Moteki and Horikoshi Abdominal Imaging • Original Research

CT of Acute Appendicitis
Downloaded from www.ajronline.org by 103.255.240.183 on 03/06/20 from IP address 103.255.240.183. Copyright ARRS. For personal use only; all rights reserved

New CT Criterion for Acute

Appendicitis: Maximum Depth of
Intraluminal Appendiceal Fluid
Takao Moteki1 OBJECTIVE. The purpose of this study was to evaluate whether a new criterion—maximum
Hiroyuki Horikoshi2 depth of the intraluminal appendiceal fluid—is useful to differentiate between a normal appendix
with diameter greater than 6 mm and appendicitis without periappendiceal inflammation.
Moteki T, Horikoshi H MATERIALS AND METHODS. The study included 59 patients showing a normal ap-
pendix with diameter greater than 6 mm and having no adjacent lesions (noncomplicated-nor-
mal-appendix group), 30 patients showing a normal appendix with diameter greater than 6 mm
and having adjacent lesions (complicated-normal-appendix group), and 38 patients showing
appendicitis without periappendiceal inflammation (appendicitis group). The following spe-
cific CT findings were retrospectively evaluated: maximum appendiceal diameter greater than
6 mm, maximum appendiceal wall thickness greater than 3 mm, appendiceal wall enhance-
ment, focal cecal wall thickening, adjacent adenopathy, appendicolith, and maximum depth of
the intraluminal appendiceal fluid.
RESULTS. The mean maximum depth of the intraluminal appendiceal fluid in the appendi-
citis group was significantly higher than in the two groups with a normal appendix (Mann-Whitney
U test: p < 0.001). When using maximum depth of the intraluminal appendiceal fluid greater than
2.6 mm for a criterion of appendicitis, sensitivity and specificity for differentiation between the ap-
pendicitis group and the other two groups with a normal appendix were both greater than 80%. In
contrast, when using another CT a criterion, either sensitivity or specificity was 50% or less.
CONCLUSION. The new CT criterion based on the maximum depth of the intraluminal
appendiceal fluid greater than 2.6 mm is particularly useful for differentiating appendicitis with-
out periappendiceal inflammation from a normal appendix with a diameter greater than 6 mm.

ecently, CT has become the im- However, no reports use CT findings to differ-

R aging test of choice for the diag-

Keywords: abdominal imaging, appendicitis, CT,
gastrointestinal imaging
DOI:10.2214/AJR.06.1180
Received September 3, 2006; accepted after revision
December 6, 2006.
1Department of Radiology, Fujioka General Hospital,
942-1 Fujioka, Fujiokashi, Gunma 375-8503, Japan.
Address correspondence to T. Moteki
(pwd8ja227h@md.point.ne.jp).
2Departmentof Radiology, Gunma Cancer Tomo Hospital,
Gunma, Japan.
AJR 2007; 188:1313–1319
0361–803X/07/1885–1313
© American Roentgen Ray Society
nosis of appendicitis. Sensitivity
and specificity range from 94%
to 98% based on findings of a thickened ap-
pendix with some degree of adjacent in-
flammation [1–7]. The use of CT has led to
a substantial decrease in the rate of unnec-
essary appendectomies and a concomitant
decrease in the perforation rate [8, 9]. The
main CT criteria for acute appendicitis are
periappendiceal inflammatory changes, an
appendix with a diameter of more than 6
mm, or a wall thickness of more than 3 mm.
However, we often encounter a normal ap-
pendix with a diameter of more than 6 mm
and sometimes encounter a normal appen-
dix with a wall thickness of more than 3
mm. Conversely, we sometimes see patients
with appendicitis without periappendiceal
inflammation despite a high clinical suspi-
cion of appendicitis.
entiate between a normal appendix with a diam-
eter of more than 6 mm and appendicitis without
periappendiceal inflammation. If CT differentia-
tion between such cases is possible, CT may be
useful to evaluate patients who present equivocal
symptoms of appendicitis, who present clinical
signs of appendicitis with little adipose tissue in
the abdomen, or who present with vague abdom-
inal pain while showing an appendix with a di-
ameter of more than 6 mm.
Hence, we undertook this study to evaluate
whether differences exist in CT findings be-
tween appendicitis without periappendiceal
inflammation and a normal appendix with a
diameter of more than 6 mm. In addition to
classic CT findings for the diagnosis of ap-
pendicitis, we also evaluated the finding of in-
traluminal appendiceal fluid because we have
the impression that cases of appendicitis tend
to show more intraluminal appendiceal fluid
than cases with a normal appendix.

AJR:188, May 2007 1313


Materials and Methods
The medical records or CT reports of all patients
in whom abdominal and pelvic CT examinations
were performed for suspected acute appendicitis
over a 14-month period (from October 2003 through
Downloaded from www.ajronline.org by 103.255.240.183 on 03/06/20 from IP address 103.255.240.183. Copyright ARRS. For personal use only; all rights reserved
November 2004) in our institution were retrospec-
tively reviewed. The institutional review board ap-
proved use of the CT database and medical records;
informed specific study-related consent was waived.
Consecutive patients (n = 339) were referred by the
attending doctors for CT when the clinical features
were suspicious for acute appendicitis or difficult to
differentiate between appendicitis and another in-
flammatory disease (such as diverticulitis, enteritis,
and pelvic inflammatory disease).
In 98 patients, periappendiceal inflammatory
changes (e.g., stranding, thickening of the lateral
conal fascia, phlegmon, or abscess) were detected
and diagnosed as appendicitis; 82 of these patients
underwent surgery, and 16 patients were conserva-
tively treated. In 130 patients, diseases other than ap-
pendicitis were detected on CT: 31 (23.8%) cases of
diverticulitis; 21 (16.2%) cases of enteritis; seven
(5.4%) cases of colon cancer, metastatic tumors, or
lymphoma in the right lower abdomen; 28 (21.5%)
cases of gynecologic pathology (e.g., ovarian tu-
mors, pelvic inflammatory disease); 13 (10%) cases
of urolithiasis; 12 (9.2%) cases of pyelonephritis
(complicating urolithiasis in eight cases); and 18
(13.8%) cases of acute cholecystitis. In 46 patients,
no obvious lesions were seen on CT, and their symp-
toms subsequently resolved. Fifteen patients were
lost to follow-up after CT examination.
Twelve patients with insufficient clinical presen-
tation for the diagnosis of acute appendicitis were
conservatively treated after CT failed to detect peri-
appendiceal inflammation. Sixteen patients without
periappendiceal inflammation on CT were proven to
have appendicitis at surgery that was performed be-
cause of worsening clinical symptoms (operations
were performed 10–31 hours [mean, 16.5 hours] af-
ter CT) and were included in the appendicitis group.
These patients did not have repeated CT examina-
tions before their operations. Twenty-two patients
without periappendiceal inflammation on CT were
also included in the appendicitis group. In these pa-
tients, the clinical symptoms were considered suffi-
cient for a diagnosis of acute appendicitis (i.e., ab-
dominal pain of recent onset that was initially
periumbilical and then localized to a point in the
right iliac fossa [such as the McBurney point] and
abdominal pain associated with vomiting, fever, and
an elevated WBC) [6, 10]; these symptoms resolved
after the administration of antibiotics.
Excluding cases of appendicitis and clinically
suspected appendicitis, we identified cases in
which a whole-length normal appendix was delin-
eated and the appendix showed a diameter of more
1314
Moteki and Horikoshi
than 6 mm on initial CT interpretation over a 14-
month period (October 2003 to November 2004). A
normal appendix was identified in 68.3%
(1,038/1,520) of the cases studied, and 132 cases
satisfying these all conditions were assigned to a
temporary-normal-appendix group. From this
group, we excluded cases in which any patients had
leukocytosis or showed pain, tumors, or inflamma-
tory lesions in the right lower abdomen. We also ex-
cluded repetitively performed CT examinations in a
patient except the representative examination.
We assigned these remaining 59 cases to the non-
complicated-normal-appendix group. From the tem-
porary-normal-appendix group, we also picked up
CT examinations in which patients had lesions at or
along the ileocecal region (except for appendicitis).
We assigned these cases (n = 30) to the complicated-
normal-appendix group. These patients consisted of
11 (36.7%) cases of diverticulitis, two (6.7%) cases
of pelvic inflammatory disease, six (20%) cases of
enteritis, two (6.7%) cases of abscess, one (3.3%)
case of pancreatitis, three (10%) cases of peritonitis,
two (6.7%) cases of ileus, and three (10%) cases of
cecal tumors (cancer, lymphoma, and metastasis
from sigmoid colon cancer).
CT images were obtained using a 4-MDCT
scanner (LightSpeed Plus, GE Healthcare). For pa-
tients who were suspected of having appendicitis,
diverticulitis, and pelvic inflammatory disease,
scanning was performed from the top of the liver to
the symphysis pubis with 120 or 140 kVp using
Auto mA (a function that automatically controls
tube current so as to stabilize image quality).
Table feed was 3.75 mm/0.5 s of scanner rotation (a
pitch of 0.75:1.0). From the raw data of each acqui-
sition, 7-mm-thick transverse sections were recon-
structed from the diaphragmatic dome to the iliac
crest, and 3.5- or 2.5-mm-thick transverse sections
of the pelvis were reconstructed.
For these patients, 2 mg/kg (when the patients
were < 50 kg in body weight) or 100 mL of IV con-
trast material (Iopamiron 300 Syringe [iopamidol],
Nippon Schering) was administered at a rate of 0.04
mL/s/kg of body weight (in patients < 50 kg) or 2
mL/s with a scanning delay of 65 seconds. This CT
protocol (the CT protocol for appendicitis) was per-
formed in all 38 (100%) patients in the appendicitis
group, eight (26.7%) patients in the complicated-
normal-appendix group, and two (3.4%) patients in
the noncomplicated-normal-appendix group. For
other patients, fine 3.5- or 2.5-mm-thick transverse
sections at the appendiceal level were also obtained
in 18 (60%) patients in the complicated-normal-ap-
pendix group and 22 (37.3%) patients in the non-
complicated-normal-appendix group. For the re-
maining patients, 5- or 7-mm-thick transverse
sections at the appendiceal level were obtained. Ex-
cept for patients who were scanned with the CT
protocol for appendicitis, injection speed and scan-
ning delay of IV contrast medium varied depending
on the clinical history provided. No patient received
oral or rectal contrast material.
Interpretation of CT images was mainly based on
the positive and negative criteria described by Rao et
al. [11] and Rao and Mueller [12]. The following
conventional specific CT findings were evaluated:
maximum appendiceal diameter, maximum appen-
diceal wall thickness, presence or absence of appen-
diceal wall enhancement, presence or absence of fo-
cal cecal wall thickening, presence or absence of
adjacent adenopathy, and presence or absence of ap-
pendicolith. In addition, we also evaluated the max-
imum depth of the intraluminal appendiceal fluid.
We judged appendiceal wall enhancement as posi-
tive when the appendiceal wall showed focally or
diffusely higher density than the surrounding small
intestinal or colon walls on enhanced CT. We did not
necessarily use the same appendiceal CT image for
measurement of maximum appendiceal wall thick-
ness as the image for measurement of maximum ap-
pendiceal diameter because optimal appendiceal CT
images for these criteria were not typically identical,
especially when an appendix was focally expanded
with intraluminal contents (such as intraluminal
fluid, gas, appendicolith, or barium).
When the appendiceal lumen was collapsed at
the image for measurement of maximum appen-
diceal wall thickness, we adopted half of diameter
of the appendix for this criterion. To evaluate the in-
traluminal appendiceal fluid, we did not use maxi-
mum width of the intraluminal appendiceal fluid
but instead used maximum depth (anteroposterior
diameter) of the intraluminal appendiceal fluid be-
cause maximum depth more closely approximates
the amount of the intraluminal appendiceal fluid
than does maximum width when an air–fluid level
is seen in the appendiceal lumen to be measured.
We obtained the maximum depth of the intralumi-
nal appendiceal fluid by measuring the maximum
anteroposterior diameter of the intraluminal con-
tent that showed lower density than the appendiceal
wall on enhanced CT images, excluding appendi-
colith, barium, and intraluminal gas.
Evaluation of CT images and measurements of
the appendix were performed on a workstation
(Centricity RA 1000, GE Healthcare) using an elec-
tronic ruler. With no knowledge of the clinical
course of the patients, a single, board-certified ra-
diologist with 15 years of experience in radiology
performed all the measurements for these CT crite-
ria for appendicitis and interpreted whether an ap-
pendix satisfied each CT criteria, except for the
equivocal cases. In equivocal cases, a decision was
made on the basis of the consensus of the same ra-
diologist and another radiologist with 15 years of
experience in radiology.
AJR:188, May 2007
 CT of Acute Appendicitis
Downloaded from www.ajronline.org by 103.255.240.183 on 03/06/20 from IP address 103.255.240.183. Copyright ARRS. For personal use only; all rights reserved

Fig. 1—34-year-old woman with appendicitis without periappendiceal inflammation
(surgically proven appendicitis case in appendicitis group). Contrast-enhanced CT
image at cecal level reveals that this appendicitis case (arrow) satisfies only two CT
criteria for appendicitis: maximum appendiceal diameter greater than 6 mm and
maximum depth of intraluminal appendiceal fluid greater than 2.6 mm.
Fig. 2—46-year-old man with rectal cancer and normal appendix (a case in
noncomplicated-normal-appendix group). Contrast-enhanced CT scan at cecal level
shows no significant intraluminal appendiceal fluid (arrow) despite satisfying some
CT criteria for appendicitis: maximum appendiceal diameter greater than 6 mm,
maximum appendiceal wall thickness greater than 3 mm, and presence of
appendiceal wall enhancement (arrowhead).

For each CT criterion, a comparison between
groups (appendicitis group vs noncomplicated- and
complicated-normal-appendix groups) was per-
formed using the Mann-Whitney U test. A p value
less than 0.05 was considered statistically signifi-
cant. For maximum appendiceal wall thickness and
maximum depth of the intraluminal appendiceal
fluid, receiver operating characteristic (ROC)
curves were constructed to depict the curves of
those features and to obtain cutoff values for the
maximum depth of the intraluminal appendiceal
fluid that best differentiated the appendicitis group
from the two normal-appendix groups.
Sensitivity and specificity for differentiation of
the appendicitis group from the two normal-appen-
dix groups were also calculated using each criterion.
For these analyses, a 6-mm maximum appendiceal
diameter, a 3-mm maximum appendiceal wall thick-
ness, and the cutoff value obtained in ROC analysis
for the maximum depth of the intraluminal appen-
diceal fluid were used as the cutoff values.
Results
The appendicitis group (Fig. 1) consisted
of 38 patients, ranging in age from 9 to 77
years (mean ± SD, 31 ± 18 years). There were
17 male patients and 21 female patients. In 16
patients, appendicitis was confirmed on the
basis of the surgical and pathologic outcome.
In 22 patients, appendicitis was clinically
confirmed. A paucity of periappendiceal fat
that might prevent periappendiceal inflamma-
tion was shown in nine patients in the appen-
dicitis group.
The noncomplicated-normal-appendix group
(Fig. 2) consisted of 59 patients, ranging in age
from 26 to 89 years (59 ± 16 years). There were
35 men and 24 women in this group. The com-
plicated-normal-appendix group (Figs. 3 and 4)
consisted of 30 patients ranging in age from 18
to 86 years (51 ± 19 years). There were 18 men
and 12 women in this group.
The number of cases satisfying each CT
criterion for appendicitis in these groups is
shown in Table 1. As the appendicitis group
in the table, the subgroup consisting of oper-
atively proven cases is also included. The cri-
teria of maximum appendiceal diameter of
more than 6 mm and maximum appendiceal
wall thickness of more than 3 mm were both
satisfied by 21 cases in the noncomplicated-
normal-appendix group (Fig. 2), 12 in the
complicated-normal-appendix group (Figs. 3
and 4), and 19 in the appendicitis group.
In comparison with the noncomplicated-
and complicated-normal-appendix groups,
the appendicitis group was not significantly
different with respect to the mean maximum
appendiceal diameter, mean maximum ap-
pendiceal wall thickness, number having ap-
pendiceal wall enhancement, and number of
appendicoliths (p > 0.05, Mann-Whitney U
test). The number of cases of focal cecal wall
thickening in the appendicitis group was less
than in the complicated-normal-appendix
group (p < 0.001) and was not significantly
different from the noncomplicated-normal-
appendix group (p > 0.05). The number of
cases of adjacent adenopathy in the appendi-
citis group was higher than in the noncompli-
cated-normal-appendix group (p < 0.05) and
was not significantly different from the com-
plicated-normal-appendix group (p > 0.05).
The mean maximum depth of the intraluminal
appendiceal fluid in the appendicitis group
was significantly greater than in the noncom-
plicated- and complicated-normal-appendix
groups (p < 0.001).
When ROC analyses were performed on
maximum depth of the intraluminal appen-
diceal fluid and maximum appendiceal wall

AJR:188, May 2007 1315

 Moteki and Horikoshi
Downloaded from www.ajronline.org by 103.255.240.183 on 03/06/20 from IP address 103.255.240.183. Copyright ARRS. For personal use only; all rights reserved

Fig. 3—44-year-old man with normal appendix with complication of enteritis (a case
in complicated-normal-appendix group). Contrast-enhanced CT scan at cecal level
shows no significant intraluminal appendiceal fluid despite satisfying some CT
criteria for appendicitis: maximum appendiceal diameter greater than 6 mm,
maximum appendiceal wall thickness greater than 3 mm (arrow), and presence of
focal cecal wall thickening (arrowhead).
Fig. 4—82-year-old man with normal appendix with complication of enteritis (a case in
complicated-normal-appendix group). Contrast-enhanced CT scan at cecal level shows
fluid collection in appendix (maximum depth of intraluminal appendiceal fluid greater
than 2.6 mm) (arrow), presence of appendiceal wall enhancement, and maximum
appendiceal diameter greater than 6 mm. In addition, large fluid collection in cecum (28
mm in depth) (arrowheads) is also seen. This intraluminal appendiceal fluid collection
may be caused by hydrostatic pressure of prominent intraluminal fluid of cecum.

TABLE 1: Diagnostic Criteria for Appendicitis and Rate of Occurrence

Rate of Occurrence (%) and Number (in parentheses)
Appendicitis Groupa
Surgically Proven Noncomplicated- Complicated-
All Cases Cases Normal-Appendix Groupb Normal-Appendix Groupc
Criterion (n = 38) (n = 16) (n = 59) (n = 30)
Maximum appendiceal diameter > 6 mm 86.8 (33) 81.3 (13) 100 (59) 100 (30)
Maximum appendiceal wall thickness > 3 mm 50.0 (19) 43.8 (7) 35.6 (21) 40.0 (12)
Appendiceal wall enhancement present 21.1 (8) 25.0 (4) 16.9 (10) 30.0 (9)
Focal cecal wall thickening present 7.9 (3) 12.5 (2) 0 (0) 60.0 (18)d
Ambient adenopathy present 18.4 (7) 12.5 (2) 3.4 (2)d 36.7 (11)
Appendicolith present 15.8 (6) 18.8 (3) 16.9 (10) 10.0 (3)
Maximum depth of intraluminal appendiceal fluid > 2.6 mm 86.8 (33) 93.8 (15) 3.4 (2)d 20.0 (6)d
a Patients with appendicitis showing no periappendiceal inflammation.
b Patients with a normal appendix > 6 mm in maximum diameter and no adjacent lesions.
c Patients with a normal appendix > 6 mm in maximum diameter and adjacent lesions.
d Mean value or incidence in these two normal-appendix groups was statistically different (p < 0.05; Mann-Whitney U test) from those in appendicitis group.

thickness comparing the appendicitis group
with the noncomplicated- and complicated-
normal-appendix groups (Fig. 5), the areas
under the ROC curves for maximum depth of
the intraluminal appendiceal fluid (0.949 ±
0.026 [SE] and 0.863 ± 0.050, respectively)
were significantly larger than the areas for
maximum appendiceal wall thickness (0.603 ±
0.059 and 0.528 ± 0.072, respectively) (p <
0.001). These results indicate that maximum
depth of the intraluminal appendiceal fluid is
more helpful for the differentiation of appen-
dicitis from normal appendix than maximum
appendiceal wall thickness.
The cutoff value of maximum depth of the
intraluminal appendiceal fluid for the best
differentiation between the appendicitis
group and the other two groups with a normal
appendix was 2.6 mm (ROC analysis). Using
a cutoff value of 2.6 mm of maximum depth
of the intraluminal appendiceal fluid for dif-

1316 AJR:188, May 2007

 CT of Acute Appendicitis

1.0 1.0

0.8 0.8
Downloaded from www.ajronline.org by 103.255.240.183 on 03/06/20 from IP address 103.255.240.183. Copyright ARRS. For personal use only; all rights reserved

True-Positive Fraction

True-Positive Fraction
0.6 0.6

0.4 0.4

0.2 0.2

0.0 0.0

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
False-Positive Fraction False-Positive Fraction

A B
Fig. 5—Receiver operating characteristic curves for wall thickness and intraluminal fluid.
A and B, Graphs compare maximum appendiceal wall thickness (solid lines) and maximum depth of intraluminal appendiceal fluid (dashed lines) of appendicitis group versus
noncomplicated-normal-appendix group (A) and of appendicitis group versus complicated-normal-appendix group (B).

TABLE 2: Sensitivity and Specificity for Differentiating Between the Appendicitis Group and the Two Normal-Appendix
Groups
Specificity Between Specificity Between
Sensitivity Between Appendicitis Group and Appendicitis Group and
Appendicitis Groupa and Noncomplicated- Complicated-Normal-
Two Normal-Appendix Groups Normal-Appendix Groupb Appendix Groupc
Criterion % No. % No. % No.
Maximum appendiceal diameter > 6 mm 86.8 33/38 35.9 33/92 52.4 33/63
Maximum appendiceal wall thickness > 3 mm 50.0 19/38 47.5 19/40 61.3 19/31
Appendiceal wall enhancement present 21.1 8/38 44.4 8/18 47.1 8/17
Focal cecal wall thickening present 7.9 3/38 100 3/3 14.3 3/21
Ambient adenopathy present 18.4 7/38 77.8 7/9 38.9 7/18
Appendicolith present 15.8 6/38 37.5 6/16 66.7 6/9
Maximum depth of intraluminal appendiceal fluid > 2.6 mm 86.8 33/38 94.3 33/35 84.6 33/39
a Patients with appendicitis showing no periappendiceal inflammation.
b Patients with a normal appendix > 6 mm in maximum diameter and no adjacent lesions.
c Patients with a normal appendix > 6 mm in maximum diameter and adjacent lesions.

ferentiation between the appendicitis group
and the noncomplicated-normal-appendix
group, the sensitivity and specificity were
86.8% (33/38) and 94.3% (33/35), respec-
tively. Using this same cutoff value for differ-
entiation between the appendicitis group and
the complicated-normal-appendix group, the
sensitivity and specificity were 86.8% (33/38)
and 84.6% (33/39), respectively.
The sensitivity and specificity for differentia-
tion between the appendicitis group and the two
normal-appendix groups, when using other
conventional criteria (maximum appendiceal
diameter > 6 mm, maximum appendiceal wall
thickness > 3 mm, presence of appendiceal wall
enhancement, presence of focal cecal wall thick-
ening, presence of adjacent adenopathy, and
presence of appendicolith) are shown in Table 2.
Discussion
The diagnosis of acute appendicitis has tra-
ditionally relied on history and physical exam-
ination. With recent reports on the accuracy of
CT examination for the diagnosis of appendi-
citis [2, 3, 11, 13, 14], CT has become part of
the standard of care in managing patients with
suspected acute appendicitis. CT signs include
appendiceal diameter of more than 6 mm, an
appendicolith, an appendiceal wall thickness
of more than 3 mm, periappendiceal inflam-
matory changes (adjacent or periappendiceal
fat stranding, fluid collections, phlegmon, or
abscess formation), extraluminal air, adjacent

AJR:188, May 2007 1317


adenopathy, adjacent bowel wall thickening,
and focal cecal wall thickening. Although a
substantial body of literature documents the
excellent accuracy of CT for the diagnosis of
appendicitis, that accuracy is mainly caused by
Downloaded from www.ajronline.org by 103.255.240.183 on 03/06/20 from IP address 103.255.240.183. Copyright ARRS. For personal use only; all rights reserved
the presence of periappendiceal inflammation.
Our results show that the incidence of a
normal appendix with maximum appendiceal
diameter of more than 6 mm is much higher
than the incidence of appendicitis showing no
significant periappendiceal inflammation.
Furthermore, the incidence of a normal ap-
pendix satisfying two major conventional CT
criteria for appendicitis (maximum appen-
diceal diameter of > 6 mm and maximum ap-
pendiceal wall thickness of > 3 mm) ex-
ceeded the incidence of appendicitis showing
no significant periappendiceal inflammation.
The incidences of appendiceal wall enhance-
ment, appendicolith, adjacent adenopathy, or
focal cecal wall thickening in the appendicitis
group were too low for clinical use for differ-
entiation between a normal appendix with
maximum diameter of more than 6 mm and
appendicitis without periappendiceal inflam-
mation and were not significantly higher than
the incidences of these criteria in the compli-
cated-normal-appendix group.
These results suggest that no reliable con-
ventional specific CT criteria exist for the dif-
ferentiation between appendicitis and a nor-
mal appendix when an appendix shows a
diameter of more than 6 mm and no signifi-
cant periappendiceal inflammation. Hence,
another specific CT criterion is needed to per-
form such differentiation.
The pathophysiology of appendicitis is most
likely the result of a closed-loop obstruction of
its lumen related to a fecalith or resulting from
hyperplasia of submucosal lymphoid follicles
caused by viral or bacterial infection. In the
presence of obstruction, the mucosa continues
to secrete, resulting in an accumulation of mu-
coid material and increasing intraluminal pres-
sure. Bacteria located in the lumen of the ap-
pendix proliferate in the presence of stasis and
obstruction. Continued mucus production and
proliferation of bacteria cause a further rise of
intraluminal pressure, resulting in the develop-
ment of acute appendicitis with edema, lym-
phatic obstruction, and necrotizing ulceration
of the mucosa [15, 16]. In other words, intralu-
minal accumulation of mucoid material (in-
traluminal fluid collection) is considered to be
closely related to appendiceal obstruction that
could cause appendicitis.
However, no reports have evaluated the
depth or width of the intraluminal appendiceal
1318
Moteki and Horikoshi
fluid for differentiation between appendicitis
and a normal appendix, except for a partially
relevant report that stated that the graded-com-
pression technique on sonography might be
useful to differentiate appendicitis from a nor-
mal appendix because the intraluminal ob-
struction that causes appendicitis also causes
the intraluminal content (including the intralu-
minal appendiceal fluid) to remain despite
compression [17].
In our study, the mean maximum depth of
the intraluminal appendiceal fluid in the ap-
pendicitis group was significantly higher than
in the other two groups with a normal appen-
dix. We consider this result to be reasonably
explained by the preceding theory; if appen-
diceal obstruction is absent, a normal appendix
might show a lower depth of the intraluminal
appendiceal fluid than appendicitis (regardless
of whether complicated lesions are present at
or along the ileocecal region). Furthermore,
when using a cutoff value of 2.6 mm for max-
imum depth of the intraluminal appendiceal
fluid to differentiate between appendicitis
without periappendiceal inflammation and a
normal appendix with maximum appendiceal
diameter of more than 6 mm, the sensitivity
and specificity were more than 80%. In con-
trast, conventional CT criteria had either sensi-
tivity or specificity of 50% or less for differen-
tiating these conditions (Table 2).
From these results, we believe that maxi-
mum depth of the intraluminal appendiceal
fluid of more than 2.6 mm (Fig. 1) is more use-
ful than conventional CT criteria for differenti-
ating appendicitis without periappendiceal in-
flammation from a normal appendix with a
maximum diameter of more than 6 mm, re-
gardless of whether complicated lesions are
present at or along the ileocecal region
(Table 1, Figs. 1–3). Therefore, maximum
depth of the intraluminal appendiceal fluid of
more than 2.6 mm should become a useful,
specific diagnostic criterion for acute appendi-
citis, especially when periappendiceal inflam-
matory changes are absent (Fig. 1) or unclear
because of poor periappendiceal fat.
In the complicated-normal-appendix group,
six cases satisfied the criterion of maximum
depth of the intraluminal appendiceal fluid of
more than 2.6 mm. Three of these cases (com-
plicated by enteritis in two cases and cecal di-
verticulitis in one case) were also associated
with large fluid collection in the cecum (> 20
mm in depth) (Fig. 4). In these cases, we be-
lieve that a larger hydrostatic pressure that is
generated from the larger cecal fluid collec-
tion might cause accumulation of more in-
traluminal appendiceal fluid. In one of these
cases (complicated by cecal cancer), the ori-
fice of the appendix was obstructed by the tu-
mor. Hence, the depth of the intraluminal ap-
pendiceal fluid may not be helpful to
differentiate between appendicitis and a nor-
mal appendix when the cecum contains a
larger amount of intraluminal fluid or when a
closed-loop obstruction of the appendix is
caused by a cecal tumor.
A limitation of our results is that the appen-
dicitis group included many cases that were
not surgically proven, although these cases
showed a certified clinical course. Thus, the
presence of a few cases of normal appendix in
the appendicitis group would reduce the diag-
nostic sensitivity of both conventional CT cri-
teria and our new criterion based on maxi-
mum depth of the intraluminal appendiceal
fluid. However, we consider that this limita-
tion does not negate the usefulness of the new
criterion because the subgroup of the appen-
dicitis group consisting of the surgically
proven cases satisfied this criterion with a
higher incidence (93.8%) than the appendici-
tis group (86.8%), as shown in Table 1.
Another limitation of our data is that our
evaluation between the appendicitis group and
the two groups with a normal appendix was
performed under the condition that two major
criteria (the presence of periappendiceal in-
flammation and maximum appendiceal diame-
ter of > 6 mm) were inefficient. Hence, we can-
not determine whether maximum depth of the
intraluminal appendiceal fluid of more than 2.6
mm is useful to differentiate between appendi-
citis and a normal appendix in comparison with
these two major criteria. However, our results
show that the maximum depth of the intralumi-
nal appendiceal fluid of more than 2.6 mm is
more useful than the other conventional CT cri-
teria, including the criterion of maximum ap-
pendiceal wall thickness of more than 3 mm.
The last limitation of our study is that an
optimal CT protocol for the diagnosis of ap-
pendicitis was principally performed in clini-
cally suspected cases of appendicitis. Thus,
all cases in the appendicitis group were
scanned with optimal enhancement and opti-
mal fine sections for evaluation of the appen-
dix, which differed from the other two groups
with a normal appendix, in which many cases
were not scanned with optimal enhancement
or optimal fine sections for evaluating an ap-
pendix. Although such optimal imaging for
evaluation of an appendix might be advanta-
geous to detect appendiceal wall enhance-
ment in the appendicitis group, the incidence
AJR:188, May 2007

of cases satisfying this CT criterion in the ap-
pendicitis group was still low and showed no
significant difference compared with the inci-
dence in the other two groups with a normal
appendix. With other CT criteria, errors of
Downloaded from www.ajronline.org by 103.255.240.183 on 03/06/20 from IP address 103.255.240.183. Copyright ARRS. For personal use only; all rights reserved
measurements and assessment of specific CT
findings in the groups with a normal appendix
might be somewhat larger than in the appen-
dicitis group. However, we consider these er-
rors to be small because only cases in which
the whole-length appendix was clearly delin-
eated were assigned to the groups having a
normal appendix, and the complicated-nor-
mal-appendix group contains cases with fine
3.5- or 2.5-mm-thick appendiceal sections at
the high rate of 86.7% (26/30). Furthermore,
in general, such errors in the groups with a
normal appendix would tend to obscure dif-
ferences between these patients and those in
the appendicitis group. Hence, the differences
in mean maximum depth of the intraluminal
appendiceal fluid between the appendicitis
group and the groups with a normal appendix
are considered to be still robust. However, this
limitation may be completely resolved if a 64-
MDCT scanner were used and fine sections
were routinely obtained.
In conclusion, when an appendix with a di-
ameter of more than 6 mm has no periappen-
diceal inflammation, conventional CT criteria
have limited efficacy in differentiating appen-
dicitis from a normal appendix. However, the
new CT criterion based on a maximum depth
AJR:188, May 2007
CT of Acute Appendicitis
of the intraluminal appendiceal fluid of more
than 2.6 mm is helpful in this differentiation.
References
1. Rao PM, Rhea JT, Novelline RA. Sensitivity and
specificity of the individual CT signs of appendicitis:
experience with 200 helical appendiceal CT exami-
nations. J Comput Assist Tomogr 1997; 21:686–692
2. Rao PM, Rhea JT, Novelline RA, Mostafavi AA,
Lawrason JN, McCabe CJ. Helical CT combined
with contrast material administered only through
the colon for imaging of suspected appendicitis.
AJR 1997; 169:1275–1280
3. Balthazar EJ, Megibow AJ, Siegel SE, Birnbaum
BA. Appendicitis: prospective evaluation with
high-resolution CT. Radiology 1991; 180:21–24
4. Weltman DI, Yu J, Krumenacker J Jr, Huang S, Moh
P. Diagnosis of acute appendicitis: comparison of 5-
and 10-mm CT sections in the same patient. Radi-
ology 2000; 216:172–177
5. Lane MJ, Katz DS, Ross BA, Clautice-Engle TL,
Mindelzun RE, Jeffrey RB Jr. Unenhanced heli-
cal CT for suspected acute appendicitis. AJR
1997;168:405–409
6. Wijetunga R, Tan BS, Rouse JC, Bigg-Wither GW,
Doust BD. Diagnostic accuracy of focused appen-
diceal CT in clinically equivocal cases of acute ap-
pendicitis. Radiology 2001; 221:747–753
7. Mullins ME, Kircher MF, Ryan DP, et al. Evaluation
of suspected appendicitis in children using limited
helical CT and colonic contrast material. AJR 2001;
176:37–41
8. Balthazar EJ, Rofsky NM, Zucker R. Appendicitis:
the impact of computed tomography imaging on
negative appendectomy and perforation rates. Am J
Gastroenterol 1998; 93:768–771
9. Rao PM, Rhea JT, Rattner DW, Venus LG, Novel-
line RA. Introduction of appendiceal CT: impact on
negative appendectomy and appendiceal perfora-
tion rates. Ann Surg 1999; 229:344–349
10. Berry J Jr, Malt RA. Appendicitis near its centenary.
Ann Surg 1984; 200:567–575
11. Rao PM, Rhea JT, Novelline RA, et al. Helical CT
technique for the diagnosis of appendicitis: pro-
spective evaluation of a focused appendix CT ex-
amination. Radiology 1997; 202:139–144
12. Rao PM, Mueller PR. Clinical and pathologic vari-
ants of appendiceal disease: CT features. AJR 1998;
170:1335–1340
13. Malone AJ Jr, Wolf CR, Malmed AS, Melliere BF.
Diagnosis of acute appendicitis: value of unen-
hanced CT. AJR 1993; 160:763–766
14. Balthazar EJ, Birnbaum BA, Yee J, Megibow AJ,
Roshkow J, Gray C. Acute appendicitis: CT and
US correlation in 100 patients. Radiology 1994;
190:31–35
15. Old JL, Dusing RW, Yap W, Dirks J. Imaging for
suspected appendicitis. Am Fam Physician 2005;
71:71–78
16. Graffeo CS, Counselman FL. Appendicitis. Emerg
Med Clin North Am 1996; 14:653–671
17. Kessler N, Cyteval C, Gallix B, et al. Appendicitis:
evaluation of sensitivity, specificity, and predictive
values of US, Doppler US, and laboratory findings.
Radiology 2004; 230:472–478
1319