Ultrasonographic Evaluation of Pulmonary Consolidation 1, 2

PAN-CHYR YANG, KWEN-TAY LUH, DUN-BING CHANG, CHONG-JEN YU,

SOW-HSONG KUO, and HUEY-DONG WU

Introduction

Most
physicians believe
that
ultrasonography has limited usage in
chest diseases. However, since the introduction of the concept of the "ultrasonic window" (1), the applications of
chest ultrasound have been greatly extended. Chest ultrasound is not only useful for chest-wall or pleural-based lesions
(2-4), it is also very useful for evaluation
of and biopsy guidance of mediastinal
and peripheral pulmonary nodules (1,
5-7). In pulmonary consolidation, the
fluid-containing air spaces become a
good ultrasonic window. Lesions deep
seated in the consolidated lung can be
visualized by ultrasound (8). Fewreports
on the ultrasonic descriptions of pulmonary consolidation are available (9, 10).
We have found that high-resolution realtime ultrasound in conjunction with
Doppler ultrasound can be very useful
in the evaluation of pulmonary consolidation. Not only the air or fluid bronchogram, but also the abscess cavity of
necrotizing pneumonia and associated
tumors in the consolidated lung can be
detected by ultrasound (8, 11). We present our experience with chest ultrasound
in the evaluation of 161 cases of pulmonary consolidation. The ultrasonographic features and technical aspects are also
discussed.
Methods
One hundred seventy consecutive patients
with lobar or segmental consolidations were
examined by chest ultrasound from July 1988
to July 1990. Patients with (1) radiographic
evidenceof pneumonic consolidation that extended to the visceral pleura (ultrasonic window) and (2) an unsatisfactory response to
antibiotic treatment for at least 1 wk were referred for ultrasonographic examinations.
There werenine patients whose consolidation
was not complete or not in contact with the
pleura to a large extent. The ultrasound examination in these patients was not successful. A total of 161 patients were evaluable;
113 were men and 48 were women, ranging
in age from 18 to 87 yr.
Forty-five patients also underwent chest
computerized tomography (CT scan) exami-

SUMMARY A total of 161 patients with lobar or segmental consolidation were examined by realtime ultrasound and Doppler ultrasound. Air bronchograms were detected In 141 patients, fluid bronchograms In 27 patients, and parapneumonic effusion In 74patients. In 36 patients with necrotizing
pneumonia, ultrasound detected mlcroabscesses in 33 (91.7%) compared with the air-fluid levels
detected by standard chest radiographs in 20 patients (55.6%; p < 0.05). Of 31 patients with tumors
causing obstructive pneumonitis, 29 (93.5%) had tumors detected by chest ultrasound, whereas
only 11patients (35.5%) had chest radiographs that suggested a tumor was causing the obstructive
pneumonitis (p < 0.05). Chest ultrasound was used to guide thoracentesis for parapneumonlc effusion In 65 patients, with a 100% success rate. Twenty-six patients with necrotizing pneumonia underwent ultrasound-guided needle aspiration of mlcroabscesses. The procedure was successful
in 24 patients (92.3%), and 21 patients (80.8%) had microbiologic confirmation. lWenty patients
with tumor-associated obstructive pneumonitis received needle aspiration biopsy under ultrasound
guidance; 19 patients (95.0%) had the histology confirmed. Five patients with malignancy manifesting as pulmonary consolidation underwent a diagnostic ultrasound-guided needle aspiration biopsy. Five patients (3.8%) developed complications of minimal pneumothorax or mild hemoptysis in
132 episodes of needle aspiration. We conclude that ultrasonography is useful for the evaluation
of pulmonary consolidation. It can also be used for needle aspiration guidance for etiologic diagnosis of patients with complicated pneumonia.
AM REV RESPIR DIS 1992; 146:757-762

nations (Somatom DR; SiemensMedical System, Iselin, NJ), and 58 patients underwent
fiberoptic bronchoscopy with or without biopsy.The final diagnoses of these 161 patients
were based on the results of the CT scan,
fiberoptic bronchoscopy, trans bronchial biopsy, percutaneous transthoracic aspiration
biopsy, and clinical follow-up.
Chest radiographs of all patients performed
during the same period of time as the ultrasound examination wereinterpreted by a radiologist and a chest physician without their
knowing the results of the bronchoscopic examination, CT scan, or chest ultrasound. Special note was made of the radiographic evidences of air bronchograms, microabscess
formation, tumor, and parapneumonic
effusion.
All patients underwent ultrasound examinations without prior knowledge of the bronchoscopy, CT scan, or chest-radiograph
results. The physicans who performed the
ultrasound examination had only information on the consolidated area present in the
chest radiographs to find the ultrasonic window. The patients were examined with realtime, linear-array, convexand sector ultrasonic units with a 3.5-MHz or a 5.0-MHz transducer (Aloka SSD 630 and Toshiba 100A;
Tokyo, Japan). The Toshiba l00A ultrasonic
units were also equipped with Doppler ultrasound, which can be used in conjunction with
sector units to evaluate blood flow. Patients
were scanned in a supine or prone position

by means of an intercostal approach. The

sonographic images were recorded on Polaroid film (Polaroid, Cambridge, MA) and analyzed for the presence of air bronchograms,
fluid bronchograms, microabscesses, parapneumonic effusions, or tumors in the consolidated lung.
The consolidated lung was defined as an
isoechoic or hypoechoic area with a triangular shape that moved with respiration. A fluid bronchogram was defined as a branching,
hyperechoic, tubular structure with an anechoic inside lumen and with no blood flow
detected by Doppler ultrasound (8). An air
bronchogram was identified by bifurcating
hyperechoiclines arising from the hilum. The
hyperechoic air densities could move with
respiration. A microabscess was defined as
an irregular hypoechoic or anechoic cystic
space present in a homogeneous consolidated lung (11). The hyperechoic air densities
(Received in original form March 12, 1991 and in
revised form January 29, 1992)
1 From the Departments of Internal Medicine
and Clinical Pathology, National Taiwan University Hospital, and Institute of Biomedical Sciences,
Academia Sinica, Taipei, Taiwan, Republic of
China.
2 Correspondence and requests for reprints
should be addressed to Pan-Chyr Yang, M.D., Ph.D.,
National Taiwan University Hospital, No.1, ChangTe Street, Taipei, Taiwan 10016,Republic of China.

757

758

YANG, WH, CHANG, YU, KUO, AND WU

could also be present inside the cavity. A tumor causing obstructive pneumonitis was a
well-defined,homogeneous, hypoechoic nodule or mass inside a consolidated lung, located near the hilum (8). Parapneumonic effusion showed as an echo-free space between
the visceralpleura and the parietal pleura that
could change shape with the respiratory
movement.
The echogenicity of the pulmonary consolidation, tumor, or microabscess was compared with the echogenicity of the liver and
defined as hypoechoic, isoechoic, or hyperechoic. The echogenicity of gallbladder was
used as a reference for anechoic lesions. The
sonographic results were compared with the
results of chest radiographs. The data were
analyzed statistically with the X! test.
After assessment of the ultrasonographic
features of the pulmonary consolidation,
those patients without a conclusivehistologic
diagnosis attained by conventional methods,
which included sputum and bronchoscopic
examinations underwent (after informed consent was obtained) ultrasound-guided needle
aspiration of the parapneumonic effusion,
microabscess, tumor, or even consolidated
lung parenchyma. The aspirated materials
were sent for bacteriologic, cytologic, and
histologic examination. A 22-gauge needle
with an outer sheath and an inner stylet was
used for aspiration. The ultrasound-guided
needle aspiration technique is detailed in
previous studies (1, 7, 8). The needle aspiration route was prechosen by the ultrasound
image to avoid passing through the bronchi
or major vessels. Repeated aspirations were
performed immediately if the aspirated material was inadequate for bacterial, cytologic, or histologic studies (8, 11). A routine chest
radiograph was taken on the day after the
needle aspiration to assess potential
complications.
Results

A total of 161 patients were included in

this study. Underlying diseases included
diabetes mellitus (21 patients), malignancy (18patients), cerebral vascular disease
(12 patients), collagen diseases (10 patients), congestive heart failure (eight patients), and alcoholism (four patients).
Eighty-eight patients did not have any underlying diseases. The localization ofthe
pulmonary consolidation, based on the
chest radiographs, showed that 33 patients had consolidation at the right upper lobe, 59 at the right lower lobe, nine
at the right middle lobe, 23 at the left
upper lobe, and 27 at the left lower lobe,
and 10 patients had consolidation with
multiple lobular involvement. The final
diagnoses of the patients are shown in
table 1. There were 89 patients with simple pneumonia who eventually recovered
after antibiotic treatment; 36patients had
necrotizing pneumonia with microab-

TABLE 1
CLINICAL DIAGNOSIS AND RESULTS OF ULTRASOUND-GUIDED
NEEDLE ASPIRATION IN 161 PATIENTS
WITH PULMONARY CONSOLIDATION

Clinical Diagnosis
Simple pneumonia
Necrotizing pneumonia
Tumor with obstructive
pneumonia
Malignancy
Parapneumonic effusion

Aspiration
Successful

Diagnosis
Confirmed
n

No. of
Cases

Aspiration
Done

89
36

16
26

12
24

75.0
92.3

6*
21

37.5
80.8

31
5
74

20
5
65

20
5
65

100.0
100.0
100.0

19
5

95.0
100.0

* Three patients had a Mycobacterium tuberculosis infection, one a cryptococcal infection, one a
Klebsiella pneumoniae pneumonia, and one a Pseudomonas aeruginosa pneumonia.

scess formation, 31 patients had tumors

with obstructive pneumonitis, and five
patients had cancer infiltration (three
with bronchioloalveolar carcinoma, one
with metastatic pancreatic carcinoma,
and one with lymphoma).

Ultrasonographic Features oj
Pulmonary Consolidation
The consolidated lung was triangular in
shape, hypoechoic, and heterogeneous in
echotexture. The air bronchograms
presented as arborizing hyperechoic lines
arising from the hilar region. The scattered hyperechoic linear echoes could
change shape with respiration (figure 1).
The fluid bronchograms showed branching, hyperechoic, parallel lines or tubular structures. The fluid bronchograms
usually paralleled the pulmonary vessels.
Doppler ultrasound was excellent in
demonstrating blood flow present in the
pulmonary vasculature (figure 2). The

vascular linear echoes could also be

traced to the hilar region, which may join
the pulmonary artery. Blood flow was absent in the fluid bronchograms. Fluid
bronchograms werepresent exclusively in
patients with obstructive pneumonitis.
None of the patients with simple pneumonia had fluid bronchograms. The
microabscesses in necrotizing pneumonia revealed a round hypoechoic to
anechoic lesion with an ill-defined margin within the consolidated lung (figure
3). The size of the microabscess cavity
varied from 0.5 to 3 em in diameter. Some
of the microabscesses had hyperechoic
speckled densities at their periphery,
representing air echoes inside the abscess
cavity. Lung tumors causing obstructive
consolidation showed as homogeneous
hypo- or hyperechoic nodules with welldefined margin (figure 4).
Table 2 summarizes the ultrasonographic features of these 161 patients with

Fig. 1. A 20-yr-old man with pulmonary

consolidation at the lingular segment.
Convex ultrasound shows a triangularshaped consolidated lesion with bifurcating hyperechoic lines of air bronchogram (arf'01lVheads). There is minimal
parapneumonic effusion (E).

759

ULTRASONOGRAPHY IN PULMONARY CONSOLIDATION

Fig. 2. (A) Chest radiograph of a 50-yr-old man shows

left upper lobe consolidation. (B) Convex ultrasound
shows a triangular-shape-consolidation with an irregular shaped hypoechoic tumor (T) near the hilar region.
Two parallel tubular structures (a, b) are seen in the consolidated lung parenchyma. (C) Doppler ultrasound of
the tubular structure (a) shows blood flow inside, indicating that this tubular structure is a pulmonary vessel.
(0) Doppler ultrasound of the tubular structure (b) shows
no blood flow inside, indicating that this tubular structure is a fluid bronchogram. (E) A CT scan with contrast
study done one week before the left upper lobe atelectasis shows a tumor located near the hilum (arrow). Arrowhead indicates a pulmonary vessel demonstrated
by ultrasound.

pulmonary consolidations in comparison

with their radiographic findings. An air
bronchogram was detected by ultrasound
in 141 patients, but by chest radiography
in only 106 patients. In 31 patients with
a tumor causing obstructive pneumonitis, the tumor wasdetected by chest ultrasound in 29 patients. In two patients with
obstructing tumors undetected by ultrasound, one tumor was confirmed by
fiberoptic bronchoscopic examination
and proven to be endobronchial papilloma, and the other, a 3-cm hilar lymph
node causing obstruction was demonstrated by CT scan. In no patients did
definite tumor shadows appear in the
consolidated lung on the chest radiographs, and only 11 of the patients had
radiographic "reverseS sign," suggesting
the presence of a tumor causing obstruction. Twenty-sevenpatients (87.1 070) had
ultrasound-demonstrable fluid bronchograms not detectable by chest radiographs. Of 36 patients with necrotizing
pneumonia, 20 had an air-fluid level on

their chest radiographs; the remaining 16

patients (without an air-fluid level) had
the necrotic cavity confirmed by CT scan.
Thirty-three patients (91.7070) had a
necrotic cavity demonstrated by ultrasound examination. In three patients in
whom ultrasound did not detect microabscessformation, CT scan and chest radiographs demonstrated necrotizing cavities
on the second week after ultrasound examination. The parapneumonic effusions were evident by ultrasound in 74
patients but evident by chest radiography in only 42 of them. Forty-twoofthese
effusions were minimal in amount and
did not cross an intercostal space, or the
echo-free space between the parietal
pleura and the visceral pleura of the consolidated lung was 1em or less (figure 5).

Ultrasound-guided Aspiration of
Pulmonary Consolidation
Of 74 patients with parapneumonic effusion, nine had pleural effusions that
wereso evident on the chest radiographs

that a thoracentesis was performed at the

bedside. The other 65 patients underwent
ultrasound-guided needle aspiration of
the effusion for etiologic diagnosis of
pneumonia. The results of the ultrasound-guided needleaspiration are shown
in table 1. Adequate fluid (> 2 mL) was
obtained from all 65 patients for cytologic and bacteriologic examination. Of
36 patients with necrotizing pneumonia,
26 underwent ultrasound-guided needle
aspiration from the microabscess cavities. The amount of material obtained
varied from 0.1to 3 mL. In two patients,
aspiration failed to obtain adequate specimens. The bacteriologic diagnosis was
thus confirmed in 21 patients (80.8%).
In three patients, although purulent fluid was obtained by needle aspiration (0.1
mL, 0.3 mL, 0.5 mL, respectively), the
bacterial cultures failed to grow any organisms. All 26 patients had undergone
empirical antibiotic treatment before needle aspiration. A total of 26 organisms
wereobtained from the abscess cultures.

760

YANG, WH, CHANG, YU, KUO, AND WU

Fig. 3. (A and B) Chest radiographs of

a 62-yr-old women suffering stroke show
segmental consol idation at anterior segment of the right upper lobe. (C)CT scan
of the chest shows a necrotizing pneumonia with a cavity inside the consolidated lung. (D) Convex ultrasound reveals a triangular-shaped consolidated
lesion with a round hypoechoic cavity
(arrowheads) inside the lung parenchyma. There are hyperechoic densities at
its periphery. Ultrasound-guided aspiration from the cavity retrieved some pus;
bacterial cultures grew Klebsiella
pneumoniae.

Fig. 4. (A) A 75-yr-old man with obstructive pneumonia at the left lower lobe. (B)
Chest ultrasound shows an irregular
hypoechoic tumor (T) inside the consolidated lung. There is a minimal amount
of parapneumonic effusion (E). The effusion cytology is negative for malignancy; however, needle aspiration biopsy
from the tumor shows squamous cell
carcinoma.

761

ULTRASONOGRAPHY IN PULMONARY CONSOLIDATION

TABLE 2

TABLE 3

COMPARISON OF THE MORPHOLOGIES OF PULMONARY CONSOLIDATION DETECTED

BY ULTRASONOGRAPHY AND CHEST RADIOGRAPHY IN 161 PATIENTS

MICROORGANISMS ISOLATED FROM

THE TRANSTHORACIC ASPIRATION
OF 21 PATIENTS WITH
NECROTIZING PNEUMONIA

Morphology

Ultrasound

Air bronchogram
Fluid bronchogram
Parapneumonic effusion
Tumor with obstructive pneumonitis
(n = 31)
Necrotizing pneumonia with cavity
(n = 36)

141
27
74

Chest Radiograph

p Value*

106
42

29 (93.5%)

11 (35.5%)

< 0.05

33 (91.7%)

20 (55.6%)

< 0.05

Analyzed by chi-square test.

Table 3 shows the microorganisms isolated from the aspirates. Twelve patients
had their treatment regimens modified
based on the culture results, and the lesions improved after an initial lack of response to antibiotics.
There were 29 patients who had
demonstrable tumors in the consolidated lung. All underwent a fiberoptic bronchoscopic examination. Nine patients
had histologic diagnoses 0 btained by
bronchoscopy with biopsy, 10 patients

Fig. 5. (A) A 32-yr-old man with a minimal volume of parapneumonic effusion,

which does not extend across one intercostal space. (B) Another 40-yr-old man
has minimal parapneumonic effusion.
The echo-free space between the parietal and visceral pleura is less than
1 cm in depth. These minimal effusions
are difficult to aspirate without ultrasound guidance. (0 = diaphragm; L =
lung; Pp = parietal pleura; Pv = visceral pleura).

No. of
Isolates

Microorganism

had brushing cytology positive for malignancy, and 10 were negative. The 10 patients whose results werenegative and the
10patients who had only cytologic diagnoses made by bronchoscopic brushing
underwent an ultrasound-guided needle
aspiration biopsy for histologic diagnosis. Definite histologic diagnoses were
made in 19 cases (95.00/0). The material
aspirated from one patient yielded only
necrotic tissue. This patient was proven
to have a squamous cell carcinoma by a

Klebsiella pneumoniae
Pseudomonas aeruginosa
Staphylococcus aureus
Streptococcus viridans
Enterobacter cloacae
Haemophilus influenzae
Bacteroides melaninogenicus
Bacteroides fragilis
Peptostreptococcus
Microaerophilic streptococci
Aspergillus fumigatus
Total

5
3
2

2
1
1
1

2
2

5
2
26

supraclavicular lymph-node biopsy during the follow-up period.

There were 16patients with pneumonic
consolidation unresponsive to antibiotic treatment and with a nondiagnostic
bronchoscopic examination. They underwent an ultrasound-guided needle aspiration biopsy from the consolidated lung.
Twelve patients (75.00/0) underwent a successful diagnostic aspiration with material appropriate for microbiologic study,
three patients were proven to have
Mycobacterium tuberculosis infection by
acid-fast stain, one patient was proven
to have cryptococcal infection by India
ink stain and culture, one culture grew
Klebsiella pneumoniae, and another grew
Pseudomonasaeruginosa. The diagnostic yield was 37.5%. The other 73 patients
who refused needle aspiration weretreated empirically with antibiotics, and their
pneumonic lesions eventually resolved.
Of fivepatients who had pulmonary consolidation due to malignancy, three were
proven to have bronchioloalveolar carcinoma, one to have metastatic pancreatic
cancer, and one to have lymphoma (by
needle aspiration biopsy).
Out of a total of 132 cases of ultrasound-guided needle aspirations, fivepatients had complications (3.8%). Three
patients experienced minimal pneumothorax and two patients had mild hemoptysis. All complications subsided
spontaneously. The complications occurred in patients undergoing needle
aspiration of the lung parenchyma. None
occurred in patients who underwent thoracentesis for parapneumonic effusions.
Discussion

The etiology of pulmonary consolidation

is diverse. Although infectious pneumo-

762

nia is the most common cause, other conditions such as cancer infiltration or
tumors with obstructive pneumonia may
manifest as pulmonary consolidation
(12-14). Conventional chest radiographs
offer little information concerning the
underlying etiology or morphologic details of pulmonary consolidation (15).
Chest radiography may not be sensitive
enough to detect minimal effusion, necrotizing pneumonia, or tumors with obstructive pneumonitis. CT scans with
contrast study are superb in demonstrating the morphologic changes of pulmonary consolidation (16); however, a routine chest CT scan for pulmonary consolidation may be not cost effective. In
this study, we demonstrated that chest
ultrasound can be useful in evaluating
pulmonary consolidation. Parapneumonic effusions, necrotizing pneumonias, or tumors with obstructive pneumonitis can be clearly identified by chest
ultrasound. The sensitivity of chest ultrasound in demonstrating necrotizing pneumonia, parapneumonic effusion, and tumor with obstructive pneumonitis may
be higher than conventional chest radiographs, but this has not been confirmed
by CT scan.
Our results also showed that chest
ultrasound can be used for thoracentesis
guidance of parapneumonic effusions.
Though most effusions in this study were
minimal, the success rate of ultrasoundguided thoracentesis was 100070. Chest
ultrasound can also be used as a guide
in the needle aspiration of microabscesses
for microbiologic study in patients with
necrotizing pneumonia. Successful aspiration was achieved in 92.3070 of our patients. Microbiologic diagnoses wereconfirmed in 80.8070, and 46.1070 of the patients had their antibiotic treatments
modified after needle aspiration. This
technique provides a reliable etiologic diagnosis for necrotizing pneumonia. It
would be particularly useful for the immunocompromised host with necrotizing pneumonia, in whom an accurate diagnosis is especially important.
Ultrasound-guided transthoracic needle aspiration through the consolidated
lung is relatively safe. The major bronchi and pulmonary vessels can be clearly
differentiated by high-resolution realtime ultrasound and Doppler ultrasound.
The route for needle aspiration can be
preselected to avoid these vital structures,
and thus the complication of major
bleeding can be minimized. Ultrasound
can select an area where the consolida-

YANG, WH, CHANG, YU, KUO, AND WU

tion is complete and where there is no

air-containing tissue interposed between
the lesion and the pleura. This may be
the main reason for the low complication rate in this series. Only five patients
(3.8070) in our study experienced minor
complications after transthoracic needle
aspiration, although a high incidence (10
to 50070) of pneumothorax in fluoroscopy or CT-guided transthoracic needle
aspiration has been reported in the literature (17, 18). However, because the case
selection may be different, it is difficult
to compare our results with those of
others.
For those patients with tumors causing obstructive pneumonitis, a fluid
bronchogram is a very useful sign of
bronchial obstruction. The fluid bronchogram rarely presents in cases of consolidation other than obstructive pneumonia. If a diagnosis cannot be made
by fiberoptic bronchoscopy, ultrasound
can be used as a guide for transthoracic
needle aspiration biopsy for histologic
study. In our study, a high diagnostic rate
of 95.0070 was attained. The yields for
microbiologic diagnosis of patients with
simple pneumonia were relatively low in
our study. In 16 patients receiving transthoracic needle aspiration, sufficient material for microbiologic study was obtained in 12,and only sixofthem (37.5070)
had their diagnoses confirmed. The low
diagnostic yield is probably due to the
fact that most ofthese patients had been
treated with antibiotics before needle
aspiration and that only scanty material
could be obtained by fine-needle aspiration with a 22-gauge needle.
Because only a limited number of patients in this study underwent a CT scan
examination, we were not able to compare the diagnostic sensitivity and specificity of ultrasound and the CT scan.
However, in a previous study, we found
that chest ultrasound was sensitive in the
demonstration of tumors causing obstructive pneumonia (8). In this study,
there weretwo patients with tumors causing obstructive pneumonia that were not
visualized by chest ultrasonography. One
of the tumors was detected by fiberoptic
bronchoscopy and one by chest CT scan.
Although ultrasound has proved useful
in the detection of and in thoracentesis
guidance to minimal parapneumonic effusions, in this study we were also unable to compare the sensitivity and specificity of ultrasound examinations with
conventional fluoroscopy in the diagnosis of pleural effusions.

In conclusion, chest ultrasonography

is a very useful tool in the evaluation of
pulmonary consolidation. Complex etiologies can be demonstrated by chest
ultrasound. Chest ultrasound can also
provide guidance for transthoracic needle aspiration for etiologic diagnosis of
patients with complicated pneumonia.
References
1. Yang PC, Luh KT, Sheu JC, Kuo SH, Yang SP.
Peripheral pulmonary lesions: ultrasonography and
ultrasonically guided aspiration biopsy. Radiology 1985; 155:451-6.
2. Saito T, Kobayashi H, Kitamura S. Ultrasonographic approach to diagnosing chest wall tumors.
Chest 1988; 94:1271-5.
3. Hirsch JH, Rogers JV, Mack LA. Real-time
sonography of pleural opacities. AJR 1981;
136:297-301.
4. Yang PC, Sheu JC, Luh KT, Yang SP. Clinical
application of real- time ultrasonography in pleural
and subpleural lesions. J Formosan Med Assoc
1984; 83:646-57.
5. Chen CR, Lee LN, Kuo SH, Luh KT. Ultrasonically guided percutaneous mediastinal cytology and biopsy. J Formosan Med Assoc 1986;
85:561-6.
6. Pedersen OM, Aasen T, Gulsvik A. Fine needle aspiration biopsy of mediastinal and peripheral pulmonary masses guided by real-time sonography. Chest 1986; 89:504-7.
7. Yang PC, Lee LN, Luh KT, Kuo SH, Yang SP.
Ultrasonography of Pancoast tumor. Chest 1988;
94:124-8.
8. Yang PC, Luh KT, Wu HD, et al. Lung tumors
associated with obstructive pneumonitis: US
studies. Radiology 1990; 174:717-20.
9. Dome HL. Differentiation of pulmonary parenchymal consolidation from pleural disease using
the sonographic fluid bronchogram. Radiology
1986; 158:41-2.
10. Weinberg B, Diakoumakis EE, Kass EG, Seife
B, Zvi ZB: The air bronchogram: sonographic
demonstration. AJR 1986; 147:593-5.
11. Yang PC, Luh KT, Lee YC, et al. Lung abscess: ultrasonography and ultrasound-guided transthoracic aspiration. Radiology 1991; 180:171-5.
12. Epstein DM, Gofter WB, Miller WT. Lobar
bronchioloalveolar cell carcinoma. AJR 1982;
139:463-8.
13. Burke M, Fraser R. Obstructive pneumonitis: a pathologic and pathogenetic reappraisal. Radiology 1988; 166:699-704.
14. Shure D, Astarita RW. Bronchogenic carcinoma presenting as an endo-bronchial mass: optimal
number of biopsy specimens for diagnosis. Chest
1983; 83:865-7.
15. TewJ, Calenoff L, Berlin BS. Bacterial or nonbacterial pneumonia: accuracy of radiographic diagnosis. Radiology 1977; 124:607-12.
16. Baber CE, Hedlund LW, Oddson TA, Putman
CEo Differentiating empyemas and peripheral pulmonary abscesses: the value of computed tomography. Radiology 1980; 135:755-8.
17. Perlmutt LM, Johnston WW, Dunnick NR.
Percutaneous transthoracic needle aspiration: a review. AJR 1989; 152:451-5.
18. Westcott JL: Percutaneous transthoracic needle biopsy. Radiology 1988; 169:593-601.