AJR 2003; 181:973-979
© American Roentgen Ray Society
Wall Thickening of the Gastric Antrum as a Normal Finding: Multidetector CT with Cadaveric Comparison
Perry J. Pickhardt1,2 and
Dean B. Asher1
1 Department of Radiology, National Naval Medical Center, 8901 Wisconsin Ave.,
Bethesda, MD 20889-5600.
2 Department of Radiology and Nuclear Medicine, Uniformed Services University of
the Health Sciences, Bethesda, MD 20889.
Received February 18, 2003;
accepted after revision April 29, 2003.
Address correspondence to P. J. Pickhardt.
The opinions and assertions contained herein are the private views of the
authors and are not to be construed as official or as reflecting the views of
the Departments of the Navy, Army, Air Force, or Defense.
Abstract
OBJECTIVE. The purpose of this study was to establish the normal
range of wall thickness and the normal appearance of the gastric antrum on
multidetector CT (MDCT).
METHODS AND MATERIALS. Soft-copy measurements of the gastric antrum
and gastric body were performed on contrast-enhanced MDCT scans in 153
consecutive patients without gastric disease. For comparison, anatomic
dissection of the stomach was performed in three cadavers.
RESULTS. Smooth thickening of the distal gastric antrum relative to
the proximal stomach on MDCT was seen in 152 (99%) of 153 patients and
appeared concentric in 96% and eccentric in 4%. The mean (± SD) antral
wall thickness was 5.1 ± 1.6 mm. The longitudinal extent of antral wall
thickening averaged 4.6 cm. At least one antral wall measurement (anterior or
posterior) exceeded 5 and 10 mm in 85 patients (56%) and seven patients (5%),
respectively. The anterior wall of the gastric body was significantly thinner
at 2.0 ± 0.4 mm (mean ± SD) than the wall of the gastric antrum
(p « 0.0001). The mean antral wall thickness when distention
was characterized as grade 1 (least), 2, 3, and 4 (most) was 6.9, 5.1, 4.9,
and 4.0 mm, respectively. Linear submucosal low attenuation (mural striation)
of the thickened portion of the gastric antrum was noted in 36 patients (24%);
fat attenuation was present in 14 cases. Cadaveric stomachs showed mild
segmental thickening of the distal gastric antrum, but this thickening was
less pronounced compared with in vivo MDCT findings.
CONCLUSION. Smooth wall thickening of the distal gastric antrum
relative to the proximal stomach on MDCT with or without submucosal low
attenuation is a normal finding. Antral wall thickness commonly exceeds 5 mm
and may measure up to 12 mm. Our MDCT findings, in conjunction with previous
anatomic and physiologic observations, suggest that normal antral wall
thickening consists of both static and dynamic components.
Introduction
CT has certain intrinsic advantages over luminal studies because of its
ability to reveal intra- and extramural aspects of the gut wall. To wit, wall
thickening on CT is an important, albeit nonspecific, indicator of disease of
the alimentary tract. A common pitfall in the CT diagnosis of abnormal wall
thickening is inadequate luminal distention, especially for the stomach. Some
sources suggest 5 mm as the cutoff for abnormal gastric wall thickening,
assuming proper distention. However, we have commonly observed prominent wall
thickening of the distal gastric antrum on multidetector CT (MDCT) that often
exceeds 5 mm, even in the setting of adequate distention. To avoid
recommending unnecessary endoscopy in patients who otherwise have no evidence
of gastric disease, we sought to establish the range of normal findings of the
gastric antrum on MDCT. We also suggest an explanation for this thickening
based on previous observations of gastric anatomy and physiology.
Materials and Methods
Contrast-enhanced MDCT scans of 153 consecutive adult patients without
known or suspected gastric disease were retrospectively analyzed. Exclusion
criteria included a clinical history of or MDCT findings of upper abdominal
malignancy, surgery, or an inflammatory process. Patients who did not receive
both oral and IV contrast material were also excluded. The mean age of the
study group was 53.1 years (range, 19–86 years). Of the 153 patients,
101 were men and 52 were women.
All studies were performed on an MDCT scanner (LightSpeed, General Electric
Medical Systems, Milwaukee, WI) after the administration of oral and IV
contrast material. Our standard abdominal protocol includes 5-mm collimation,
11.25 mm/sec table speed, and 5-mm reconstruction interval. Our standard IV
contrast regimen consists of 150 mL of nonionic contrast material administered
at a rate of 3 mL/sec and a 60-sec scanning delay. Our standard oral contrast
regimen consists of 1000 mL of water-soluble contrast material consumed over
45–60 min. Neither water nor effervescent crystals are used routinely
but are occasionally used.
Soft-copy measurements of the gastric antrum during the portal venous phase
with actual-size magnification (i.e., life-size representation) were performed
in the axial plane on a diagnostic workstation using electronic calipers. Two
reviewers performed all the measurements, and disagreement was resolved by
consensus opinion. Anterior and posterior antral measurements were taken in
the mid portion of the relative wall thickening—away from the pylorus
and excluding the rugal folds. Care was also taken to avoid portions of the
greater or lesser curve that would result in pseudothickening from oblique
measurement. A similar representative measurement of the anterior wall of the
proximal gastric body was obtained. The nondependent gastric body was chosen
for comparison over the fundus and cardia to minimize the frequency of
pseudothickening from poor distention. SD values were obtained for all gastric
wall measurements.
The morphologic and attenuation characteristics of the antral wall were
also evaluated. Morphologic features evaluated were smooth versus irregular
contour, concentric versus eccentric thickening, and the estimated
longitudinal extent of thickening. In addition to subjective assessment,
eccentric thickening was defined as one antral wall (anterior or posterior)
measuring less than 50% of the other. If submucosal linear low attenuation
(mural striation) was apparent on MDCT, attenuation values in the region of
low attenuation were recorded. Antral distension was subjectively graded on a
4-point scale, with grades 1 through 4 roughly indicating 0–25%,
26–50%, 51–75%, and 76–100% of perceived distention,
respectively.
A review of the medical records revealed that 10 (7%) of the 153 patients
had undergone upper endoscopy within 8 months of the MDCT examination (mean
interval, 3.3 months). Endoscopy was not performed for suspected antral
disease in any of these patients. Indications for performing upper endoscopy
were esophageal disease in five patients (dysphagia in two, gastroesophageal
reflux disease in two, and screening for esophageal varices in one),
percutaneous endoscopic gastrostomy in two patients, anemia in two patients,
and persistent nausea in one patient. The endoscopic results in these patients
were correlated with the MDCT findings.
Three unembalmed cadavers were dissected to supplement the in vivo MDCT
findings. Permission from the medical school was obtained before cadaveric
dissection. All three patients had died of cardiopulmonary causes, and none
had a history of gastric disease. From each cadaver, the entire stomach was
carefully excised without disrupting the pyloric region. The stomachs were
then longitudinally sectioned to evaluate the walls of the gastric body and
gastric antrum. Gross tissue samples consisting of the full wall thickness
were placed in formaldehyde solution. These specimens were later prepared for
histologic evaluation with standard H and E staining.
Statistical analysis of our results was performed using the Student's
t test. A p value of less than 0.05 was considered
significant. Ninety-five percent confidence intervals were calculated for mean
antral thickness for each degree of distention.
Results
The gastric antrum, pylorus, and duodenal bulb were readily identified at
MDCT in each patient. Smooth wall thickening of the distal gastric antrum
relative to the gastric body was seen in 152 (99%) of the 153 patients. In the
remaining patient, mild concentric thickening of the gastric antrum was
present but measured less than the anterior gastric body. Antral thickening
extended to the pylorus in all 153 patients and appeared circumferential in
147 patients (96%) (Figs. 1,
2,
3) and eccentric in six
patients (4%) (Fig. 4A,
4B). Mean antral thickness
measured 5.0 ± 1.9 mm for the anterior wall, 5.2 ± 1.7 mm for
the posterior wall, and 5.1 ± 1.6 mm overall (range, 1.8–12.0 mm)
(Fig. 5). The difference in
mean thickness between the anterior and posterior antral walls was not
significant. The mean anterior wall thickness of the gastric body, measuring
2.0 ± 0.4 mm, was significantly less than that of the gastric antrum
(p « 0.0001).
View larger version (161K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1. —Contrast-enhanced axial multidetector CT (MDCT) image in
36-year-old man without gastric disease shows mild to moderate uniform
concentric wall thickening involving well-distended gastric antrum
(arrowheads). Pylorus and duodenal bulb are well depicted on this
image. This appearance of gastric antrum was most common on MDCT.
|
|
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2. —Contrast-enhanced axial multidetector CT image in 45-year-old
woman without gastric disease shows prominent short-segment circumferential
wall thickening (arrowheads) of distal gastric antrum. Note
thin-walled gastric body (arrows).
|
|
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3. —Contrast-enhanced axial multidetector CT image in 65-year-old
man without gastric disease shows moderate antral wall thickening
(arrowheads) relative to gastric body. Note linear submucosal fat
attenuation in area of antral thickening.
|
|
View larger version (144K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A. —Transient asymmetric antral wall thickening in 34-year-old
man without gastric disease. Contrast-enhanced axial multidetector CT (MDCT)
image during dynamic phase shows asymmetric antral wall thickening
(arrowhead). Subtle submucosal low attenuation is present focally at
or near pylorus. Note thin wall of more proximal gastric body
(arrow).
|
|
View larger version (161K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B. —Transient asymmetric antral wall thickening in 34-year-old
man without gastric disease. Contrast-enhanced axial MDCT image during delayed
phase shows resolution of asymmetric antral thickening (short
arrows), leaving only mild uniform thickening relative to proximal
gastric body (long arrow).
|
|
View larger version (18K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5. —Bar graph shows distribution of measurements of anterior
(white bars) and posterior (black bars) antral wall
thicknesses on multidetector CT in 153 consecutive patients without gastric
disease.
|
|
At least one of the two antral measurements (anterior or posterior)
exceeded 5 mm in 85 patients (56%) and 10 mm in seven patients (5%). In four
patients, both antral measurements were less than 3 mm, but each of these
measurements was still thicker than the gastric body. Anterior and posterior
measurements were within 1 and 3 mm of each other in 100 patients (65%) and
147 patients (96%), respectively. In no cases were both antral measurements
less than 2 mm, but one wall was less than 2 mm (1.8 or 1.9 mm) in three
patients. No correlation between the degree of antral thickening and either
patient age or sex was apparent (mean thickening in men vs women, 5.1 vs 4.9
mm; p = 0.4). The gastric body measurement exceeded 3 mm in only
three patients (2%). Although not actually measured, the wall of the duodenal
bulb was noticeably thinner than the adjacent gastric antrum in nearly every
patient and never appeared thicker (Figs.
1 and
4A,
4B).
The estimated mean longitudinal extent of the antral wall thickening was
4.6 ± 1.1 cm. Longitudinal involvement ranged from thickening of a
relatively long segment that gradually tapered (Fig.
6A,
6B,
6C) to thickening of a short
segment with a more abrupt termination
(Fig. 2). Luminal distention of
the gastric antrum on MDCT was considered grade 1 (least distended) in 65
patients, grade 2 in 22 patients, grade 3 in 37 patients, and grade 4 in 29
patients. The mean antral wall thickness for grades 1, 2, 3, and 4 distention
was 6.9, 5.1, 4.9, and 4.0 mm, respectively
(Fig. 7). The difference in
antral thickening between each group was statistically significant (p
< 0.001) except when comparing grades 2 and 3 distention (p =
0.58).
View larger version (148K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6A. —Images show effect of degree of luminal distention on
appearance of concentric long-segment antral thickening in 62-year-old woman
without gastric disease. Contrast-enhanced axial multidetector CT (MDCT) image
during dynamic phase shows concentric antral thickening (arrowheads)
with moderate luminal distention. Note conspicuity of bright mucosal
enhancement using water as oral contrast material, which would be obscured by
positive oral contrast material.
|
|
View larger version (141K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6B. —Images show effect of degree of luminal distention on
appearance of concentric long-segment antral thickening in 62-year-old woman
without gastric disease. Contrast-enhanced axial MDCT image during delayed
phase shows decreased luminal distention and further increase in antral
thickening (arrowheads).
|
|
View larger version (148K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6C. —Images show effect of degree of luminal distention on
appearance of concentric long-segment antral thickening in 62-year-old woman
without gastric disease. Delayed axial MDCT image with optimal distention and
patient in decubitus position after ingesting effervescent crystals shows
decreased but persistent antral wall thickening (arrowheads).
|
|
Intramural low attenuation (mural striation) of the thickened portion of
the gastric antrum was apparent on MDCT in 36 (24%) of the 153 patients
(Fig. 3). In each case of mural
striation, the area of low attenuation appeared thin and linear and was
restricted to the submucosal region. Visually subtle cases were often limited
to the immediate pyloric and prepyloric regions (Fig.
4A,
4B). Submucosal low attenuation
was not seen elsewhere in the stomach in any patient. The mean antral wall
thickness in the presence of mural striation was 5.6 mm, compared with 4.9 mm
in its absence (p = 0.045). Repeated sampling with the cursor along
the area of submucosal low attenuation produced a consistently negative value
for attenuation (Hounsfield units) in 14 (39%) of the 36 patients. Negative
attenuation values were suggestive of fat deposition in these patients, which
was further supported by subjective visual assessment
(Fig. 3). The thin linear
geometry of the low-attenuation area precluded placing a region of interest
for a more accurate determination of attenuation.
Upper endoscopy performed in 10 of the 153 patients showed no evidence of
antral inflammation, fold thickening, or other antral abnormality in any
patient. The mean antral wall thickness on MDCT of the patients who underwent
endoscopy was 4.6 mm.
The dissected stomachs from the three cadavers showed mild but perceptible
wall thickening of the distal gastric antrum that extended to the pylorus
(Fig. 8A,
8B,
8C). Precise measurement of
wall thickness using calipers proved difficult, but the distribution of antral
thickening was similar to the typical MDCT findings in terms of longitudinal
extent and circumferential involvement. Histologic examination of full wall
thickness specimens from the distal gastric antrum and proximal gastric body
showed that this static increase in antral wall thickness primarily resulted
from underlying differences in the muscularis propria (Fig.
8A,
8B,
8C). The approximate ratio of
antrum-to-body wall thickness was generally less than 2:1 for the cadaveric
specimens, which is less than the ratio of 2.6:1 for the mean thickness (5.1
mm:2.0 mm) seen in vivo on MDCT. Possible reasons for this discrepancy are
discussed in the next section of this article.
View larger version (122K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8A. —Gastric evaluation from dissection of cadaver of 85-year-old
woman who died of cardiopulmonary cause and who did not have history of
gastric disease. Photograph of distal gastric antrum and pylorus after
longitudinal incision shows mild wall thickening of distal gastric antrum
(arrowheads) leading to pylorus (arrows). Wall of gastric
body was perceptibly thinner (not shown).
|
|
View larger version (124K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8B. —Gastric evaluation from dissection of cadaver of 85-year-old
woman who died of cardiopulmonary cause and who did not have history of
gastric disease. Low-power photomicrographs of histologic sections through
anterior wall of distal gastric antrum (B) and proximal gastric body
(C) at same magnification show that relative antral thickening is
primarily caused by differences in circular smooth muscle layer
(asterisks). (H and E, x20)
|
|
View larger version (149K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8C. —Gastric evaluation from dissection of cadaver of 85-year-old
woman who died of cardiopulmonary cause and who did not have history of
gastric disease. Low-power photomicrographs of histologic sections through
anterior wall of distal gastric antrum (B) and proximal gastric body
(C) at same magnification show that relative antral thickening is
primarily caused by differences in circular smooth muscle layer
(asterisks). (H and E, x20)
|
|
Discussion
For more than 50 years, barium studies have been used to evaluate the
stomach for thickening of the rugal folds
[1,
2]. However, conventional
radiographic techniques are limited in their ability to depict the actual
thickness of the gastric wall because they are primarily luminal examinations.
Studies with conventional CT that emerged in the late 1970s and early 1980s
showed that CT is useful for more direct assessment of the gastric wall
[3–6].
Despite the advantages of CT, reliable differentiation between abnormal and
normal wall thickening on CT has proven difficult.
Gastric pseudothickening from incomplete distention is a well-recognized
phenomenon, particularly in the region of the esophagogastric junction. In an
adequately distended stomach, however, most sources have stated that gastric
wall thickness should not exceed 5 mm
[7–10].
Our findings show that, although this cutoff is appropriate for the gastric
body, this cutoff is often not appropriate for the distal gastric antrum.
Although most radiologists who interpret abdominal CT scans likely assume that
this degree of antral thickening is normal, surprisingly little evidence is
available in the literature to actually support this assumption. Earlier
studies with conventional CT attributed this apparent antral thickening to
artifacts related to the scanning section because images were generally
obtained with 10-mm collimation and were sometimes obtained at 20-mm intervals
[3]. Gross pseudothickening
related to motion artifacts or to oblique sectioning of the gastric antrum is
substantially reduced or eliminated with single-detector CT and MDCT because
of faster scanning; thinner sections; and contiguous, or even overlapping,
image display. Despite this reduction in gross pseudothickening, we have
anecdotally noted a high frequency of prominent antral thickening on MDCT
scans in patients without suspected gastric disease, which served as the
motivation for conducting our study. In our experience, further diagnostic
workup in these patients, including endoscopy, generally yields negative
findings.
Our findings show that relative wall thickening of the distal gastric
antrum compared with the proximal stomach on MDCT is a normal finding. In our
study group of patients without suspected gastric disease, antral thickness
averaged 5.1 mm, measured greater than 5 mm in the majority of patients, and
was 10–12 mm in 5% of the patients. In comparison, the wall thickness of
the nondependent gastric body averaged only 2.0 mm and exceeded 3.0 mm in only
2% of the patients. Therefore, normal gastric wall thickness is a
site-specific measurement: a 5-mm threshold seems appropriate for the
well-distended gastric body but not for the distal gastric antrum.
Furthermore, the use of a simple linear CT measurement as the sole diagnostic
criterion for gastric disease is not sufficient; morphologic assessment of
wall uniformity and static versus dynamic thickening must also be
considered.
We have also shown that although incomplete distention has a measurable
effect on antral wall thickness, it is not the sole factor. Furthermore,
positioning appears to affect antral wall thickening less than the remainder
of the stomach, because the mean thickness of the dependent posterior wall did
not significantly differ from the nondependent anterior wall (5.2 vs 5.0 mm).
Unlike observations reported for the proximal stomach
[10], underlying wall
thickening in the antral region can persist despite proper distention and
nondependent positioning (Fig.
6A,
6B,
6C).
Antral wall thickening seen on MDCT is perhaps best explained in terms of
both static and dynamic components. The primary static or anatomic contributor
to antral wall thickening is its augmented muscular coat. Anatomic studies
have shown that gastric smooth muscle, particularly the circular layer, is
thicker and denser around the gastric antrum than around the rest of the
stomach [11]. Our gross and
histologic findings from cadaveric dissections support this observation (Fig.
8A,
8B,
8C). However, this underlying
muscular thickening, which helps promote the antral role of grinding
[12,
13], is not the only
contributor to antral wall thickening seen on MDCT.
Dynamic or physiologic contributors to antral thickening are more complex.
We have shown that incomplete distention is one factor. Gastric motility and
the resulting antral contractions, however, are likely equally important.
Whereas the gastric fundus and proximal gastric body largely serve as a
passive reservoir, the distal stomach is more electrically active, generating
intense peristaltic contractions at a rate of approximately three cycles per
minute [14]. These antral
contractions can be readily observed on real-time examination of the stomach,
such as at endoscopy or sonography (Fig.
9A,
9B,
9C). Periodic antral
contractions also likely contribute to the additional thickening (concentric
and eccentric) seen in some patients on MDCT, as shown by temporal changes
seen on separate series (Fig.
4A,
4B). The thickening of the
distal gastric antrum that develops during a contraction on sonography
strongly resembles the short-segment thickening seen on MDCT in some patients,
as shown by comparison of Figures
2 and
9A,
9B,
9C. These dynamic events are
perhaps captured more effectively on MDCT than on conventional CT because of
faster scanning.
View larger version (120K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9A. —Sonographic appearance of contraction of normal gastric
antrum in 34-year-old man without gastric disease. Initial transverse
sonographic image of water-filled stomach shows good luminal distention of
distal gastric antrum (asterisk) and mild associated wall thickening
(arrows). Note similarity of this image to
Figure 1.
|
|
View larger version (119K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9B. —Sonographic appearance of contraction of normal gastric
antrum in 34-year-old man without gastric disease. Sequential sonographic
images show progressive luminal narrowing and wall thickening of distal
gastric antrum (arrowheads, C) from active contraction. Note
similarity of C to Figure
2.
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9C. —Sonographic appearance of contraction of normal gastric
antrum in 34-year-old man without gastric disease. Sequential sonographic
images show progressive luminal narrowing and wall thickening of distal
gastric antrum (arrowheads, C) from active contraction. Note
similarity of C to Figure
2.
|
|
The relatively high prevalence (24%) of linear submucosal low attenuation
or mural striation in the region of antral thickening was initially surprising
to us. The presence of negative attenuation values in a significant percentage
of these cases suggests submucosal fat deposition as the cause of mural
striation. Although the mean antral thickness was slightly greater in patients
with mural striation, the difference was of only borderline significance.
Similar-appearing submucosal fat deposition from inflammatory bowel conditions
has been well documented on CT in the rectum, colon, and small bowel
[15,
16]. Although mural striation
in the gastric antrum on MDCT may represent chronic or subclinical
inflammation, further investigation is needed before firm conclusions can be
drawn about its cause or significance. However, based on our findings, its
presence alone should not raise concern for clinical disease and likely
excludes a significant infiltrative process of the mucosa or submucosa.
Given the high frequency of normal antral wall thickening on MDCT, when
should antral wall thickening be considered abnormal? If a 10-mm threshold is
used, the positive predictive value is likely to be low because we encountered
this value in 5% of our patients without suspected disease. Antral wall
thickness never exceeded 12 mm in our series. This finding suggests that 12 mm
may be a more appropriate threshold. It is vital, however, not to simply apply
a ruler in isolation without considering other factors such as the degree of
luminal distention and morphologic features. Eccentric antral thickening was
seen in only 4% of the patients in our series. Although each case of eccentric
wall thickening was considered smooth, true disease is difficult to exclude if
this appearance persists on separate series. Extensive circumferential
thickening should also be viewed with suspicion if it appears rigid or
irregular, especially if unchanged over time. Beyond these situations, our
findings suggest that irregular or eccentric antral thickening that exceeds 12
mm on CT should be considered abnormal.
Differential diagnostic considerations for abnormal antral wall thickening
include a wide array of inflammatory, neoplastic, infiltrative, and
miscellaneous conditions [17].
Gastritis caused by Helicobacter pylori may be present in nearly 50%
of Americans older than 60 years
[2]. According to one study,
the most common CT manifestation of H. pylori gastritis is prominent
circumferential antral wall thickening, averaging 1.5–2.0 cm
[18]. Other causes of antritis
include nonsteroidal antiinflammatory medications, atypical infections (e.g.,
cytomegalovirus, tuberculosis, and syphilis), perigastric inflammatory
processes such as pancreatitis, caustic ingestion, Zollinger-Ellison syndrome,
inflammatory bowel disease, vasculitis, and chronic granulomatous disease of
childhood
[19–21].
Gastric adenocarcinoma (Fig.
10) and lymphoma represent the most common malignant causes, with
metastatic disease being less common
[21,
22]. Rarely, antral thickening
can result from an infiltrative process, such as sarcoidosis, amyloidosis, or
eosinophilic gastritis. Conversely, some causes of gastric wall thickening
tend to spare the gastric antrum, such as Ménétrier's disease
and gastric varices.
View larger version (164K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10. —Abnormal antral wall thickening in 44-year-old man with
gastric adenocarcinoma. Contrast-enhanced axial MDCT image shows irregular,
eccentric antral wall thickening (arrowheads) with shouldering.
Gastric adenocarcinoma was subsequently proven.
|
|
Our study has several limitations. We did not formally evaluate MDCT scans
in patients with known gastric disease, such as those with symptomatic H.
pylori gastritis. However, the goal of our study was to establish a
normal range of MDCT findings in patients without suspected antral disease to
avoid unnecessary further workup. Although a blinded comparison with abnormal
cases would have been useful, it was beyond the scope of our study. Given the
reportedly high prevalence of H. pylori gastritis, however,
subclinical gastritis as a cause of antral thickening in some patients cannot
be excluded. Another limitation was the lack of endoscopic correlation in most
of the cases. However, this limitation was anticipated because the patients in
our series did not have suspected gastric disease and therefore were also not
tested for H. pylori. We included the normal endoscopic findings in
the small cohort of 10 patients to lend further credence that the antral wall
thickening seen on MDCT in our population was not caused by antral
inflammation or another disease condition. A final limitation to our study was
that the use of positive oral contrast material might have caused gastric wall
measurements to be underestimated in some patients because adjacent
high-attenuation fluid can obscure the enhancing mucosa. This problem is not
encountered when water is used as the oral contrast medium (Fig.
6A,
6B,
6C).
In conclusion, smooth and uniform wall thickening of the distal gastric
antrum relative to the proximal stomach is a normal finding on MDCT. Antral
wall thickening commonly exceeds 5 mm but may measure up to 12 mm. Associated
linear submucosal low attenuation (mural striation) is a relatively common
feature. Our imaging findings, in conjunction with previous anatomic and
physiologic observations, suggest that normal antral wall thickening is likely
caused by an anatomic component (muscular thickening) that can be further
accentuated by dynamic factors (antral contraction and incomplete distention).
In the absence of persistent substantial antral wall irregularity, asymmetry,
or thickening (> 12 mm) on MDCT, recommending further workup is generally
unnecessary in patients without suspected gastric disease.
References
- Berg HM. Antral gastritis. Radiology1952; 59:324
–335
- Dheer S, Levine MS, Redfern RO, Metz DC, Rubesin SE, Laufer I.
Radiographically diagnosed antral gastritis: findings in patients with and
without Helicobacter pylori infection. Br J
Radiol 2002;75:805
–811[Abstract/Free Full Text]
- Balfe DM, Koehler RE, Karstaedt N, Stanley RJ, Sagel SS. Computed
tomography of gastric neoplasms. Radiology1981; 140:431
–436[Abstract/Free Full Text]
- Lee KR, Levine E, Moffat RE, Bigongiari LR, Hermreck AS. Computed
tomographic staging of malignant gastric neoplasms.
Radiology1979; 133:151
–155[Abstract]
- Brown BM, Federle MP, Jeffrey RB. Gastric wall thickening and
extragastric inflammatory processes: a retrospective CT study. J
Comput Assist Tomogr 1982;6:762
–765[Medline]
- Komaki S. Normal or benign gastric wall thickening demonstrated by
computed tomography. J Comput Assist Tomogr1982; 6:1103
–1107[Medline]
- Desai RK, Tagliabue JR, Wegryn SA, Einstein DM. CT evaluation of
wall thickening in the alimentary tract. RadioGraphics1991; 11:771
–783[Abstract]
- Scatarige JC, DiSantis DJ. CT of the stomach and duodenum.
Radiol Clin North Am1989; 27:687
–706[Medline]
- Gossios KJ, Tsianos EV, Demou LL, et al. Use of water or air as
oral contrast media for computed tomographic study of the gastric wall:
comparison of the two techniques. Gastrointest Radiol1991; 16:293
–297[Medline]
- Hammerman AM, Mirowitz SA, Susman N. The gastric air-fluid sign:
aid in CT assessment of gastric wall thickening. Gastrointest
Radiol 1989;14:109
–112[Medline]
- Torgersen J. The muscular build and movements of the stomach and
duodenal bulb. Acta Radiol1942; 45[suppl]:1
–100
- Kelly KA. Motility of the stomach and gastroduodenal junction. In:
Johnson LR, ed. Physiology of the gastrointestinal
tract. New York: Raven, 1981:393
–410
- Kumar D. Gastric motor physiology and pathophysiology. In:
Gustavsson S, Kumar D, Graham DY, eds. The stomach.
London: Churchill Livingstone, 1992:129
–142
- Hinder RA, Kelly KA. Human gastric pacesetter potential: site of
origin, spread, and response to gastric transection and proximal gastric
vagotomy. Am J Surg1977; 133:29
–33[Medline]
- Jones B, Fishman EK, Hamilton SR, et al. Submucosal accumulation of
fat in inflammatory bowel disease: CT-pathologic correlation. J
Comput Assist Tomogr 1986;10:759
–763[Medline]
- Muldowney SM, Balfe DM, Hammerman A, Wick MR. "Acute"
fat deposition in bowel wall submucosa: CT appearance. J Comput
Assist Tomogr 1995;19:390
–393[Medline]
- Pickhardt PJ. The alimentary tract. In: Body CT: a
practical approach. Slone RM, Fisher AJ, Pickhardt PJ, Gutierrez,
FR, Balfe DM, eds. New York: McGraw-Hill, 2000:95
–112
- Urban BA, Fishman EK, Hruban RH. Helicobacter
pylori gastritis mimicking gastric carcinoma at CT evaluation.
Radiology1991; 179:689
–691[Abstract/Free Full Text]
- Farman J, Lerner ME, Ng C, et al. Cytomegalovirus gastritis:
protean radiologic features. Gastrointest Radiol1992; 17:202
–206[Medline]
- Kenney PJ, Brinsko RE, Patel DV. Gastric involvement in chronic
granulomatous disease of childhood: demonstration by computed tomography and
upper gastrointestinal studies. J Comput Assist Tomogr1985; 9:563
–565[Medline]
- Fishman EK, Urban BA, Hruban RH. CT of the stomach: spectrum of
disease. RadioGraphics1996; 16:1035
–1054[Abstract]
- Megibow AJ, Balthazar EJ, Naidich DP, Bosniak MA. Computed
tomography of gastrointestinal lymphoma. AJR1983; 141:541
–547[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
= grade 2, 
= grade 3, 
= grade 4.