Sei sulla pagina 1di 8

The British Journal of Radiology, 73 (2000), 930937

2000 The British Institute of Radiology

Rapidly growing small peripheral lung cancers detected


by screening CT: correlation between radiological
appearance and pathological features
1

J-C WANG, MD, 1S SONE, PhD, MD, 1L FENG, PhD, MD, 1Z-G YANG, MD,
1
S TAKASHIMA, PhD, MD, 1Y MARUYAMA, PhD, MD, 1M HASEGAWA, MD, 1S KAWAKAMI, MD,
2
T HONDA, PhD, MD and 3T YAMANDA, PhD, MD
Departments of 1Radiology, 2Laboratory Medicine and 3Surgery, Shinshu University School of Medicine,
Asahi, Matsumoto 390-8621, Japan

Abstract. 12 peripheral small lung cancers (,20 mm) of rapid growth (volume doubling time
,150 days), detected by repeated low dose CT screening, were evaluated to examine their CT
features and to correlate such features with histopathological ndings. Each patient's CT images,
including follow-up and thin section CT images, were studied retrospectively to determine tumour
growth rate and CT morphological features. Nine of the tumours exhibited a solid tumour
growth pattern: seven of these showed a well dened, homogeneous, soft tissue density with
spicular or lobulated margin. These seven tumours included small cell lung cancer (n53),
moderately differentiated adenocarcinoma (n52), poorly differentiated adenocarcinoma (n51)
and squamous cell carcinoma (n51). The other two tumours, a moderately differentiated
adenocarcinoma and a well differentiated adenocarcinoma, appeared as irregular, soft tissue
density nodules with poorly dened margins. The latter exhibited an air bronchogram pattern and
a small cavity. The remaining three tumours exhibited a lepidic tumour growth pattern. They
showed ground glass opacity or ground glass opacity with a higher density central zone on CT
images and were well differentiated adenocarcinomas. In conclusion, most peripheral small lung
cancers of rapid growth were adenocarcinomas. They also included small cell lung cancer and
squamous cell carcinoma. The majority showed solid tumour growth pattern and lacked an air
bronchogram and/or small air spaces in the nodule. Some well differentiated adenocarcinomas
with lepidic tumour growth pattern also showed rapid growth.
The prognosis of lung cancer correlates well
with the tumour volume doubling time (VDT) [1,
2] and the latter in turn correlates with histopathological type [24]. Furthermore, some studies have indicated that the prognosis of
surgically resectable tumour is better than that
of non-resectable tumour. The essential issue for a
successful surgical excision is detection of early
stage lung cancers. Thus, accurate characterization of the CT features of rapidly growing small
lung cancers is crucial for early diagnosis and
treatment to improve patient prognosis.
Low dose spiral CT has recently been introduced to detect lung cancers in a population-based
screening study. As many lesions detected in low
dose CT images are small and do not qualify for
immediate histological evaluation, they are often
Received 25 January 2000 and in revised form 28 April
2000, accepted 4 May 2000.
Address correspondence to Shusuke Sone, PhD, MD,
Department of Radiology, Shinshu University School
of Medicine, 3-1-1 Asahi, Matsumoto, Nagano 3908621, Japan.
930

re-examined after an interval. At the later


examination, a change in morphological appearances is considered highly signicant for making a
nal diagnosis since most benign and malignant
nodules change in size, shape and density in
different ways. This enables an evaluation of the
natural history of rapidly growing small lung
cancers (,20 mm) on CT images.
To our knowledge, no studies have previously
characterized the natural history of rapidly
growing small lung cancers based on CT
images, and a correlation between the CT
characteristics of such tumours on thin section
CT image and histopathological features is not
available. Our goal was to determine the growth
rate and CT features of rapidly growing small
lung cancers and correlate these to histopathological ndings.

Materials and methods


Between May 1996 and May 1999, annual
screening for lung cancer was conducted in
Nagano Prefecture of Japan using low dose
The British Journal of Radiology, September 2000

CT features of rapidly growing small lung cancers

spiral CT. A total of 81 lung cancers was found in


this study; 34 of these lung cancers received initial
and annual repeat low dose CT screenings,
permitting us to observe the interval change of
the lesions. We reviewed the repeat CT of these
patients to determine the VDTs, and we evaluated
diagnostic CT images to assess the morphological
features of rapidly growing peripheral small lung
cancers. We dened a rapidly growing peripheral
small lung cancer as a tumour smaller than 2 cm
in maximum diameter, a solitary parenchymal
nodule in the lung, with a VDT,150 days. The
following inclusion criteria were selected: (1) the
nodule was found incidentally on annual rescreening CT, and follow-up examinations were
performed by diagnostic CT, including thin
section CT; (2) tumours were peripheral, i.e.
located within the peripheral two-thirds of the
lung on CT images, without contact with lobar or
segmental bronchi; (3) the period between initial
screening CT and nal diagnostic CT was .150
days; (4) the VDT of the tumour was ,150 days;
(5) targeted, thin section (high resolution) CT
(HRCT) images with collimation of 1 mm taken
within 1 month prior to surgery were available for
review; and (6) the lesion was subsequently
treated by surgery.
12 patients (10 male and 2 female), ranging in
age from 56 years to 76 years (mean 70 years)
fullled the inclusion criteria. Initial screening CT
examination was performed using a low dose
spiral CT (model CT-W950SR, Hitachi Medical
Co., Japan). The technical scan parameters were:
120 kVp, 50 mA (11 cases) or 25 mA (1 case),
10 mm s21 table speed, 2 s per rotation of the Xray tube, 10 mm collimation, 10 mm reconstruction interval with a standard reconstruction
algorithm and data processing algorithm of 180
linear interpolation. A state-of-the-art CT scanner
(Hi-speed Advantage, GE Medical System,
Milwaukee, WI) was used in the diagnostic CT
examination. CT images were obtained from the
lung apices to the lung bases with 10 mm
collimation.
Technical
parameters
were:
120 kVp, 200 mA, 1 s scan time, 10 mm collimation and 320 mm eld of view. One additional
targeted spiral CT sequence was performed
through the nodule with 1 mm collimation in
each patient and CT images were reconstructed
with a bone algorithm, 20 cm eld of view and
0.5 mm reconstruction interval. All CT images
were obtained during breath-holding at mid
inspiration.
The lung tumour VDT was calculated using the
Schwartz formula [5]. Initial tumour size was
measured on CT images shown on the cathode
ray tube monitor, using a window width of
1000 HU and window level of2700 HU. Lesions
in ve subjects were invisible or ,3 mm in
The British Journal of Radiology, September 2000

diameter on the initial screening CT images.


Taking partial volume effect into account, we
dened the size of each of these ve tumours to be
3 mm in diameter. Because intervals between
patients' repeated low dose CT screenings and
nal thin section CT imaging varied, and the
longest interval between initial and nal CT
images will yield a more accurate VDT, nal
tumour size was measured on nal thin section
CT images. Two observers (J-CW, SS), blinded
to the pathological ndings, independently
reviewed each CT image. They recorded the CT
features of each tumour, including density,
margin and internal texture. Discrepancies in
interpretation between the observers were
resolved by consensus. The medical records were
examined for history of exposure to well documented pulmonary carcinogens such as cigarette
smoke and asbestos.
All surgical specimens were xed in an inated
state by transbronchial infusion of formalin liquid
and were sliced transversely at the centre of the
tumour to provide optimal correlation with the
HRCT image. The gross appearance of the nodule
and the microscopic ndings on the hematoxylin
eosin stained pathological material was examined
in each case. The margin and internal texture of
each tumour were assessed on thin section CT
images and then correlated with pathological
ndings.

Results
Among the ten male patients, eight were heavy
smokers (>30 pack-years), one was a mild smoker
(,30 pack-years) and the other one was a nonsmoker with a family history of malignancy. Of
the two female patients, one was a passive smoker
while the other was a non-smoker with a family
history of malignancy. The mean size of the 12
lesions on initial screening CT images was 4.7 mm
(range 3.08.5 mm), and 11.5 mm (range
6.014.5 mm) on the nal, thin section CT
images. The tumour VDT ranged from 54 days
to 132 days (Table 1).
Among the 12 cases of rapidly growing lung
cancer, four of eight adenocarcinomas were well
differentiated, three were moderately differentiated and one was a poorly differentiated
adenocarcinoma. The remaining four cases
included one poorly differentiated squamous cell
carcinoma (SCC) and three small cell lung cancers
(SCLCs) in the lung periphery. 9 of the 12 cancers
were located in the right lung and 3 were in the
left lung. The proportion of rapidly growing
cancers among the 34 small peripheral cancers
that were detected by annual repeat CT was 28%
(8/29) of adenocarcinomas, 50% (1/2) of SCCs
and 100% (3/3) of SCLCs.
931

J-C Wang, S Sone, L Feng et al


Table 1. Data for 12 patients with rapidly growing small lung cancer
No.

Age (years)
/sex

Smoking history
(pack-years)

Diameter on
initial CT
(mm)

Time to
detection
(days)

Diameter at
diagnosis
(mm)

VDT
(days)

Location

Cell type

1
2
3
4
5
6
7
8
9
10
11
12

67/M
75/M
68/M
73/M
76/M
70/F
64/M
56/M
69/M
70/F
74/M
71/M

34
25
72
40
40
Passive smoker
30
30
34
0
30
0

3Z3
3Z3
3Z3
3Z3
3Z6
3Z3
5Z4
5Z8
3Z5
8Z5
10Z7
6Z8

366
299
372
357
396
299
363
380
376
361
159
386

15Z14
17Z7
11Z11
13Z8
14Z10
6Z6
13Z9
15Z13
7Z7
17Z9
14Z9
18Z11

54
60
66
72
84
100
120
127
128
130
131
132

RLL
RUL
LLL
RLL
LUL
RUL
RML
RUL
RUL
RLL
RLL
LUL

SCLC
SCLC
PD SCC
MD Adeno
WD Adeno
WD Adeno
MD Adeno
PD Adeno
WD Adeno
WD Adeno
MD Adeno
SCLC

VDT, tumour volume doubling time.


RLL, right lower lobe; RUL, right upper lobe; LLL, left lower lobe; LUL, left upper lobe; RML, right middle lobe.
SCLC, small cell lung carcinoma; PD SCC, poorly differentiated squamous cell carcinoma; MD Adeno, moderately differentiated
adenocarcinoma; WD Adeno, well differentiated adenocarcinoma; PD Adeno, poorly differentiated adenocarcinoma.

All cancers were treated by surgical resection.


The pathological and clinical stages were
T1N0M0 in eight patients, T1N1M0 in two,
T1N2M0 in one and T2N1M0 in one patient. The
patient with a moderately differentiated adenocarcinoma (T2N1M0) had invasion of the visceral
pleura by the tumour and died 8 months after
operation owing to pulmonary metastases. The
remaining 11 patients were still alive at 1524
months after surgery.
Most of the rapidly growing lung cancers
appeared as soft tissue density nodules on CT
images (n59) (Figure 1); however, two nodules
exhibited ground glass opacities (GGOs)
(Figure 2) and one nodule showed GGO with a
higher density central zone. Tumours with
homogeneous attenuation and well dened,
smooth margins (n57) were more common than
those with heterogeneous attenuation and ill
dened, irregular margins (n55). Half of the
tumours had spiculation and few of them showed
a pleural tag (n53) (Figure 3), convergence of
pulmonary vessels towards the tumour (n51) or
halo sign (n51). Air bronchogram and small air
spaces were found in only one tumour (Figure 4;
Table 2).
Regarding the CT manifestations according to
histology, well differentiated adenocarcinomas
tended to show GGO (n53), although one
tumour showed soft tissue density. GGO in the
adenocarcinoma was based on lepidic tumour
growth (alveolar lining tumour cell growth)
histologically (Figure 2c). Moderately differentiated adenocarcinomas were more likely to
932

show homogeneous soft tissue density with a


well dened margin (Figure 3), which corresponded well to the histology showing solid
tumour growth. Poorly differentiated adenocarcinomas, SCC and parenchymal SCLC appeared as
homogeneous, soft tissue density nodules with
well dened, lobulated margins and ne spiculation, which correlated with solid tumour growth
on the histological specimen (Figure 1). Small cell
lung cancers characteristically showed a well
dened, distinctly lobulated appearance.

Discussion
The VDT of lung nodules has been widely
accepted as an index of tumour growth rate.
Steele and Buell [6] suggested that a VDT of
30490 days represents a malignant zone indicative of malignancy, while a VDT outside the
above range was referred to as benign zone. The
VDTs of lung cancer were noted to have a wide
range among the same histological type [1, 4] and,
furthermore, several studies have suggested varying VDTs at different stages of lung cancers [1, 7].
The currently available data on VDT of lung
cancer were based on chest radiographs [3, 7],
which usually do not allow detection of small lung
cancer (,10 mm). There is therefore a need to
dene the growth patterns of smaller lung cancers
such as those detected initially on CT scan.
There is currently no consensus on the denition of rapidly growing lung cancer. According to
Usuda et al [1] and Hayabuchi et al [8], who
dened the cut-off VDTs between rapidly and
The British Journal of Radiology, September 2000

CT features of rapidly growing small lung cancers

(a)

(b)

(c)

Figure 1. Case 1: a small cell carcinoma in the right


lower lobe in a 67-year-old male patient. (a) Initial
screening CT (left) suggests a small nodule (arrow)
(,3 mm) with an ill dened margin in the right lower
lobe. On the re-screening CT image (right), 1 year
later, the nodule had increased in size and presented
as a soft tissue density nodule with a lobulated
margin. This interval change strongly suggested that
the nodule needed further thin section CT examination to rule out cancer. (b) Pre-operative thin section
CT image shows a homogeneous soft tissue density
nodule with a well dened lobulated margin (arrows).
The tumour size is 14 mm Z 15 mm on the thin section CT. (c) Pathological specimen (Z1.25) shows
small cell carcinoma (intermediate cell type) with
smooth margin and homogeneous internal texture.

Table 2. Thin section CT ndings according to histopathological type in rapidly growing small peripheral lung
cancer
Thin section CT ndings

n
Nodule density
Soft tissue density
GGO
Nodule margin
Well dened
Poorly dened
Smooth
Irregular
Spiculation
Lobulation
Pleural tag
Convergence of vessels and bronchi
Halo sign
Internal features
Homogeneous
Heterogeneous
Air bronchogram
Small air space
Calcication

Histological type
WD Adeno

MD Adeno

PD Adeno

PD SCC

SCLC

Total

12

1
3

3
0

1
0

1
0

3
0

9
3

1
3
0
4
1
0
0
0
1

1
2
2
1
1
2
3
1
0

1
0
1
0
1
1
0
0
0

1
0
1
0
1
1
0
0
0

3
0
3
0
2
3
0
0
0

7
5
7
5
6
7
3
1
1

0
4
1
1
0

2
1
0
0
1

1
0
0
0
0

1
0
0
0
0

3
0
0
0
0

7
5
1
1
1

GGO, ground glass opacity; WD Adeno, well differentiated adenocarcinoma; MD Adeno, moderately differentiated
adenocarcinoma; PD Adeno, poorly differentiated adenocarcinoma; PD SCC, poorly differentiated squamous cell carcinoma;
SCLC, small cell lung carcinoma.

The British Journal of Radiology, September 2000

933

J-C Wang, S Sone, L Feng et al

(a)

(b)

Figure 2. Case 5: a well differentiated adenocarcinoma in the left upper lobe in a 76-year-old male
patient. (a) Initial screening CT (left) shows a 3 mm
Z 6 mm ground glass opacity nodule (arrow) in the
left upper lobe identied retrospectively. The rescreening CT image 1 year later (right) shows that
the size of the ground glass opacity has enlarged at
the same position (arrow). (b) Pre-operative thin section CT image shows a ground glass opacity nodule
with an ill dened margin. The nodule is 14 mm Z
10 mm in size. (c) Pathological specimen (Z1.25)
shows well differentiated adenocarcinoma with alveolar lining tumour growth and mild thickening of the
alveolar septa (type A of Noguchi's classication).
(c)

slowly growing lung cancers at 113 days and 150


days, respectively, and taking into account the
slower growth rate shown by small lung cancers
compared with large ones [1], in this study we
dened lung cancers with a VDT of less than 150
days as being rapidly growing.
Peripheral lung cancers arise from the small
airways of the lung. Their growth pattern has
traditionally been classied into two types: hilic
(solid) and lepidic growth [9]. In the hilic growth
pattern, the tumour develops as a solid mass,
displacing the surrounding lung. The tumour may
be smooth or irregular in margin, round or
lobulated in conguration, and shows mostly
homogeneous soft tissue density on CT images
[10]. In this study, most (n59) of the rapidly
growing lung cancers showed solid tumour
growth pattern in pathological specimens; three
were SCLCs, three moderately differentiated
adenocarcinomas, one well differentiated adenocarcinoma, one poorly differentiated adenocarcinoma and one SCC. Seven of these showed a well
dened homogeneous soft tissue density.
934

However, the remaining two, one well differentiated and one moderately differentiated adenocarcinoma, showed soft tissue density nodules
accompanied by ill dened margins, the latter
correlated with lepidic growth pattern in a small
portion of the tumour periphery on pathological
specimens.
On the other hand, a lepidic growth pattern is
characterized by tumour cell growth replacing
normal alveolar lining cells [9]. A heterogeneous
hazy density with ill dened opacity may be seen
on HRCT and is characteristic of well to
moderately differentiated adenocarcinoma [11,
12]. This type of lung cancer was regarded as
very slow growing [12] and has the most favorable
prognosis among adenocarcinomas in general
[11]. According to Noguchi et al [11], the 5-year
survival rate of 28 cases of this type of cancer was
100%. It should be noted that three localized
bronchioloalveolar carcinoma in this study presented as rapidly growing. Our present study
indicated that this type of adenocarcinoma might
The British Journal of Radiology, September 2000

CT features of rapidly growing small lung cancers

(a)

(b)

(c)

Figure 3. Case 11: a moderately differentiated adenocarcinoma in the right lower lobe in a 74-year-old
male patient. (a) Initial screening CT (left) shows a
7 mm Z 10 mm heterogeneously low density nodule
(arrow). The re-screening CT image 6 months later
(right) demonstrates a soft tissue density nodule with
an increasing size and density. Cancer was highly suspected. (b) Pre-operative thin section CT image shows
a soft tissue density nodule with irregular shape, ne
spiculation and a pleural tag (arrow). The tumour
size of 9 mm Z 14 mm was measured on thin section
CT image. (c) Pathological specimen (Z1.25) shows
moderately differentiated adenocarcinoma with pleural
reactive thickening and irregular margin. No air
bronchogram pattern or small air spaces are found in
the tumour.

grow rapidly, with a possibility of unfavorable


outcome.
Kuriyama et al [13] reported HRCT ndings of
20 small lung cancers (,20 mm); 13 (72%) of 18
adenocarcinomas were reported to exhibit an air
bronchogram pattern on CT images. They
suggested that the air bronchogram pattern in a
lung nodule was helpful in discriminating adenocarcinomas from benign lesions. Other groups
found that an air bronchogram also appeared on
CT images in 28.766.7% of lung cancers of
different sizes and pathological types [14, 15]. In
eight cases of adenocarcinoma in this study, we
found only one adenocarcinoma with an air
bronchogram on CT images. We also found
only one case having small air spaces in rapidly
growing lung cancers, although in one study [16]
this pattern was reported in 39% of CT images of
small lung cancers. Our study demonstrated that
most of the rapidly growing small lung cancers
lacked an air bronchogram or small air spaces in
the tumour on CT images. Thus, lung nodules

that do not exhibit an air bronchogram or small


air spaces are likely to be rapidly growing
tumours and this indicates the need for early
diagnosis and treatment.
The CT features of small nodules of SCLCs
have not been adequately described. In this study,
we encountered three cases of small nodule SCLC
and we were able to study their interval changes.
They appeared as well dened, homogeneous, soft
tissue density nodules with smooth margins and a
lobulated conguration on CT images. The VDTs
of these tumours were 54, 60 and 132 days,
respectively (Table 1). According to Quoix et al
[17], less than 5% of cases of SCLC present as
solitary pulmonary nodules. This type of SCLC
has been regarded as a biologically distinct subset
characterized by relatively slow growth and which
is potentially curable by surgery [17, 18].
According to our calculation, using the formula
of Schwartz [5], the VDTs of three cases of
peripheral SCLC reported by Urschel [18] were
180, 210 and 360 days. Our cases had a noticeably

The British Journal of Radiology, September 2000

935

J-C Wang, S Sone, L Feng et al

(a)

(c)

shorter VDTs than those in previous studies, and


were similar to those of the central SCLCs
[19].
There are several limitations in this study. First,
the study included a relatively small sample size.
Thus, further studies using a larger number of
cases are necessary to dene the spectrum of CT
and pathological ndings of rapidly growing lung
cancers. The other limitation was measurement of
initial tumour size on low dose spiral CT images
rather than on HRCT images. To avoid an
overestimation of tumour VDT, we dened the
minimal size of small lung cancer on screening CT
images as 3 mm when they were invisible or
smaller than 3 mm in the initial screening CT
image. However, we believe that the measurements performed in this study were more accurate
than those reported in previous studies in which
tumour VDT was measured on conventional chest
radiographs.
936

(b)

Figure 4. Case 10: a well differentiated adenocarcinoma in the right lower lobe in a 70-year-old female
patient. (a) Initial screening CT scan (left) shows a
5 mm Z 8 mm irregular opacity in the right lower
lobe. Annual re-screening CT image 1 year later
(right) shows an increase in size of the same nodule,
with heterogeneous density. (b) Pre-operative thin section CT shows a lobulated, soft tissue density nodule
with an air bronchogram pattern. The margin of the
nodule is irregular and partly ill dened. The tumour
size is measured as 9 mm Z 17 mm. (c) Pathological
specimen (Z1.25) shows that the lesion is well differentiated adenocarcinoma with an irregular margin
and air bronchogram pattern. Tumour shows replacement of the growth pattern with active broblastic
proliferation. Some remnant small air spaces are
found in the tumour.

In conclusion, in the present study we described


the characteristics of rapidly growing small lung
cancers as they appear on thin section CT images.
Peripheral small lung cancers of rapid growth
mostly were adenocarcinomas; they also included
SCLCs and SCCs. However, as most peripheral
lung cancers were adenocarcinomas in our repeat
CT screening study, the proportion of rapidly
growing lung cancer among adenocarcinomas is
lower than that of SCLC and SCC. CT features of
these included a soft tissue density tumour that
lacked an air bronchogram pattern or small air
spaces in the nodules. We also showed the rare
occurrence of rapid growth in cases of well
differentiated adenocarcinomas, which are characterized by lepidic tumour growth with GGO on
CT images and which have otherwise been
reported in the past to show a slow growth. A
long-term study of patients treated surgically is
needed to correlate the growth rate of small lung
The British Journal of Radiology, September 2000

CT features of rapidly growing small lung cancers

cancers with prognosis and to establish the effects


of early diagnosis and surgical treatment of lung
cancer on prognosis.

Acknowledgments
We thank Kazuhisa Hanamura, BS, and
Kazuhiro Asakura, EE, from the Telecommunications Advancement Organization of
Japan Matsumoto Research Center for their
contribution to this study.

References
1. Usuda K, Saito Y, Sagawa M, Sato M, Kanma K,
Takahashi S, et al. Tumor doubling time and
prognostic assessment of patients with primary lung
cancer. Cancer 1994;74:223944.
2. Arai T, Kuroishi T, Saito Y, Kurita Y, Naruke T,
Kaneko M. Tumor doubling time and prognosis in
lung cancer patients: evaluation from chest lms
and clinical follow-up study. Japanese Lung Cancer
Screening Research Group. Jpn J Clin Oncol
1994;24:199204.
3. Geddes DM. The natural history of lung cancer: a
review based on rates of tumour growth. Br J Dis
Chest 1979;73:117.
4. Kerr KM, Lamb D. Actual growth rate and tumour
cell proliferation in human pulmonary neoplasms.
Br J Cancer 1984;50:3439.
5. Schwartz M. A biomathematical approach to
clinical tumor growth. Cancer 1961;14:127294.
6. Steele JD, Buell P. Asymptomatic solitary pulmonary nodules, host survival, tumor size, and growth
rate. J Thorac Cardiovasc Surg 1973;65:14054.
7. Spratt JS, Meyer JS, Spratt JA. Rate of growth of
human neoplasms: Part II. J Surg Oncol 1996;61:
6883.
8. Hayabuchi N, Russell WJ, Murakami J. Slowgrowing lung cancer in a xed population sample.
Radiological assessments. Cancer 1983;52:1098104.

The British Journal of Radiology, September 2000

9. Theros EG. Varying manifestations of peripheral


pulmonary neoplasms: a radiologicpathologic correlative study. AJR 1977;128:893914.
10. Sone S, Sakai F, Takashima S, Honda T, Yamanda
T, Kubo K, et al. Factors affecting the radiologic
appearance of peripheral bronchogenic carcinomas.
J Thorac Imaging 1997;12:15972.
11. Noguchi H, Morikawa A, Kawasaki M, Matsuno
Y, Yamada T, Hirohashi S, et al. Small adenocarcinoma of the lung: histologic characteristics and
prognosis. Cancer 1995;75:284452.
12. Kuriyama K, Seto M, Kasugai T, Higashiyama M,
Kido S, Sawai Y, et al. Ground-glass opacity on
thin-section CT: value in differentiating subtypes of
adenocarcinoma of the lung. AJR 1999;173:
4659.
13. Kuriyama K, Tateishi R, Doi O, Higashiyama M,
Kodama K, Inoue E, et al. Prevalence of air
bronchograms in small peripheral carcinoma of the
lung on thin-section CT: comparison with benign
tumors. AJR 1991;156:9214.
14. Gaeta M, Pandolfo I, Volta S, Russi EG,
Bartiromo G, Girone G, et al. Bronchus sign on
CT in peripheral carcinoma of the lung: value in
predicting results of transbronchial biopsy. AJR
1991;157:11815.
15. Ma K, Templeton PA, White CS, Cai Z-L, Bai
Y-X, Cai Y-Q. Evaluation of the air bronchogram sign on CT in solitary pulmonary lesions. J
Comput Assist Tomogr 1996;20:9836.
16. Kuriyama K, Tateishi R, Doi O, Kodama K,
Tatsuta M, Matsuda M, et al. CTpathological
correlation in small peripheral lung cancers. AJR
1987;149:113943.
17. Quoix E, Fraser R, Wolkove N, Finkelstein H,
Kreisman H. Small cell lung cancer presenting as a
solitary pulmonary nodule. Cancer 1990;66:57782.
18. Urschel JD. Pretreatment natural history of small
cell lung cancer presenting as a solitary pulmonary
nodule. J Cardiovasc Surg 1994;35:2735.
19. Brigham BA, Bunn PA Jr, Minna JD, Cohen MH,
Ihde DC, Shackney SE. Growth rate of small cell
bronchogenic carcinomas. Cancer 1978;42:28806.

937

Potrebbero piacerti anche