Benign and Malignant Lesions
Radiology
Erik K. Insko, MD, PhD
Marc S. Levine, MD
Bernard A. Birnbaum, MD2
Jill E. Jacobs, MD2
Index terms:
Gastritis, 72.291
Gastrointestinal tract, CT, 72.12112,
72.12115
Gastrointestinal tract, radiography,
72.123
Stomach, CT, 72.12112, 72.12115
Stomach, neoplasms, 72.31, 72.32
Published online before print
10.1148/radiol.2281020623
Radiology 2003; 228:166 –171
Abbreviation:
UGI ⫽ upper gastrointestinal
1
From the Department of Radiology,
Hospital of the University of Pennsyl-
vania, 3400 Spruce St, Philadelphia,
PA 19104. Received May 28, 2002;
revision requested July 26; revision re-
ceived September 19; accepted No-
vember 18. Address correspondence
to E.K.I. (e-mail: insko@rad.upenn.
edu).
Current address:
2
Department of Radiology, New York
University Medical Center, New York,
NY.
Author contributions:
Guarantor of integrity of entire study,
E.K.I.; study concepts and design, all
authors; literature research, E.K.I.; clin-
ical studies, B.A.B., J.E.J.; data acquisi-
tion, all authors; data analysis/inter-
pretation, E.K.I., M.S.L.; statistical
analysis, E.K.I.; manuscript prepara-
tion, definition of intellectual content,
and editing, E.K.I., M.S.L.; manuscript
revision/review and final version ap-
proval, all authors.
© RSNA, 2003
166
of the Stomach: Evaluation
of CT Criteria for
Differentiation1
PURPOSE: To determine the sensitivity and specificity of computed tomographic
(CT) criteria for differentiating benign from malignant stomach lesions in patients
with a thickened gastric wall at CT.
MATERIALS AND METHODS: A radiology department file search revealed 36
patients with a thickened gastric wall at CT who underwent double-contrast barium
suspension upper gastrointestinal tract examinations within 6 weeks before or after
CT. The authors reviewed the CT images without knowledge of the final radiologic,
endoscopic, or pathologic findings to determine the degree of gastric wall thicken-
ing and the symmetry, distribution, and enhancement of the thickened wall. The
sensitivity and specificity of these findings for detection of malignancy were calcu-
lated.
RESULTS: Two of 36 patients had two gastric abnormalities each. The final diag-
noses in the 38 cases were gastritis in 19, hiatal hernia in four, benign ulcer in three,
benign (n ⫽ 3) or malignant (n ⫽ 8) gastric neoplasm in 11, and no gastric
abnormality in one case. Mean wall thickness was 1.5 cm (range, 0.7–7.5 cm). The
finding of gastric wall thickness of 1 cm or greater had a sensitivity of 100% but a
specificity of only 42% for detection of malignant or potentially malignant stomach
lesions. The finding of focal, eccentric, or enhancing wall thickening had a sensitivity
of 93%, 71%, or 43%, respectively, and a specificity of 8%, 75%, or 88%, respec-
tively, for detection of these lesions. Gastric wall thickening that was 1 cm or greater
and was focal, eccentric, and enhancing had a specificity of 92% but a sensitivity of
only 36% for detection of these lesions.
CONCLUSION: Gastric wall thickness of 1 cm or greater at CT had a sensitivity of
100% but a specificity of less than 50% for detection of malignant or potentially
malignant stomach lesions that necessitated further diagnostic evaluation.
© RSNA, 2003
When evaluating the stomach with computed tomography (CT), it is important to be able
to differentiate benign conditions such as gastritis from ulcers or neoplasms that necessi-
tate further assessment with barium or endoscopic examinations. However, evaluation of
the stomach at CT may be limited by a variety of factors, including gastric peristalsis,
incomplete distention, an inadequate volume of oral contrast material, and retained food.
The results of several previously performed studies (1– 4) have shown that a normal gastric
wall usually has a thickness of 5 mm or less at CT, whereas a gastric wall involved by tumor
usually has a thickness of 1 cm or greater. However, in the radiologic literature, apart from
the data on degree of wall thickness, there is little information regarding the CT criteria for
differentiating normal stomach or benign conditions such as gastritis from potentially
malignant lesions that necessitate further diagnostic evaluation.
Our experience has been that patients who are suspected of having gastric neoplasms
because of a thickened wall at CT not infrequently are found to have gastritis or even a
normal stomach at barium or endoscopic examinations. Considering the established CT
criteria for differentiating benign from malignant lesions of the small or large bowel (5), we
wondered whether the degree of gastric wall thickening combined with other criteria such
 as symmetry, distribution, and enhance-
ment of the thickened gastric wall might
enable radiologists to better differentiate
benign from malignant conditions in-
volving the stomach at CT. With this
background, we performed a blinded ret-
Radiology
rospective study involving a large series
of patients who underwent abdominal
CT, as well as barium suspension exami-
nation, endoscopy, and/or surgery, to de-
termine the sensitivity and specificity of
CT criteria in the differentiation of be-
nign and malignant disease involving
the stomach in patients with a thickened
gastric wall at CT.
MATERIALS AND METHODS
Patient Selection
A computerized search of our hospi-
tal’s radiology files by one author (E.K.I.)
during a 3-year period from 1997 to 1999
revealed 152 patients with a thickened
gastric wall at abdominal CT (keyword
searches: thickened, bulky, prominent,
abnormal) who had also undergone dou-
ble-contrast barium suspension examina-
tions of the upper gastrointestinal (UGI)
tract. One hundred sixteen patients were
excluded from analysis for any of the fol-
lowing reasons: (a) The CT study was per-
formed without intravenously adminis-
tered contrast material, (b) the CT images
could not be retrieved from our image
archives, (c) a blinded review of the CT
images revealed no definite abnormali-
ties of the gastric wall and no inadequate
distention of the stomach, (d) the inter-
val between CT and barium suspension
examination was longer than 6 weeks,
and (e) the patients had undergone gas-
tric surgery previously.
The remaining 36 patients comprised
our study group. These 36 patients had a
total of 38 suspected gastric abnormali-
ties at CT. The mean interval between CT
and double-contrast UGI barium suspen-
sion examination was ⫾12.1 days (range,
1–32 days). The mean age of the patients
was 52 years (age range, 35– 80 years).
Seventeen patients were men and 19
women. Our institutional review board
approved all aspects of this retrospective
study and did not require informed con-
sent from the patients whose records
were included in our study.
CT Technique
All patients included in the study un-
derwent helical CT of the abdomen (with
a CT HiSpeed Advantage or HiSpeed CT/I
unit; GE Medical Systems, Milwaukee,
Wis) according to an established proto-
Volume 228 䡠 Number 1
col. All patients received oral contrast
material (600 – 800 mL of 2%–3% dia-
trizoate meglumine and diatrizoate so-
dium [Gastrografin]; Bristol-Myers Squibb,
Princeton, NJ, or a 2.1% wt/vol barium
suspension [Readi-cat], E-Z-Em, Westbury,
NY) 30 – 45 minutes before the CT exam-
ination and an additional 400 –500-mL
dose of the oral contrast material fol-
lowed by 3 g of an oral effervescent agent
dissolved in 8 ounces of water (Baros;
Lafayette Pharmaceuticals, Lafayette, Ind)
immediately before the examination. All
patients also received a 150-mL injection
of 60% iodinated contrast material (dia-
trizoate meglumine [Hypaque] or iohexol
[Omnipaque 300]; Nycomed Amersham,
Princeton, NJ) through an antecubital
vein at a rate of 2–3 mL/sec. The CT im-
ages were routinely obtained with the pa-
tient in a supine position during full in-
spiration. In one case, prone-position CT
images of the stomach also were ac-
quired. CT images of the upper abdomen
were obtained by using either 5-mm col-
limation and a pitch of 1.5:1.0 or 7-mm
collimation and a pitch of 1.3:1.0 (200 –
220 mAs); transverse images were recon-
structed with a soft-tissue algorithm.
UGI Examination Technique
All patients underwent double-con-
trast examinations of the UGI tract that
were performed by using digital fluoro-
scopic equipment (Diagnost 76 Plus;
Philips, Eindhoven, the Netherlands).
These studies, which were performed as
biphasic examinations by using an effer-
vescent agent (Baros) and a 250% wt/vol
barium suspension (E-Z-HD; E-Z-Em) fol-
lowed by a 50% wt/vol barium suspen-
sion (Entrobar; Lafayette Pharmaceuti-
cals), have been described previously (6).
All UGI examinations were performed by
a supervised resident (E.K.I.) or an at-
tending UGI radiologist (M.S.L.).
Image Review
The images acquired at CT scanning in
the 36 patients were reviewed by consen-
sus between two CT radiologists (B.A.B.,
13 years experience; J.E.J., 10 years expe-
rience) who had no knowledge of the
final radiologic, endoscopic, or patho-
logic findings. The greatest visible gastric
wall thickness was measured with elec-
tronic calipers at a picture archiving and
communications, or PACS, workstation.
All 38 cases involved a gastric wall thick-
ness of greater than 5 mm, which is a
previously established CT criterion for an
CT Differentiation of Benign and Malignant Stomach Lesions
abnormally thickened gastric wall (1,2).
The thickened wall was also evaluated for
symmetry (ie, eccentric or asymmetric vs
circumferential or symmetric), distribu-
tion (ie, focal vs diffuse), and presence or
absence of enhancement after intrave-
nous contrast material administration.
Final Diagnosis
The reports from the double-contrast
UGI examinations, as well as those from
the endoscopic, surgical, and/or patho-
logic examinations, were reviewed by
one author (E.K.I.). If the patient under-
went double-contrast UGI examination
but not endoscopy, the UGI findings
were accepted as the final diagnosis or
the reference standard. If the patient un-
derwent endoscopy or surgery, as well as
double-contrast UGI examination, the
endoscopic or histopathologic findings
in the endoscopic biopsy specimens or
resected surgical specimens were ac-
cepted as the reference standard. Al-
though it is conceivable that one or more
malignant tumors could have been
missed on the barium images, we believe
that the findings on these images were a
reasonable reference standard in the ab-
sence of endoscopic or pathologic corre-
lation, because double-contrast UGI stud-
ies have been shown to have a sensitivity
of greater than 95% in the detection of
gastric carcinomas (7). The final diagno-
sis was based on the UGI findings in 16
patients, on the histopathologic findings
at endoscopy alone in 13 patients, on the
endoscopic and surgical findings in six
patients, and on the surgical findings
alone in one patient. Two patients each
were found to have two separate abnor-
malities in the stomach. The diagnoses of
these lesions were based on the his-
topathologic findings at endoscopy alone
in one patient and on the findings at
endoscopy and surgery in the other pa-
tient. There were no discrepancies be-
tween the UGI examination findings and
the endoscopic or histopathologic find-
ings in the 19 patients who underwent
endoscopy or surgery and barium (ie,
UGI) examinations.
Criteria for Evaluating Accuracy
of CT
For the purposes of this study, all pa-
tients with a final diagnosis of a gastric
ulcer or neoplasm were considered to
have potentially malignant lesions that
warranted further diagnostic evaluation
with a double-contrast UGI or endo-
scopic examination. Ulcers were in-
䡠 167

Radiology
Figure 1. Antral gastritis in a 73-year-old
woman. (a) Transverse contrast material– en-
hanced CT image shows focal symmetric thick-
ening (arrow) of the wall of the gastric antrum
and no evidence of enhancement. The antral
wall has a thickness of 1.2 cm in this region.
(b) Left posterior oblique double-contrast UGI
image shows slightly thickened folds (arrow)
in the gastric antrum that are compatible with
mild antral gastritis. Endoscopic biopsy speci-
mens (not shown) revealed chronic inflamma-
tory changes in the antrum and no evidence of
tumor.
Figure 2. Transverse contrast-enhanced CT
image obtained in a 59-year-old woman with
antral gastritis shows focal asymmetric thick-
ening (arrow) of the anterior wall of the gastric
antrum and no evidence of enhancement. The
antral wall has a thickness of 1.0 cm in this
region. Subsequently obtained double-contrast
UGI tract image (not shown) revealed antral
gastritis with thickened antral folds but no ev-
idence of tumor.
168 䡠 Radiology 䡠 July 2003
Figure 3. Hiatal hernia in a 77-year-old
woman. (a) Transverse contrast-enhanced CT
image shows focal asymmetric thickening (ar-
row) of the posteromedial wall of the gastric
fundus in the region of the cardia and no evi-
dence of enhancement. The fundal wall has a
thickness of 2.0 cm in this region. (b) Right
lateral double-contrast UGI tract image shows
a hiatal hernia (arrow) and no evidence of tu-
mor in the gastric fundus. In retrospect, the
focal wall thickening of the stomach that was
depicted at CT probably resulted from incom-
plete distention of the hiatal hernia, with pro-
lapse of the hernia into the fundus.
cluded in this group, because we are
aware of no reliable criteria for differen-
tiating benign ulcers from malignant ul-
cers on the basis of CT findings. Gastro-
intestinal stromal tumors also were
included in this group, because 10%–30%
of these neoplasms are found to be ma-
lignant (8). Conversely, patients with a
final diagnosis of gastritis, hiatal hernia,
or a normal stomach were considered to
have findings that did not warrant fur-
ther diagnostic evaluation. We then cal-
culated the sensitivity and specificity of
the various parameters of gastric wall
thickening in the detection of malignant
or potentially malignant stomach lesions
in our study group.
RESULTS
In the 38 cases of a thickened gastric wall
at CT, the mean wall thickness was 1.5
cm (range, 0.7–7.5 cm). The wall thick-
ness was 1 cm or greater in 28 (74%) cases
and less than 1 cm in the remaining 10
(26%). The final diagnoses were gastritis
in 19 (Figs 1, 2), hiatal hernia in four (Fig
3), benign ulcer in three (Fig 4), and gas-
tric neoplasm in 11 cases: seven gastric
carcinomas (four infiltrating carcinomas,
one ulcerated carcinoma, and two scir-
rhous carcinomas [Figs 5, 6]), three be-
nign gastrointestinal stromal tumors (Fig
7), and one esophageal carcinoma invad-
ing the gastric fundus. No abnormalities
were found in the stomach in the re-
maining case. Thus, in the 38 cases of a
thickened gastric wall at CT, the stomach
was either involved by benign disease
(n ⫽ 23) or normal (n ⫽ 1) in 24 cases,
which warranted no further diagnostic
evaluation, whereas malignant or poten-
tially malignant lesions were present in
14 cases, which warranted further evalu-
ation with barium or endoscopic exami-
nations.
With use of the 1-cm wall thickness
threshold, the mean wall thickness of the
stomach was 1 cm or greater (mean, 2.5
cm; range, 1.2–7.5 cm) in all 14 (100%)
cases of malignant or potentially malig-
nant lesions that warranted further diag-
nostic evaluation (Figs 4 –7). Conversely,
the mean wall thickness was 1 cm or
greater in 14 (58%) (Figs 1–3) and less
than 1 cm in 10 (42%) of the 24 cases of
benign conditions that warranted no fur-
ther diagnostic evaluation (mean, 1.0
cm; range, 0.7–2.4 cm). Therefore, the CT
finding of a gastric wall thickness of 1 cm
or greater had a sensitivity of 100% but a
specificity of only 42% in the detection
of malignant or potentially malignant
stomach lesions.
Further analysis of our study data re-
vealed that the mean wall thickness of
the stomach was 2 cm or greater (Figs 6,
7) in seven (50%) and less than 2 cm (Figs
4, 5) in seven (50%) of the 14 cases of
malignant or potentially malignant le-
sions that warranted further diagnostic
evaluation. Conversely, the mean wall
thickness was 2 cm or greater (Fig 3) in
three (12%) and less than 2 cm (Figs 1, 2)
in 21 (88%) of the 24 cases of benign
conditions that did not warrant further
diagnostic evaluation. Therefore, when
the threshold for a thickened gastric wall
was increased to 2 cm or greater at CT,
the specificity of this finding in the de-
tection of malignant or potentially ma-
Insko et al
 lignant lesions increased to 88%, but the
sensitivity decreased to 50%.
The thickened gastric wall had a focal
distribution in 13 (93%) (Figs 4 –7) and a
diffuse distribution in one (7%) of the 14
Radiology

cases of malignant or potentially malig-

nant lesions. Wall thickening was eccen-
tric or asymmetric in 10 (71%) (Figs 4, 5,
7) and circumferential or symmetric in
four (29%) (Fig 6) of the 14 cases of ma-
lignant or potentially malignant lesions.
After intravenous administration of con-
trast material, the thickened wall en-
hanced in six (43%) (Figs 5, 6) but did not
enhance in eight (57%) (Figs 4, 7) of the
14 cases of malignant or potentially ma-
lignant lesions.
Conversely, the thickened gastric wall
had a focal distribution in 22 (92%) (Figs
1–3) and a diffuse distribution in two
(8%) of the 24 cases of benign conditions
that did not warrant further diagnostic
evaluation. Wall thickening was circum-
ferential or symmetric in 18 (75%) of
these 24 cases (Fig 1) and eccentric or
asymmetric in six (25%) (Figs 2, 3). The
thickened wall enhanced after intrave-
nous administration of contrast material
in three (12%) of these 24 cases but did
not enhance in 21 (88%) (Figs 1–3). Thus,
the finding of focal, eccentric, or enhanc-
Figure 5. Ulcerated carcinoma of the stom-
ing wall thickening had a sensitivity of ach in an 82-year-old man. (a) Transverse con-
93%, 71%, or 43%, respectively, and a trast-enhanced CT image shows focal asym-
specificity of 8%, 75%, or 88%, respec- Figure 4. Benign gastric ulcer in a 72-year-old metric thickening (arrow) of the medial wall of
tively, in the detection of malignant or man. (a) Transverse contrast-enhanced CT im- the proximal gastric body and heterogeneous
age shows focal asymmetric thickening (arrow) enhancement. The antral wall has a thickness
potentially malignant stomach lesions. of the wall of the gastric antrum and no evidence of 1.6 cm in this region. (b) Left posterior
Gastric wall thickening that was 1 cm of enhancement. The antral wall has a thickness oblique double-contrast UGI tract image
or greater, focal, eccentric, and enhanc- of 1.3 cm in this region. In retrospect, barium shows a polypoid mass (black arrows) with a
ing had a specificity of 92% (22 of 24 was probably trapped in an antral ulcer (arrow- flat central ulcer (white arrow) in the lesser
cases) but a sensitivity of only 36% (five head). (b) Frontal double-contrast UGI tract im- curvature of the proximal gastric body. Endo-
of 14 cases) in the detection of malignant age shows an ulcer (black arrow) in the distal scopic results (not shown) confirmed the pres-
antrum with a large surrounding mound of ence of an ulcerated mass in the stomach, and
or potentially malignant lesions at CT edema (white arrows). Endoscopic biopsy speci- biopsy specimens (not shown) revealed gastric
(Figs 5, 6). mens (not shown) confirmed the presence of a adenocarcinoma. The lesion was surgically re-
Four patients had wall thickening in benign gastric ulcer with Helicobacter pylori gas- sected.
the gastric fundus that resulted from an tritis.
incompletely distended hiatal hernia (Fig
3). The mean wall thickness in these
cases was 1.4 cm (range, 0.7–2.0 cm). The fervescent agents can be used to improve
thickened wall enhanced after intrave- DISCUSSION gastric distention, and, in questionable
nous administration of contrast material cases, supplemental CT can be performed
in three (75%) of these four cases. Because abdominal CT is being per- with the patient in the prone position to
All three patients with benign condi- formed with greater frequency, it is in- better evaluate the proximal portion of
tions in whom the thickened gastric wall creasingly becoming possible to detect a the stomach. Ultimately, however, radi-
enhanced after intravenous administra- variety of abnormalities in the stomach ologists must decide which patients re-
tion of contrast material were found to of both symptomatic and asymptomatic quire further diagnostic evaluation to
have hiatal hernias. Thus, the presence of patients who have gastric disease. The rule out neoplastic lesions in the stom-
a thickened gastric wall that enhanced challenge for radiologists when interpret- ach.
after intravenous administration of con- ing these CT images is to differentiate a In our study, the primary CT finding in
trast material had an overall specificity of normal stomach or gastritis from malig- patients with suspected abnormalities of
88% in the detection of malignant or po- nant or potentially malignant lesions the stomach was a thickened gastric wall. A
tentially malignant stomach lesions at such as gastric ulcers and neoplasms that wall thickness of 1 cm or greater as the
CT, but the specificity increased to 100% warrant further investigation with dou- threshold criterion had a sensitivity of
when the cases of hiatal hernia were ex- ble-contrast UGI tract or endoscopic ex- 100% in the detection of malignant or po-
cluded. aminations. Oral contrast material or ef- tentially malignant gastric lesions that

Volume 228 䡠 Number 1 CT Differentiation of Benign and Malignant Stomach Lesions 䡠 169

Radiology
Figure 6. Scirrhous carcinoma of the stom-
ach in a 61-year-old man. (a) Transverse con-
trast-enhanced CT image shows focal cir-
cumferential thickening (arrow) of the wall
of the gastric antrum and heterogeneous en-
hancement. The antral wall has a thickness
of 2.0 cm in this region. (b) Left posterior
oblique double-contrast UGI tract image
shows irregular narrowing of the gastric an-
trum (arrows). Endoscopic biopsy specimens
(not shown) revealed scirrhous adenocarci-
noma of the stomach.
warranted further diagnostic evaluation
with barium or endoscopic examinations.
A wall thickness of 1 cm or greater has also
been used as the threshold criterion for the
presence of malignant tumor at CT in
other studies (1,2). In our study, however,
this threshold had a specificity of only
42%, so unnecessary endoscopic or barium
examinations were performed in 58% of
patients with gastric wall thickening of 1
cm or greater at CT. When the threshold
for malignant tumor was increased to 2 cm
or greater at CT, the specificity of this find-
ing increased to 88% but the sensitivity
decreased to 50%. Thus, no single thresh-
old enabled gastric wall thickness to be
both a sensitive and a specific criterion for
the detection of malignant or potentially
malignant stomach lesions at CT.
In the patients with a gastric wall thick-
ness of 1 cm or greater, the additional CT
finding of focal, eccentric, or enhancing
170 䡠 Radiology 䡠 July 2003
Figure 7. Gastrointestinal stromal tumor in a
52-year-old man. (a) Transverse contrast-en-
hanced CT image shows focal masslike thicken-
ing (arrow) of the posterior wall of the fundus.
This wall has lower attenuation than the adja-
cent wall and no evidence of enhancement. The
lesion has a thickness of 2.2 cm. (b) Right lateral
double-contrast UGI tract image shows a smooth
submucosal-appearing mass (arrow) on the pos-
terior wall of the gastric fundus. Endoscopic bi-
opsy specimens (not shown) revealed a gastroin-
testinal stromal tumor in the gastric fundus. The
lesion was surgically resected.
wall thickening had a sensitivity of 93%,
71%, or 43%, respectively, and a specificity
of 8%, 75%, or 88%, respectively, in the
detection of malignant or potentially ma-
lignant stomach lesions. Thus, focal wall
thickening was a sensitive but very non-
specific CT finding, whereas eccentric wall
thickening and enhancing wall thickening
were more specific but considerably less
sensitive CT findings in the detection of
these lesions.
Although no individual CT finding was
both sensitive and specific for the detec-
tion of malignant stomach lesions, the
presence of a gastric wall that was 1 cm or
greater in thickness, focal, eccentric, and
enhancing after intravenous administra-
tion of contrast material had a specificity
of 92% in the detection of these lesions
(Figs 5, 6). Therefore, this combination of
findings at CT should be highly sugges-
tive of malignant tumor, necessitating
further evaluation with barium or endo-
scopic examinations for a definitive diag-
nosis. Nevertheless, such an appearance
of the thickened gastric wall was present
at CT in only 36% of patients with ma-
lignant tumors in the stomach, limiting
the utility of CT for this constellation of
findings.
Particular diagnostic difficulties were
encountered at CT in four patients with
hiatal hernias: Incomplete filling of the
hernia created an erroneous impression
of gastric wall thickening in the region of
the gastric fundus. This so-called pseudo-
tumor at the gastroesophageal junction
has been well documented at CT in pa-
tients with hiatal hernias (9,10). In such
cases, performing CT in patients in the
prone position may be helpful for im-
proving the gaseous distention of the
fundus and differentiating a pseudotu-
mor caused by a partially collapsed hiatal
hernia from a true neoplastic lesion in
this region.
Our study was limited by the fact that
the primary inclusion criterion was an
abnormally thickened gastric wall at CT.
As a result, we have no way of knowing
how many patients with benign or ma-
lignant stomach lesions had no evidence
of gastric wall thickening at CT. Also,
because this was a retrospective study,
there was no set protocol for performing
CT in patients in the prone position to
optimize the gaseous distention in the
proximal portion of the stomach. An er-
roneous impression of gastric wall thick-
ening at CT can also result from inade-
quate distention by the oral contrast
material or effervescent agent (11,12). Be-
cause of this problem, we excluded those
cases in which the stomach was inade-
quately distended at CT to minimize the
number of false-positive cases related to
technical artifacts. In general, however,
additional oral contrast material or effer-
vescent agent should be administered
when the assessment of gastric wall
thickness is compromised by inadequate
gastric distention in patients with signs
or symptoms of gastric disease. Because
of the limitations of our study, we believe
that additional prospective studies to fur-
ther delineate the CT findings of benign
versus malignant conditions of the stom-
ach are warranted (13).
In conclusion, a gastric wall thickness of
1 cm or greater at CT had a sensitivity of
100% but a specificity of less than 50% in
the detection of malignant or potentially
malignant stomach lesions that warranted
further diagnostic evaluation. Focal wall
thickening was found to be a sensitive but
very nonspecific CT criterion for assessing
Insko et al
 the malignancy potential of a thickened
gastric wall, whereas eccentric wall thick-
ening and enhancing wall thickening were
found to be more specific but considerably
less sensitive criteria. Wall thickening that
Radiology
was focal, eccentric, and enhancing almost
always indicated malignant tumor, but
this constellation of findings was present
in only a small percentage of cases. Radiol-
ogists should be aware of the usefulness of
these specific criteria in the evaluation of
gastric wall thickening at CT to better dif-
ferentiate gastritis or a normal stomach
from malignant or potentially malignant
lesions that warrant further diagnostic
evaluation with barium or endoscopic ex-
aminations.
References
1. Lee KR, Levine E, Moffat RE, Bigongiari LR,
Hermreck AS. Computed tomographic
Volume 228 䡠 Number 1
staging of malignant gastric neoplasms. Ra-
diology 1979; 133:151–155.
2. Balfe DM, Koehler RE, Karstaedt N, Stan-
ley RJ, Sagel SS. Computed tomography of
gastric neoplasms. Radiology 1981; 140:
431– 436.
3. Moss AA. Computed tomography in the
staging of gastrointestinal carcinoma. Ra-
diol Clin North Am 1982; 20:761–780.
4. Scatarige JC, DiSantis DJ. CT of the stom-
ach and duodenum. Radiol Clin North
Am 1989; 27:687–706.
5. Macari M, Balthazar EJ. CT of bowel wall
thickening: significance and pitfalls of in-
terpretation. AJR Am J Roentgenol 2001;
176:1105–1116.
6. Levine MS, Rubesin SE, Herlinger H, Laufer
I. Double contrast upper gastrointestinal
examination: technique and interpreta-
tion. Radiology 1988; 168:593–602.
7. Low VHS, Levine MS, Rubesin SE, Laufer
I, Herlinger H. Diagnosis of gastric carci-
noma: sensitivity of double-contrast bar-
ium studies. AJR Am J Roentgenol 1994;
162:329 –334.
CT Differentiation of Benign and Malignant Stomach Lesions
8. Miettinen M, Sarlomo-Rikala M, Lasota J.
Gastrointestinal stromal tumors: recent
advances in understanding of their biol-
ogy. Hum Pathol 1999; 30:1213–1220.
9. Pupols A, Ruzicka FF. Hiatal hernia caus-
ing a pseudomass on computed tomogra-
phy. J Comput Assist Tomogr 1984;
4:699 –700.
10. Halvorsen RA. Thompson WA. Com-
puted tomography of the gastrointestinal
junction. Crit Rev Diagn Imaging 1984;
21:83–228.
11. Kaye MD, Yound SW, Hayward R, Cas-
tellino RA. Gastric pseudotumor on CT
scanning. AJR Am J Roentgenol 1980; 135:
190 –193.
12. Komaki S. Normal or benign gastric wall
thickening demonstrated by computed
tomography. J Comput Assist Tomogr
1982; 6:1103–1107.
13. Rockey DC, Halvorsen RA, Higgins JL,
Cello JP. Prospective evaluation of pa-
tients with bowel wall thickening. Am J
Gastroenterol 1995; 90:99 –103.
䡠 171