Downloaded from www.ajronline.org by Washington State Univ. on 10/03/15 from IP address 134.121.47.100. Copyright ARRS.

For personal use only; all rights reserved

Helical CT with Sagittal and

Coronal Reconstructions:
Accuracy for Detection of
Diaphragmatic Injury
Anna R. Larici 1 OBJECTIVE. The objectives of our study were to determine the accuracy of single-detector
Michael B. Gotway 2,3 helical CT (including coronal and sagittal reconstructions) for the diagnosis of traumatic diaphrag-
Harold I. Litt 2 matic injury, establish measurements for the thickness of the normal diaphragmatic crus, and de-
Gautham P. Reddy 2 scribe an additional sign of diaphragmatic injury: active arterial extravasation of contrast material
at the level of the diaphragm.
W. Richard Webb 2
MATERIALS AND METHODS. The CT scans of 25 patients with surgically proven dia-
Carol A. Gotway 4
phragmatic injury and 22 patients with surgically confirmed uninjured diaphragms were blindly re-
Samuel K. Dawn 2,3 viewed by five thoracic radiologists. Sagittal and coronal reconstructions were performed for 20 of
Shelley R. Marder 2,3 the 25 patients with a proven diaphragmatic injury and for all the patients without a diaphragmatic in-
Maria Luigia Storto 1 jury. Scans were evaluated for findings suggestive of diaphragmatic injury and for associated injuries.
Reviewers scored the usefulness of the reconstructed images for establishing the final diagnosis.
Measurements of the right and left crura were performed to establish a threshold measurement that
would enable radiologists to discriminate between a normal diaphragm and an injured diaphragm.
RESULTS. The sensitivity, specificity, positive predictive value, negative predictive value,
and accuracy of helical CT were 84%, 77%, 81%, 81%, and 83%, respectively. Scans showing
active arterial extravasation of contrast material enabled reviewers to correctly identify diaphrag-
matic injury in two patients. Reconstructed images confirmed the correct diagnosis in three pa-
tients but supported an incorrect diagnosis in two. The mean thickness of the diaphragmatic crura
(right and left) was not significantly greater in patients with an injured diaphragm than in those
with an uninjured diaphragm.
CONCLUSION. Helical CT shows good sensitivity, specificity, and accuracy for the diag-
nosis of diaphragmatic injury. Coronal and sagittal reconstructions are of limited use in establish-
ing or refuting this diagnosis. Active arterial extravasation of contrast material near the diaphragm
should raise suspicion for injury. Crus measurements cannot be used to reliably distinguish be-
tween injured and uninjured diaphragms.

Received January 4, 2002; accepted after revision

February 11, 2002.
1
Department of Radiology, SS Annunziata Hospital,
University of G. d’Annunzio, Via P. Valignani 1, Chieti
H elical CT is widely used to screen
patients for thoracic and visceral
abdominal injuries sustained from
blunt or penetrating trauma. Although diaphrag-
CT for the diagnosis of acute diaphragmatic in-
jury [9–11]; a wide range of sensitivities and
specificities have been reported. Only a few
studies have evaluated the accuracy of single-
66100, Italy. matic injuries are present in only approximately detector helical CT for the diagnosis of dia-
2
Department of Radiology, Rm. M-391, University of California, 5–6% of patients with blunt thoracoabdominal phragmatic injury [2, 3, 12–14]. These studies
505 Parnassus Ave., San Francisco, CA 94143-0628. trauma [1–3] and 10% of patients with penetrat- focused primarily on the helical CT evaluation
3
Department of Radiology, Thoracic Imaging Section, ing [4] thoracoabdominal trauma, diaphrag- of patients with blunt trauma [2, 3, 12], and lit-
Rm. 1X 55A, Box 1325, San Francisco General Hospital,
1001 Potrero Ave., San Francisco, CA 94110. Address
matic injuries are commonly associated with tle data specifically examining the use of heli-
correspondence to M. B. Gotway. other torso injuries and can contribute to signifi- cal CT for diagnosing penetrating traumatic
4
Centers for Disease Control and Prevention, National cant morbidity and mortality if not promptly di- injury to the diaphragm exist.
Center for Environmental Health Centers for Disease agnosed [2, 3, 5–8]. Although traditionally many patients with
Control and Prevention, MS E70, 1600 Clifton Rd., N.E., The normal diaphragm is a thin dome- penetrating injuries are taken directly to sur-
Atlanta, GA 30323.
shaped structure, a portion of which is oriented gery without undergoing preoperative imag-
AJR 2002;179:451–457
in the axial plane, which can make the imaging ing, there is a growing trend toward more
0361–803X/02/1792–451 of diaphragmatic injury challenging [8]. Several conservative treatment of such patients [15–
© American Roentgen Ray Society studies have examined the use of conventional 17]. Therefore, CT is playing an increasingly

AJR:179, August 2002 451


important role in the assessment of penetrating
thoracoabdominal injuries [15, 16]. Further-
more, although several reports [12, 14] have
suggested that sagittal and coronal reconstruc-
tions have proven useful in confirming the diag-
Downloaded from www.ajronline.org by Washington State Univ. on 10/03/15 from IP address 134.121.47.100. Copyright ARRS. For personal use only; all rights reserved
nosis of diaphragmatic injury, little data are
available that specifically evaluate the use of
these techniques in trauma cases. The purposes
of our investigation were to determine the accu-
racy of helical CT for the diagnosis of diaphrag-
matic injury in a patient population with a
relatively high proportion of penetrating inju-
ries, establish a threshold measurement of the
crus that could be used to distinguish between
an uninjured and an injured diaphragm, and de-
scribe an additional sign suggesting diaphrag-
matic injury: active arterial extravasation of
contrast material at the level of the diaphragm.
Materials and Methods
Study Population
The trauma registry at a large urban trauma cen-
ter was searched for cases of surgically or autopsy-
proven diaphragmatic injury from March 1998 to
March 2001. The search yielded 120 cases. Among
these 120 patients, 23 patients underwent helical CT
of the abdomen with or without dedicated chest CT
before surgery or before death. An additional two
patients with diaphragmatic injury who underwent
preoperative CT were identified at a collaborating
institution, yielding a total study population of 25
patients with proven diaphragmatic injury.
The average age of the study population was 36.2
years (range, 17–82 years). Seventeen patients sus-
tained right-sided injuries, and eight patients had
left-sided injuries; there were no bilateral injuries.
The mechanism of diaphragmatic injury was pene-
trating trauma (i.e., gunshot wound or stab wound)
in 14 patients and blunt trauma (i.e., injury in a mo-
tor vehicle crash, struck by a car, or fall from a
height) in 11 patients.
An additional 22 patients who sustained either
blunt (n = 14) or penetrating (n = 8) trauma, had sur-
gically confirmed intact diaphragms, and had under-
gone preoperative CT were randomly chosen and
included as the control group, bringing the total
study population to 47 patients.
CT Scans
All patients underwent helical CT (CTi; General
Electric Medical Systems, Milwaukee, WI) of the ab-
domen and pelvis after receiving water-soluble oral
contrast material (Gastrografin [meglumine diatri-
zoate]; Bracco Diagnostics, Princeton, NJ) and IV
contrast material (Omnipaque 300 [iohexol]; Ny-
comed, Princeton, NJ). Scanning with a 7-mm colli-
mation and a 7-mm reconstruction interval was
performed 70 sec after IV injection of contrast mate-
rial at a rate of 2 mL/sec. Imaging began at the lung
bases, above the diaphragm, in all patients.
452
Larici et al.
Nine patients also underwent chest CT, which
was performed before abdominal CT. CT scans of
the chest were ordered at the discretion of the
trauma service. Chest CT using a 5-mm collimation
with a 3-mm reconstruction increment was per-
formed 30 sec after IV injection of contrast mate-
rial; the contrast material was administered at a rate
of 3 mL/sec. Only one injection of contrast material
was performed for patients who underwent both
thoracic and abdominal CT examinations. One pa-
tient had dedicated 1-mm helical imaging through
the lower chest and upper abdomen in addition to
dedicated thoracic and abdominal CT. Chest CT ex-
tended from the cervicothoracic junction to the up-
per abdomen. Because the entire length of the
diaphragm was not always completely covered by
chest CT alone, axial and reformatted data from
both chest and abdominal studies were presented
for retrospective review. Images were photographed
using soft-tissue window settings (level, 40 H;
width, 440 H).
Digital CT data from 20 patients were available,
thus allowing sagittal and coronal reconstructions to
be performed. Reconstructed images were calculated
using 0.7-mm spacing between images with com-
mercially available software (Advantage Windows
4.0; General Electric Medical Systems). Approxi-
mately 40 reconstructed images were generated for
each of the two planes.
CT Review
The CT studies from these patients were retro-
spectively reviewed by five thoracic radiologists
who were unaware of the surgical or autopsy find-
ings. The reviewers were aware only of the context
of the study. Reviewers specifically evaluated the
imaging studies for an abnormally elevated hemidia-
phragm, diaphragmatic discontinuity, visceral herni-
ation with or without the collar sign, an abnormally
thick-appearing crus, the “dependent viscera” sign,
and any other finding that could suggest diaphragm
injury. Sagittal and coronal reformations were re-
viewed contemporaneously with axial images. Re-
viewers described each diaphragm as injured,
uninjured, or unknown and the reconstructed images
for each case as either helpful or not helpful in cor-
roborating the diagnosis based on the axial images.
The reviewers understood that a diagnosis of “un-
known” would prompt surgical exploration to defin-
itively exclude diaphragmatic injury. The presence
or absence of associated injuries was also noted.
Diaphragm Measurements
For the entire study population, the thickness of
the diaphragmatic crura was measured in an at-
tempt to establish a value that would enable radiol-
ogists to distinguish between normal and injured
diaphragms. The diaphragm crura were measured
by drawing perpendicular lines bisecting one an-
other through the center of the spinal canal at the
level of L1 and then measuring the thickness of the
crus along a line drawn at a 45° angle to the inter-
section of these two lines.
Statistical Analysis
The sensitivity, specificity, positive and negative
predictive values, and accuracy for the helical CT di-
agnosis of traumatic diaphragmatic injury were cal-
culated for the entire population, for patients with
right- versus left-sided injuries, and for those with a
penetrating versus blunt mechanism of injury. The
sensitivity and specificity of the individual findings
of traumatic diaphragmatic injury were also calcu-
lated. The paired two-tailed t test was used to com-
pare the average thicknesses of the right and left
diaphragmatic crura of patients with an injured dia-
phragm and those of patients with an uninjured dia-
phragm for the entire study population as well as for
patients with trauma from a blunt mechanism of in-
jury and those with trauma from a penetrating mech-
anism of injury.
Odds ratios were calculated for the ability of CT
to detect diaphragmatic injuries caused by blunt and
penetrating mechanisms of injury. The odds ratio
represents the likelihood of CT detecting a diaphrag-
matic injury versus the likelihood of CT suggesting
a diaphragmatic injury that was subsequently proven
incorrect at surgery. These odds ratios were then
compared with the Breslow and Day’s test for ho-
mogeneity of odds ratios to determine whether CT
performed significantly differently for the diagnosis
of diaphragmatic injury in patients with a blunt ver-
sus penetrating mechanism of injury. Statistics were
performed using commercially available software
(SAS 8.2; Statistical Analysis System, Cary, NC).
Results
Associated abdominal injuries were
present in 14 (56%) of the 25 patients with
diaphragmatic injury.
The sensitivity, specificity, positive and neg-
ative predictive values, and accuracy for the
detection of diaphragmatic injuries on helical
CT were 84% (21/25), 77% (17/22), 81% (21/
26), 81% (17/21), and 81% (38/47), respec-
tively. The sensitivity and specificity of helical
CT for penetrating injury were 86% (12/14)
and 79% (11/14), respectively, and those for
blunt injury were 82% (9/11) and 75% (6/8).
The odds ratio for the CT detection of blunt di-
aphragmatic injury was 13.5 (CI, 1.47–123)
and that for penetrating injury was 22.0 (CI,
3.1–157). Such large odds ratios imply a
strong correlation between the CT diagnosis of
traumatic diaphragmatic injury and surgical
confirmation of diaphragmatic injury. The
ability of CT to detect diaphragmatic injury on
the basis of the mechanism—blunt or penetrat-
ing—did not differ significantly ( p = 0.747).
The sensitivity and specificity in detecting
diaphragmatic injuries in patients with a right-
versus left-sided injury and in those with a
penetrating versus blunt mechanism of injury
as well as the individual helical CT findings of
AJR:179, August 2002
 Helical CT of Diaphragmatic Injury

diaphragmatic injury are presented in Table 1. Sensitivity and Specificity of Helical CT for Diagnosis of Diaphragmatic
The most sensitive finding (52%) for the de- TABLE 1
Injury Based on Side of Injury, Mechanism of Injury, and Imaging Findings
tection of diaphragmatic injury was the re-
Sensitivity Specificity
cently described dependent viscera sign [2] Variable
(Fig. 1). Direct visualization of diaphragmatic % No. % No.
Downloaded from www.ajronline.org by Washington State Univ. on 10/03/15 from IP address 134.121.47.100. Copyright ARRS. For personal use only; all rights reserved

injury (Fig. 2) and the “thick crus” sign [18] Side of injury
(Fig. 3) were both 36% sensitive. The depic-
Right 79 13/17 82 18/22
tion of a missile or puncturing instrument tra-
Left 100 8/8 95 21/22
jectory was the most sensitive finding in
patients with a penetrating diaphragmatic in- Mechanism of injury
jury (Fig. 4). Active arterial extravasation of Blunt 82 9/11 75 6/8
contrast material near the diaphragm (Fig. 2) Penetrating 86 12/14 76 11/14
suggested diaphragmatic injury in two patients Helical CT finding
and was the only sign of diaphragmatic injury Collar sign 24 6/25 100 22/22
in one of these two patients. For the entire pop- Direct visualization of injury 36 9/25 95 21/22
ulation, the collar sign was the most specific Thick crus 36 9/25 77 17/22
(100%) manifestation of diaphragmatic injury Herniation without collar sign 8 2/25 95 21/22
(Fig. 5), although the trajectory of the injury in Trajectory 36 5/14 100 22/22
patients with penetrating injury was also 100%
Active extravasation of contrast material 8 2/25 100 22/22
specific (Fig. 4).
Dependent viscera sign 52 13/25 71 9/22
False-negative interpretations (4/25) oc-
curred most commonly in patients with other
injuries abutting the injured diaphragm, such
as a pleural effusion (Fig. 6) or perisplenic he- sifying the cases according to mechanism of ologists at major trauma centers will frequently
matoma (Fig. 7). injury, we also found that the mean thickness encounter patients with diaphragmatic injuries
Reviewers indicated that diagnostic confi- of the diaphragmatic crura was not signifi- [19]. Diaphragmatic injuries are commonly as-
dence of a true-positive finding was increased cantly greater in patients with an injured dia- sociated with other severe thoracoabdominal in-
by the sagittal and coronal reconstructions in phragm than in those with an uninjured juries [4, 8, 20], and the overall morbidity and
three of the 20 true-positive cases (Fig. 5). Dia- diaphragm (Table 2). mortality associated with traumatic diaphrag-
phragmatic injury was incorrectly suggested matic injuries may both exceed 30% [2, 21–24];
on reconstructed images in two cases (Fig. 8). this mortality rate reflects the frequency and se-
For the entire study population, the mean Discussion verity of other organ injuries associated with di-
thickness of the diaphragmatic crura (right and Diaphragmatic injuries occur in only 5–10% aphragmatic tears rather than the diaphragmatic
left) was not significantly greater in patients of patients with severe thoracoabdominal injury injury itself. The diagnosis of diaphragmatic in-
with an injured diaphragm than in those with [1, 3, 4], but the large number of trauma cases jury usually necessitates prompt surgical repair
an uninjured diaphragm (Table 2). When clas- each year in the United States implies that radi- [5, 6, 22]. If the diagnosis is missed or delayed,

Fig. 1.—30-year-old man involved in motor vehicle collision. Axial CT image Fig. 2.—17-year-old boy with right upper quadrant stab wound. Axial CT image reveals
shows bowel (arrows) is resting against posterior ribs in left lower hemithorax. discontinuity of right hemidiaphragm (curved arrow), which is diagnostic of diaphrag-
This finding represents “dependent viscera” sign. Rupture of left hemidiaphragm matic injury. Active arterial extravasation of contrast material (straight arrow) indi-
was surgically confirmed. cates that injury is in close proximity to diaphragm.

AJR:179, August 2002 453

 Larici et al.
Downloaded from www.ajronline.org by Washington State Univ. on 10/03/15 from IP address 134.121.47.100. Copyright ARRS. For personal use only; all rights reserved

Fig. 3.—60-year-old man involved in motor vehicle collision. Axial CT image shows
subjectively thickened right diaphragmatic crus (arrows), suggesting injury. Dia-
phragmatic rupture was confirmed at surgery.
Fig. 4.—17-year-old girl with right upper quadrant gunshot injury. Axial CT image re-
veals subcutaneous emphysema in superficial tissues of right flank (straight arrow).
Linear, irregular low-attenuation area in liver (curved arrow) is consistent with lac-
eration from missile. Extrapolating missile trajectory indicates that projectile must
have traversed diaphragm. Diaphragmatic injury was confirmed at surgery. Streak
artifact emanates from bullet fragment in posterior soft tissues of thorax.

the pressure difference between the thorax and
the abdomen tends to favor enlargement, rather
than healing, of the injury, which results in pro-
gressive herniation of the abdominal contents
into the thorax [3, 6].
CT has the major advantages of being
widely available, rapid, noninvasive, and accu-
rate for the assessment of the numerous injuries
that are often associated with diaphragmatic in-
juries. Helical CT offers major advantages over
conventional CT, particularly the ability to per-
form rapid volumetric acquisitions through the
area of interest using a narrow collimation.
Such techniques also allow high-quality sagit-
tal and coronal reformations to be performed.
These reformations have been shown to be
beneficial in experimental models and in prac-
tice [3, 12–14].
Several studies have evaluated the utility of
conventional and helical CT for the evaluation
of diaphragmatic injury, with sensitivities and
specificities ranging from 0% to 61% [2, 3, 9–
13, 25, 26] and from 76% to 99% [2, 3, 10, 12,
13, 26], respectively. Several CT findings sug-
gestive of diaphragmatic injury have been de-
scribed, including apparent elevation of a
hemidiaphragm, herniation of the abdominal
contents into the thorax (with or without waist-
like constriction of the herniated viscus, the so-
called collar sign) [3], direct visualization of

A B

Fig. 5.—33-year-old man involved in motor vehicle collision.

A, Axial CT image shows waistlike constriction of stomach (long arrows), suggesting herniation of stomach through injured diaphragm (short arrow).
B, Coronal reformatted CT image clearly shows herniation of stomach (long arrows) through diaphragmatic defect into left hemithorax; this finding represents collar sign.
Diaphragm (short arrow) can be seen lateral to stomach.

454 AJR:179, August 2002

 Helical CT of Diaphragmatic Injury
Downloaded from www.ajronline.org by Washington State Univ. on 10/03/15 from IP address 134.121.47.100. Copyright ARRS. For personal use only; all rights reserved

Fig. 6.—24-year-old man with right upper quadrant stab wound. Axial CT image re-
veals moderate-sized right pleural effusion (asterisk) but does not reveal direct ev-
idence of diaphragmatic injury. Large effusion may obscure direct visualization of
diaphragmatic injury that was proven at surgery.
Fig. 7.—65-year-old man involved in motor vehicle collision. Axial CT image ob-
tained with 1-mm collimation reveals evidence of splenic parenchymal injury
(curved arrow). Curvilinear high-attenuation structure (straight arrows) posterior
to spleen was thought to represent injured diaphragm, but no evidence of diaphrag-
matic injury was found at surgery.

diaphragmatic discontinuity, an abnormally
thickened diaphragmatic crus [18], evidence of
penetrating injury with a trajectory projected
to involve the diaphragm [15, 16, 27], and the
recently described dependent viscera sign [2].
The latter manifests as the cranial aspects of
the abdominal viscera (the liver on the right
and stomach or bowel on the left) resting
against posterior ribs because of the loss of di-
aphragmatic support after injury [2].
One of the purposes of our study was to ex-
amine the utility of helical CT for the detection
of diaphragmatic injury in a patient population
with a predominance of penetrating injuries
(gunshot or stab wounds) as opposed to patients
with blunt trauma injuries. Most CT studies of
diaphragmatic injuries to date have focused on
patients who sustained blunt trauma [3, 12], but
these results may not be generalizable to patients
with penetrating injuries. Mechanisms of dia-
phragmatic rupture from blunt trauma include
rapid, large increases in intraabdominal pressure
or shear stress [3, 6], whereas the injury mecha-
nism from a penetrating injury is related to the
missile or stabbing weapon itself. These differ-
ences in mechanism raise the possibility that the
average extent of injury to the diaphragm may
differ in patients who sustain blunt versus pene-
trating trauma [16, 28]. Many diaphragmatic in-
juries related to blunt trauma are large, often

A B

Fig. 8.—30-year-old man involved in motor vehicle collision.

A, Sagittal reformation reveals waistlike constriction (arrows) of dome of liver, raising possibility of right hemidiaphragm injury. No evidence of injury was found at surgery.
This appearance may occur as artifact of reconstruction or of patient breathing during acquisition of imaging volume.
B, Coronal reformation again shows constriction (arrows) or collar sign.

AJR:179, August 2002 455


Mean Thickness of
Diaphragmatic Crura:
TABLE 2
Injured Diaphragm Versus
Uninjured Diaphragm
Downloaded from www.ajronline.org by Washington State Univ. on 10/03/15 from IP address 134.121.47.100. Copyright ARRS. For personal use only; all rights reserved
Mean Thickness of
Group Crus (mm) p detected on helical CT [3, 12], perhaps in part
Injured Uninjured
Entire study
Right 7.53 5.83 0.326
Left 6.27 4.33 0.278
Blunt trauma
Right 10 6 0.314
Left 5.75 4.25 0.182
Penetrating trauma
Right 6.62 6 0.588
Left 5.5 3.5 0.201
more than 10 cm long [2, 7]. Gunshot wounds
can cause large blast injuries to the diaphragm,
but stab wounds may result in smaller injuries
[3, 16]. Therefore, helical CT might not detect
penetrating diaphragmatic injuries as readily as
blunt injuries. However, our study found that the
overall sensitivity and specificity of helical CT
for the detection of diaphragmatic injury were
84% and 77%, respectively. These results com-
pare favorably with the results of other CT inves-
tigations that focused primarily on patients who
sustained blunt trauma [3, 12]. Our favorable re-
sults, despite the fact that patients with penetrat-
ing injuries constituted a large portion of the
study population, are at least in part due to the
fact that injury trajectories seen on CT are useful
markers of injuries caused by penetrating trauma
[15, 16]. The trajectory of a penetrating injury
may be visible on CT scans as localized soft-tis-
sue swelling, subcutaneous emphysema, or focal
extravasation of IV contrast material (Fig. 4).
When injury to a solid or hollow organ is also
evident along this same path, diaphragmatic in-
jury is likely [15, 16, 27].
Although CT showed a moderately good
overall sensitivity of 84% for the detection of
diaphragmatic injury, the fact that sensitivity
was not higher is not unexpected. Studies of the
natural aging of the diaphragm [29] and MR
imaging of the diaphragm [30] have revealed
that the entire diaphragm often cannot be com-
pletely visualized with cross-sectional imaging.
Couple this concept with the fact that penetrat-
ing injuries are often smaller (and therefore
harder to detect) than blunt injuries and with the
limited specificity of several CT findings of dia-
phragmatic injury, and it becomes clear why the
CT diagnosis of traumatic diaphragmatic injury
is often a difficult one.
456
Larici et al.
Similar to Killeen et al. [12], we found that
helical CT performed better for left-sided inju-
ries (100% sensitivity) than for right-sided inju-
ries (76% sensitivity). Other investigators have
found that left-sided injuries are more readily
because of the relatively greater tissue contrast
between fat in the left upper quadrant and the
adjacent left hemidiaphragm than the poor tis-
sue contrast between the right hemidiaphragm
and the adjacent liver. These differences in tis-
sue contrast may enhance direct visualization of
diaphragmatic injuries.
The most sensitive sign of diaphragmatic in-
jury in our study was the newly described depen-
dent viscera sign. This sign showed 100%
sensitivity for left- and 83% sensitivity for right-
sided diaphragmatic injuries in a recent investi-
gation of patients who sustained blunt trauma
[2]. The sensitivity of this sign in our investiga-
tion was only 52%. This lower sensitivity is per-
haps related to the predominance of right-sided
injuries, which may be more difficult to diag-
nose on imaging [12], and to the relatively high
proportion of injuries caused by penetrating in-
jury mechanisms in our study population. The
dependent viscera sign might require a large tear
to become evident, and large tears may be more
prevalent with blunt mechanisms of injury. Also,
on the left side, this sign may be observed when
the tear occurs in the classic location for blunt
trauma: the posterolateral aspect of the dia-
phragm, between the spleen and abdominal
aorta [2]. This site is structurally weaker than the
remaining areas of the diaphragm [2, 3]. Given
that penetrating injuries do not occur preferen-
tially at this site, this sign may be encountered
less frequently in patients with penetrating dia-
phragmatic injury. Because this sign relies on the
depiction of the cranial aspect of an abdominal
viscus contacting posterior ribs, we found that
the most common cause of the absence of this
CT sign in patients with proven diaphragmatic
injury was a large pleural effusion. Other investi-
gators have also noted that pleural effusions may
obscure diaphragmatic injury on CT [3, 10].
Although the dependent viscera sign was
the most sensitive indicator of diaphragmatic
injury in our series, the sensitivity of this sign
was only 52%. Given that the overall sensitiv-
ity of helical CT for the detection of dia-
phragmatic injury was 84%, it is apparent
that the various CT findings of diaphragm in-
jury are often seen in isolation. This fact em-
phasizes the idea that those who interpret CT
scans of trauma patients must be familiar
with all the CT findings of diaphragmatic in-
jury and must be prepared to suggest the
presence of an injury on the basis of a single
CT finding.
We found sagittal and coronal reformations
added little to the reviewers’ ability to diagnose
diaphragmatic injury. In contrast, Killeen et al.
[12] found that reformatted images aided in the
diagnosis of diaphragm injury, and experimen-
tal investigations reported that sagittal reforma-
tions were 92% sensitive for the diagnosis of
diaphragm injury and were diagnostically supe-
rior to both coronal and axial reformations [13,
14]. In our series, reformatted images increased
diagnostic confidence for only five (20%) of the
25 true-positive cases; the sagittal images were
considered the most useful plane in three of
these cases. In none of the cases did reformatted
images detect an injury unseen on axial images,
and in two cases the reformatted images were
considered misleading.
A recently described sign of diaphragmatic
injury, the thick crus sign, represents the subjec-
tive impression that the crus of the diaphragm is
abnormally thickened [18]. We sought to estab-
lish a threshold measurement that could distin-
guish between normal and injured diaphragms
to avoid the subjective nature of this sign. How-
ever, the mean thickness of the diaphragmatic
crura was not significantly greater for patients
with an injured diaphragm than for those with
an uninjured diaphragm for the study popula-
tion as a whole or when segregating the popula-
tion on the basis of blunt versus penetrating
injury mechanisms. It is possible that injured di-
aphragms appear significantly thicker than un-
injured diaphragms on CT, but our study
population was too small to detect this differ-
ence. Additionally, because the thickness of the
normal crus varies with patient age, the range of
measurements for the normal diaphragmatic
crus would require analysis of a large number of
patients representing a wide age range.
The thick crus sign may be seen more fre-
quently in patients with a blunt injury, in
whom the site of injury is often more consis-
tent. Penetrating mechanisms cause injury on
the basis of the type and trajectory of the pro-
jectile; therefore, a single measurement at one
portion of the diaphragm will probably be of
little use. The subjective observation of dia-
phragm thickening as a manifestation of injury
may be more useful if extended to all portions
of the diaphragm, not just the crus.
The two major limitations of our work are the
retrospective nature of the study and the rela-
tively small sample size. The latter is a problem
common to most studies about the utility of CT
for diagnosis of diaphragmatic injury and is asso-
ciated with the relative rarity of preoperative CT
AJR:179, August 2002

for an uncommon injury. For example, almost
80% of the patients with proven diaphragmatic
injury over the study period at our institution did
not undergo CT before surgery; such a situation
introduces selection bias in the review.
Downloaded from www.ajronline.org by Washington State Univ. on 10/03/15 from IP address 134.121.47.100. Copyright ARRS. For personal use only; all rights reserved
The sizes of the diaphragmatic injuries were
not available from the surgical notes in most of
the patients. Without knowing the sizes of the
injuries, we cannot directly compare the perfor-
mance of helical CT for the detection of blunt
versus penetrating diaphragmatic injuries. Al-
though our results compare favorably with
those from studies performed primarily on pa-
tients who sustained blunt trauma, without the
average size of the penetrating diaphragmatic
injuries in our population, the role of helical CT
in the diagnosis of penetrating diaphragmatic
injury cannot be firmly established.
Our results might have been improved if a
specific CT protocol had been implemented
for imaging the diaphragm. Other investiga-
tors advocate CT evaluation of the region of
the diaphragm with narrow collimation and
reconstruction increments after routine CT
evaluation for trauma to the abdomen [3].
These protocols increase spatial resolution and
the quality of the reformatted images. With the
increasing use of multidetector CT, rapid acqui-
sition of narrowly collimated images may be
possible without increased scanning time or in-
creased heat-loading stress on the CT tube. Fu-
ture investigations using this equipment may
show improved results.
Our attempt to establish a crus measurement
that would enable radiologists to distinguish be-
tween normal and injured diaphragms was lim-
ited by several factors. Our sample size was
small; thus, our power to detect a difference be-
tween normal and injured diaphragms was lim-
ited. Because of the small size of our sample,
we were not able to stratify measurements ac-
cording to patient sex and age. Both of these
factors are known to influence the thickness of
the normal diaphragm. Finally, we performed
our measurements at the level of the L1 verte-
bral body so that these measurements could be
standardized across our population. However,
the diaphragm has multiple muscular origins, so
any measurement that attempts to differentiate
normal from injured diaphragms will be useful
only if the injury occurs at a site where the mea-
surements were studied.
In conclusion, helical CT has good accuracy
for the detection of diaphragmatic injury, but
AJR:179, August 2002
Helical CT of Diaphragmatic Injury
sagittal and coronal reconstructions are of lim-
ited use in confirming or refuting this diagnosis.
The most sensitive sign for the detection of dia-
phragmatic injury was the recently described
dependent viscera sign, although observation of
the trajectory of missile or stab injuries in pa-
tients with penetrating trauma was also useful.
The collar sign and the trajectory of the missile
or stab injuries in patients with penetrating in-
jury mechanisms were 100% specific for the
detection of diaphragmatic injury. Because the
sensitivity of any individual sign of diaphrag-
matic injury is not particularly high, familiarity
with all CT findings of diaphragmatic injury is
required, and the diagnosis must occasionally
be made on the basis of a single finding.
References
1. Ward RE, Flynn TC, Clark WP. Diaphragmatic dis-
ruption secondary to blunt abdominal trauma. J
Trauma 1981;21:35–38
2. Bergin D, Ennis R, Keogh C, Fenlon HM, Murray
JG. The “dependent viscera” sign in CT diagnosis of
blunt traumatic diaphragmatic rupture. AJR 2001;
177:1137–1140
3. Shanmuganathan K, Killeen K, Mirvis SE, White
CS. Imaging of diaphragmatic injuries. J Thorac Im-
aging 2000;15:104–111
4. Meyers BF, McCabe CJ. Traumatic diaphragmatic
hernia: occult marker of serious injury. Ann Surg
1993;218:783–790
5. Morgan AS, Flancbaum L, Esposito T, Cox EF.
Blunt injury to the diaphragm: an analysis of 44 pa-
tients. J Trauma 1986;26:565–568
6. Kearney PA, Rouhana SW, Burney RE. Blunt rup-
ture of the diaphragm: mechanism, diagnosis, and
treatment. Ann Emerg Med 1989;18:1326–1330
7. Estrera AS, Landay MJ, McClelland RN. Blunt trau-
matic rupture of the right hemidiaphragm: experience
in 12 patients. Ann Thorac Surg 1985;39:525–530
8. Ilgenfritz FM, Stewart DE. Blunt trauma of the dia-
phragm: a 15-county, private hospital experience. Am
Surg 1992;58:334–338; discussion 338–339
9. Chen JC, Wilson, SE. Diaphragmatic injuries: recog-
nition and management in sixty-two patients. Am
Surg 1991;57:810–815
10. Murray JG, Caoili E, Gruden JF, Evans SJ, Hal-
vorsen RA Jr, Mackersie RC. Acute rupture of
the diaphragm due to blunt trauma: diagnostic
sensitivity and specificity of CT. AJR 1996;166:
1035–1039
11. Voeller GR, Reisser JR, Fabian TC, Kudsk K, Man-
giante EC. Blunt diaphragm injuries: a five-year ex-
perience. Am Surg 1990;56:28–31
12. Killeen KL, Mirvis SE, Shanmuganathan K. Helical
CT of diaphragmatic rupture caused by blunt trauma.
AJR 1999;173:1611–1616
13. Israel RS, McDaniel PA, Primack SL, Salmon CJ,
Fountain RL, Koslin DB. Diagnosis of diaphrag-
matic trauma with helical CT in a swine model. AJR
1996;167:637–641
14. Israel RS, Mayberry JC, Primack SL. Diaphragmatic
rupture: use of helical CT scanning with multiplanar
reformations. AJR 1996;167:1201–1203
15. Grossman MD, May AK, Schwab CW, et al. Deter-
mining anatomic injury with computed tomography
in selected torso gunshot wounds. J Trauma 1998;
45:446–456
16. Shanmuganathan K, Mirvis SE, Chiu WC, Killeen
KL, Scalea TM. Triple-contrast helical CT in pene-
trating torso trauma: a prospective study to determine
peritoneal violation and the need for laparotomy.
AJR 2001;177:1247–1256
17. Chiu WC, Shanmuganathan K, Mirvis SE, Scalea TM.
Determining the need for laparotomy in penetrating
torso trauma: a prospective study using triple-contrast
enhanced abdominopelvic computed tomography. J
Trauma 2001;51:860–869
18. Leung JC, Nance ML, Schwab CW, Miller WT Jr.
Thickening of the diaphragm: a new computed to-
mography sign of diaphragm injury. J Thorac Imag-
ing 1999;14:126–129
19. Gelman R, Mirvis SE, Gens D. Diaphragmatic rup-
ture due to blunt trauma: sensitivity of plain chest ra-
diographs. AJR 1991;156:51–57
20. Shah R, Sabanathan S, Mearns AJ, Choudhury AK.
Traumatic rupture of diaphragm. Ann Thorac Surg
1995;60:1444–1449
21. Madden MR, Paull DE, Finkelstein JL, et al. Occult
diaphragmatic injury from stab wounds to the lower
chest and abdomen. J Trauma 1989;29:292–298
22. Reber PU, Schmied B, Seiler CA, et al. Missed dia-
phragmatic injuries and their long-term sequelae. J
Trauma 1998;44:183–188
23. Rodriguez-Morales G, Rodriguez A, Shatney CH.
Acute rupture of the diaphragm in blunt trauma:
analysis of 60 patients. J Trauma 1986;26:438–444
24. Wiencek RG Jr, Wilson RF, Steiger Z. Acute injuries
of the diaphragm: an analysis of 165 cases. J Thorac
Cardiovasc Surg 1986;92:989–993
25. Shapiro MJ, Heiberg E, Durham RM, Luchtefeld
W, Mazuski JE. The unreliability of CT scans and
initial chest radiographs in evaluating blunt
trauma induced diaphragmatic rupture. Clin Ra-
diol 1996;51:27–30
26. Worthy SA, Kang EY, Hartman TE, et al. Diaphrag-
matic rupture: CT findings in 11 patients. Radiology
1995;194:885–888
27. Shackleton KL, Stewart ET, Taylor AJ. Traumatic di-
aphragmatic injuries: spectrum of radiographic find-
ings. RadioGraphics 1998;18:49–59
28. Mueller CS, Pendarvis RW. Traumatic injury of the
diaphragm: report of seven cases and extensive liter-
ature review. Emerg Radiol 1994;1:118–132
29. Caskey CI, Zerhouni EA, Fishman EK, Rahmouni
AD. Aging of the diaphragm: a CT study. Radiology
1989;171:385–389
30. Gierada DS, Curtin JJ, Erickson SJ, et al. Fast gradi-
ent echo magnetic resonance imaging of the normal
diaphragm. J Thorac Imaging 1997;12:70–74
457