽ .09). By using multiple logistic Bronchial dilation 9 (53) 15 (65) NS†
regression analysis, no statistically sig- Bronchial wall thickening 9 (53) 13 (57) NS†
nificant relationships between smoking Ground glass opacity 0 (0) 0 (0) ...
history (number of pack-years) and the Interlobular septal thickening 1 (6) 6 (26) NS‡
following CT features were observed in Centrilobular emphysema 2 (12) 0 (0) NS‡
the older group: reticular pattern (P ⫽ Note.—NS ⫽ not statistically significant.
.4), bronchial dilation (P ⫽ .43), bron- * Data are numbers of subjects, with percentages in parentheses.
chial wall thickness (P ⫽ .82), and cysts †
␹2 Test.
(P ⫽ .20). Table 4 shows no statistically ‡
Fisher exact test.
significant differences in CT features be-
tween ex-smokers and nonsmokers in
the older group. There was no statisti-
cally significant correlation between
Table 5
PFT data and CT features in the older
group. Characteristics of Reticular Pattern in the Older Group by Observer
Table 5 shows the distribution, ex-
Characteristic Observer 1 Observer 2
tent, and coarseness of the reticular pat-
terns noted by the two observers (S.J.C. No. of subjects with reticular pattern 30 23
and D.M.H.). The predominant finding Distribution
was a limited extent (median, 5%; maxi- Upper lobes 1 (3) 1 (4)
mum, 20%) of pure reticular pattern Right middle and lingula 0 (0) 0 (0)
without honeycombing with lower lobe, Lower lobes 29 (97) 22 (96)
bilateral, peripheral distribution (Figs 1 Random 3 (10) 4 (17)
and 2). Only two older individuals were Widespread 0 (0) 2 (9)
Unilateral 0 (0) 0 (0)
recorded as having centrilobular emphy-
Bilateral 30 (100) 23 (100)
sema, and one younger individual had
Asymmetric 2 (7) 0 (0)
limited (5%) ground glass opacity. The
Central 1 (3) 0 (0)
distribution of these features was not an-
Peripheral 28 (93) 20 (87)
alyzed further. Of the seven older individ-
Extent*
uals with interlobular septal thickening, Upper lobes 2.5 (2.5–5) 2.5 (2.5–5)
two had limited distribution, and five had Right middle and lingula 5 (5) 5 (5)
widespread bilateral distribution. The Lower lobes 5 (2.5–20) 5 (5–20)
distribution of cysts, bronchial dilation, Coarseness
and bronchial wall thickening by observer Grade 1 13 (43) 5 (22)
is shown in Table 6. Cysts were distrib- Grade 2 17 (57) 18 (78)
uted uniformly throughout the lungs (Figs Grade 3 0 (0) 0 (0)
3 and 4), whereas there was striking
Note.—Unless otherwise specified, data are numbers of patients, with percentages in parentheses.
lower lobe predominance for both bron-
* Data are medians, with ranges in parentheses. Values are given in percentages.
chial dilation and thickening (Fig 5).

Discussion
We have documented that asymptom-
atic older individuals have a relatively
high frequency of a basal subpleural re-
ticular pattern and cysts, independent
of smoking history. Bronchial dilation
and bronchial wall thickening were also
common in the older group, similar to
results of previous studies (14,15). All
of the described CT features were inde-
pendent of smoking history. Our find-
ings are important because there is the
potential for confusion with the CT ap-
pearance of clinically relevant intersti-
tial lung disease, which could result in
unnecessary follow-up and potentially
harmful treatment. However, the CT
patterns described in our study may not
be of clinical importance because the
majority of individuals had normal pul-
monary function, and thus, these find-
ings likely reflect the normal spectrum
of morphology of aging lung.
The cause of the CT findings could
not be evaluated because correlation
with histopathologic findings was not
possible. However, previous histopatho-

570 radiology.rsnajnls.org ▪ Radiology: Volume 251: Number 2—May 2009

THORACIC IMAGING: CT Lung Morphology in the Elderly Copley et al

logic studies (3,16) have established that Another explanation could be alveo- with aging, thus predisposing the alveoli
there is increased collagen deposition and lar collapse in relatively underventilated to collapse (1). Furthermore, in a large
progressive fibrosis in the lungs of rats portions of the lung; however, the retic- longitudinal radiographic study from the
with aging. The fine subpleural basal re- ular pattern was identifiable in nonde- 1950s (4), the authors noted that the
ticular pattern observed in our study pendent subpleural lung. In animal stud- volume of the lower lungs diminished
was not associated with traction bron- ies, the collagen matrix that supports with time and that the ribs were more
chiectasis, an ancillary CT feature of the alveolar walls becomes less elastic closely approximated basally. Interest-
fibrosis, and therefore, the observed ingly, a study (17) of pulmonary auscul-
finding may not represent actual es- tation has shown a small but statistically
tablished fibrosis (although this may Figure 2 significant increase in higher frequency
simply be owing to the limited extent sounds with increasing age, which the
of the reticular pattern). authors did not deem to be clinically
important; however, they did not spec-
ulate on the genesis of these sounds.
Figure 1 Histopathologic studies have shown in-
creased thickening of interlobular septa
without inflammation or fibrosis in the
lungs of elderly individuals (1). In-
creased interlobular septal thickening
on CT images was seen in a 6-year fol-
low-up study of 60-year-old smokers
and nonsmokers in a Swedish cohort
(15). In our study, thickening of inter-
lobular septa was observed more fre-
Figure 1: Axial prone thin-section CT image of quently in the older group (seven sub-
a 92-year-old woman who had normal pulmonary Figure 2: Axial prone thin-section CT image of jects in the older group vs none in the
function shows a subtle bilateral subpleural basal an asymptomatic 78-year-old woman who was a younger group), although the observa-
reticular pattern of limited extent (grade 2). This nonsmoker and had been a city dweller for 39 tion did not reach statistical signifi-
individual was a nonsmoker and had been a city years. Image shows a subpleural bilateral fine cance.
dweller for 76 years. reticular pattern of limited extent (grade 2). It may not be possible to accurately
assess the histopathologic entity of “se-
nile emphysema” (4), as judged with CT
signs of overt emphysema, by using our
data. Only two individuals in our study
Table 6
were reported to have centrilobular em-
Distribution of Other CT Features in the Older Group by Observer physema, and both were ex-smokers.
Characteristic Observer 1 Observer 2 Subjective CT detection of subtle em-
physema is likely to be unreliable and
Cystic Air Spaces
associated with substantial interob-
No. of subjects 16 15 server variation, which is a disadvan-
Upper lobes 6 (38) 6 (40) tage in comparison to objective meth-
Middle lobe and lingula 3 (19) 3 (20)
ods (18). However, both observers
Lower lobes 7 (44) 6 (40)
identified these two cases indepen-
Size (mm)* 11 (7–20) 13 (5–22)
dently in our study. Despite the sub-
Bronchial Dilation
tlety of the various CT findings, interob-
No. of subjects 32 30
Upper lobes 3 (9) 4 (13)
server agreement was moderate to
Middle lobe and lingula 3 (9) 2 (7)
good for all features (13). Future stud-
Lower lobes 26 (81) 24 (80) ies that use objective densitometric
Bronchial Wall Thickening analysis and more sophisticated post-
No. of subjects 30 29 processing may reveal cryptic emphy-
Upper lobes 3 (10) 4 (14) sematous changes. The histopatho-
Middle lobe and lingula 4 (13) 2 (7) logic definition of senile emphysema is
Lower lobes 23 (77) 23 (79) not straightforward. Conventionally,
emphysema is defined as abnormal en-
Note.—Unless otherwise specified, data are numbers of patients, with percentages in parentheses.
largement of airspaces distal to termi-
* Data are means, with ranges in parentheses.
nal bronchioles with destruction of al-

Radiology: Volume 251: Number 2—May 2009 ▪ radiology.rsnajnls.org 571

THORACIC IMAGING: CT Lung Morphology in the Elderly
veolar walls (19). However, there may
simply be an increase in airspace size
associated with advancing age in adult
lungs (20). Therefore, the term “senile
emphysema” should be discouraged ow-
ing to the lack of parenchymal destruc-
tion, and “senile” or “senescent lung”
may be a better descriptor (4,21).
Other CT studies have not demon-
strated a clear relationship between ag-
ing and emphysema. Tylén et al (22)
performed a study on both smoking and
nonsmoking asymptomatic 60 year olds,
and a subsequent study after 6-year fol-
low-up was published (15). The initial
study (22) found emphysematous changes
(defined vaguely by “stretched vessels”)
in only one nonsmoker, which the au-
thors dismissed as a false-positive
finding. The follow-up study (15) of the
original cohort demonstrated septal
lines, subpleural nodules, and focal
ground glass opacity in both smokers
and nonsmokers, with increased fre-
quency in smokers, suggesting a syner-
gistic effect between age and smoking in
the development of these CT abnormal-
ities. A reticular pattern and parenchy-
mal cysts were not findings in the study
by Vikgren et al (15) and were not spe-
cifically assessed. The subjects in our
study were substantially older (median
age, 80 years) than the 66 year olds in
the follow-up study.
An interesting and new CT finding in
our study was the presence of thin-
walled cysts in the older population in a
substantial proportion of the nonsmok-
ers, with no correlation with pack-years
of cigarette smoking. In histopathologic
studies of lung diseases in which cysts
are a major feature (eg, lymphoid inter-
stitial pneumonia), the mechanism likely
to account for this CT finding is unclear,
although a “check-valve” phenomenon
owing to partial bronchiolar obstruction
has been postulated (23).
Previous CT studies (14,24) have
described both large and small airway
abnormalities in apparently healthy older
individuals. The frequency and extent of
air trapping significantly increased with
age in a prospective study by Lee et al
(24), but relatively few individuals over
75 years old were included. In a pro-
spective study of the large airways by
572
Copley et al
Figure 3 Figure 4
Figure 4: Axial prone thin-section CT image of
Figure 3: Axial prone thin-section CT image of
a 76-year-old man shows a thin-walled cystic
a 77-year-old woman who was a nonsmoker
airspace in the posterior left lower lobe. He had a
shows a 2-cm (diameter) thin-walled cystic air-
15 pack-year smoking history but had stopped
space within the right upper lobe. Bronchial wall
smoking 30 years previously.
thickening was scored as 1 in right and left lower
lobes.
Figure 5
Matsuoka et al (14), a significant corre-
lation between age and increasing bron-
choarterial ratio was demonstrated in
asymptomatic healthy subjects, which
included overlap with the CT definition
of bronchiectasis. In contrast to our
study, no PFT data were obtained, and
there were only 27 subjects greater than
65 years old (mean, 74 years). The ex-
planation for these large and small air-
way changes may be impaired large (25)
and small airway clearance mechanisms
with advancing age (26). Decreased al-
veolar defenses (25) and low-grade in- Figure 5: Axial prone thin-section CT image of
flammation of the respiratory tract have an asymptomatic 75-year-old man shows mild
also been described in asymptomatic el- bilateral lower lobe bronchial dilation. He had
derly volunteers (27,28), which could been a city dweller for 31 years and had a 10 pack-
year smoking history but had not smoked for 38
contribute to bronchial wall thickening
years.
and bronchial dilation.
A limitation of our study was that
the group did not purely consist of non-
smokers. However, an advantage of in- factor, and forced expiratory volume in
cluding ex-smokers, apart from the 1 sec–to–forced vital capacity ratio are
group being more representative of the well recognized (1), which makes accu-
general population, was that we were rate functional assessment of normality
able to demonstrate that there was no difficult. However, in the individuals
correlation between the presence of the with PFT data, there was no correlation
observed CT features and smoking his- between pulmonary function and CT
tory. features.
Another limitation of our study was The use of 1-mm sections every 20
that not all of our study group had PFT mm was specifically designed to mini-
results. Accurate assessment of pulmo- mize the radiation dose of this non–
nary function in the elderly is notori- clinically indicated imaging, particularly
ously problematic, and an age-related since we included individuals younger
decrease in lung volume, gas transfer than 55 years of age. The use of noncon-
radiology.rsnajnls.org ▪ Radiology: Volume 251: Number 2—May 2009
THORACIC IMAGING: CT Lung Morphology in the Elderly
tiguous sections could potentially have
resulted in underestimation of the pres-
ence of features that were limited in
extent; however, despite this limitation,
our results were still statistically signif-
icant.
There may have been selection bias
owing to the fact that we identified indi-
viduals without respiratory symptoms
who were undergoing a clinically indi-
cated brain or abdominal CT and, there-
fore, were not truly representative of the
healthy aged population. The patients at
our institution are mainly urban dwellers;
therefore, many of the individuals had
been lifelong city dwellers. It was not
possible to evaluate the independent re-
lationship of years of city dwelling with
the presence of observed CT features
because of strong colinearity between
age and years of city dwelling. It is
therefore possible that there is an envi-
ronmental cause for the CT findings we
identified, which might not be repro-
duced in a rural population. However,
the findings are still relevant to the ur-
ban aged population.
In conclusion, we identified CT fea-
tures in asymptomatic older individuals
that were significantly different from
those seen in younger subjects. The
main findings of our study were a lim-
ited extent of predominantly bilateral
basal subpleural reticular pattern and
occasional scattered cystic air spaces.
The clinical relevance of recognizing
these findings is that they may be part of
the normal spectrum of senescent lung
and should not be over-interpreted to
represent clinically important disease.
In an increasingly elderly population,
definition of normality in such individu-
als will become important.
References
1. Janssens JP, Pache JC, Nicod LP. Physiolog-
ical changes in respiratory function associ-
ated with ageing. Eur Respir J 1999;13(1):
197–205.
2. Gibson GJ, Pride NB, O’Caine C, Quagliato
R. Sex and age differences in pulmonary me-
chanics in normal nonsmoking subjects.
J Appl Physiol 1976;41(1):20 –25.
Radiology: Volume 251: Number 2—May 2009 ▪ radiology.rsnajnls.org
3. Calabresi C, Arosio B, Galimberti L, et al.
Natural aging, expression of fibrosis-related
genes and collagen deposition in rat lung.
Exp Gerontol 2007;42(10):1003–1011.
4. Mayer E, Rappaport I, Blazsik C. Emphy-
sema and the lungs of the aged: a clinical
study—preliminary report. Dis Chest 1958;
34(3):247–256.
5. Ensor RE, Fleg JL, Kim YC, de Leon EF,
Goldman SM. Longitudinal chest x-ray
changes in normal men. J Gerontol 1983;
38(3):307–314.
6. Mathieson JR, Mayo JR, Staples CA, Müller
NL. Chronic diffuse infiltrative lung disease:
comparison of diagnostic accuracy of CT and
chest radiography. Radiology 1989;171(1):
111–116.
7. Padley SP, Hansell DM, Flower CD, Jennings
P. Comparative accuracy of high resolution
computed tomography and chest radiogra-
phy in the diagnosis of chronic diffuse infil-
trative lung disease. Clin Radiol 1991;44(4):
222–226.
8. Grenier P, Valeyre D, Cluzel P, Brauner
MW, Lenoir S, Chastang C. Chronic diffuse
interstitial lung disease: diagnostic value of
chest radiography and high-resolution CT.
Radiology 1991;179(1):123–132.
9. Standardized questionaries on respiratory
symptoms. Br Med J 1960;2:1665.
10. Otake S, Takahashi M, Ishigaki T. Focal pul-
monary interstitial opacities adjacent to tho-
racic spine osteophytes. AJR Am J Roentge-
nol 2002;179(4):893– 896.
11. Hansell DM, Bankier AA, MacMahon H,
McLoud TC, Müller NL, Remy J. Fleischner
Society: glossary of terms for thoracic imag-
ing. Radiology 2008;246(3):697–722.
12. Roberts CM, MacRae KD, Winning AJ,
Adams L, Seed WA. Reference values and
prediction equations for normal lung func-
tion in a non-smoking white urban popula-
tion. Thorax 1991;46(9):643– 650.
13. Brennan P, Silman A. Statistical methods for
assessing observer variability in clinical mea-
sures. BMJ 1992;304(6840):1491–1494.
14. Matsuoka S, Urchiyama K, Shima H, Ueno
N, Oish S, Nojiri Y. Bronchoarterial ratio
and bronchial wall thickness on high-resolu-
tion CT in asymptomatic subjects: correla-
tion with age and smoking. AJR Am J Roent-
genol 2003;180(2):513–518.
15. Vikgren J, Boijsen M, Andelid K, et al. High-
resolution computed tomography in healthy
smokers and never-smokers: a 6-year fol-
Copley et al
low-up study of men born in 1933. Acta Ra-
diol 2004;45(1):44 –52.
16. Mays PK, Bishop JE, Laurent GJ. Age-re-
lated changes in the proportion of types I and
III collagen. Mech Ageing Dev 1988;45(3):
203–212.
17. Gross V, Dittmar A, Penzel T, Schüttler F,
von Wichet P. The relationship between nor-
mal lung sounds, age and gender. Am J Re-
spir Crit Care Med 2000;162(3):905–909.
18. Bankier AA, de Maertelaer V, Keyzer C,
Gevenois PA. Pulmonary emphysema: sub-
jective visual grading versus objective quan-
tification with macroscopic morphometry
and thin-section CT densitometry. Radiology
1999;211(3):851– 858.
19. The definition of emphysema: report of a
National Heart, Lung and Blood Institute,
Division of Lung Diseases workshop. Am
Rev Respir Dis 1985;132(1):182–185.
20. Gillooly M, Lamb D. Microscopic emphy-
sema in relation to age and smoking habit.
Thorax 1993;48(5):491– 495.
21. Verbeken EK, Cauberghs M, Mertens I,
Clement J, Lauweryns JM, Van der Woesti-
jne KP. The senile lung: comparison with
normal and emphysematous lungs. I. Struc-
tural aspects. Chest 1992;101(3):793–799.
22. Tylén U, Boijsen M, Ekberg-Jansson A, Bake
B, Löfdahl CG. Emphysematous lesions and
lung function in healthy smokers 60 years of
age. Respir Med 2000;94(1):38 – 43.
23. Nicholson AG. Lymphocytic interstitial pneu-
monia and other lymphoproliferative disor-
ders of the lung. Semin Respir Crit Care Med
2001;22(4):409 – 422.
24. Lee KW, Chung SY, Yang I, Lee Y, Ko EY,
Park MJ. Correlation of aging and smoking
with air trapping at thin-section CT of the
lung in asymptomatic subjects. Radiology
2000;214(3):831– 836.
25. Gyetko MR, Toews GB. Immunology of the
aging lung. Clin Chest Med 1993;14(3):379 –
391.
26. Svartengren M, Falk R, Philipson K. Long
term clearance from small airways decreases
with age. Eur Respir J 2005;26(4):609 – 615.
27. Meyer KC, Ershler W, Rosenthal NS, Lu XG,
Peterson K. Immune dysregulation in the ag-
ing human lung. Am J Respir Crit Care Med
1996;153(3):1072–1079.
28. Meyer KC, Rosenthal NS, Soergel P, Peterson
K. Neutrophils and low-grade inflammation in
the seemingly normal aging human lung. Mech
Ageing Dev 1998;104(2):169 –181.
573