|
 Abstract
To evaluate the efficacy of emergency CT in patients
with blunt pancreatic trauma on their
admission to the hospital and to
describe the CT findings of pancreatic
parenchymal and main pancreatic duct
injuries, we performed a retrospective
study. During a 4-year period
(2001-2005), 15 patients (twelve males
and three females; age range 4-38 years)
were investigated by spiral CT.
Ultrasound at admission was obtained in
7 cases and endoscopic retrograde
cholangiopancreatography (ERCP) was also
obtained in 7 cases. Serum amylase was
reviewed at admission in all cases. The
imaging findings obtained by the spiral
CT were correlated with ultrasound, ERCP
and operative findings (when available).
At admission, the diagnosis of
pancreatic injury was missed by CT in
four patients (26.7%). Thus, the
sensitivity of CT was 73.3%. ERCP showed
rupture of the main pancreatic duct in
all cases in which it was performed. One
of them was negative on the CT obtained
at admission. Serum amylase at admission
was elevated in 53.3% of cases. No
correlation between the CT features and
type of outcome of surgical management
could be noted in this study. On
retrospective review of the CT
examinations obtained at admission, we
found that two of the four false
negative results could have been
avoided. Therefore, proper CT technique
and accurate film reading are mandatory
for diagnosis of pancreatic injuries. If
the CT at admission with proper
technique and proper film reading do not
reveal evidence of pancreatic injury
with the presence of persistent
unexplained abdominal pain or elevated
serum amylase, we advise to repeat the
CT after 24-48 hours.
Introduction
Injuries to the pancreas are uncommon and account for
1-4% of severe abdominal injuries(1).
Most pancreatic injuries occur in young
men. Blunt trauma to the pancreas and
duodenum usually results from road
traffic accidents, when an unrestrained
driver is thrown on to the steering
wheel. Handlebars may cause similar
injuries to motorcyclist or children on
bicycle (1, 2).
Although pancreatic trauma is uncommon, it causes
significant morbidity and mortality
because of the delay in recognition and
consequent treatment (5).
Transection of the pancreas is more common to the left
of mesenteric vessels and occurs when
there is compression over the spine.
Injuries to the right and the midline
may cause more serious crushing injuries
of the pancreatic head and duodenum
(17).
Isolated pancreatic injuries are uncommon due to its
retroperitoneal location and the
proximity of multiple organs and major
vascular structures (12).
As a result of associated injuries, pancreatic trauma is
often overlooked at presentation.
Radiological findings may be subtle in
the acute phase, compared with the
lesions in the liver and spleen.
Associated injuries may be assumed to be
responsible for free intraperitoneal
fluid or inflammatory changes in the
upper abdomen (17).
The purpose of this study is to describe the spiral CT
findings of blunt pancreatic injuries
and to evaluate retrospectively the
sensitivity of this technique in
assessment of pancreatic contusions and
main duct injuries in correlation to ERP
(endoscopic retrograde pancreatography)
and operative findings when available.
Materials and Methods
During a 4-year period (2001-2005), 15 patients (twelve
males and three females; age range 4-38
year, mean age 19.9 years) were
investigated with CT after having
sustained blunt abdominal trauma (Table
1). The medical charts, laboratory data,
CT scans and operative details were
analyzed in this retrospective study.
Eight patients were investigated by CT
within 6 hours after injury. A CT
grading scheme of Yon-Cheong et al (24)
was applied to predict the presence or
absence of duct disruption (Table 2).
The modified classification of Lucas
(14) was also used, which also considers
the combined duodenal and main
pancreatic duct lesions for scaling of
the CT findings (Table 3).
Patients were examined on a fourth generation CT
scanner with spiral capability (Somaton
Plus 4, Siemens, Germany). A spiral
volume was acquired using 10 mm
collimation from level of diaphragm
until the level of symphysis pubis after
oral administration of Gastrografin 3%
(Schering, Germany) and after I.V.
injection of 120-170 ml of non-ionic
contrast medium (iohexol 300 mg
iodine/ml, Schering, Germany), at an
injection rate of 2 or 3 ml/s.
ERP was obtained preoperatively in seven patients. Ultrasound
was done prior to CT in 7 cases. Ten
patients underwent surgical treatment.
Two cases had
|
Table 2 : CT grading for
blunt pancreatic injuries |
|
Table 3 : pancreatic injury,
modified calassification of
lucas |
CT-guided drainage. No surgical
intervention was performed for 3 cases.
Results
Pancreatic injury occurred in the tail
in 8 cases (53.3%), neck in 5 cases
(33.3%), and body in two patients
(13.3%). Eight patients (53.3%) had an
isolated pancreatic injury and seven
(46.7%) had multiple visceral injuries.
An extraluminal contrast material was
observed in one patient (6.7%) in which
duodenal perforation was confirmed at
surgery. Hepatic contusion was found in
4 cases (26.7%), left renal contusion
was found in two patients (13.3%). The
same number was also observed on the
right kidney. Left suprarenal hematoma
was also found in 1 case (6.7%). Brain
contusions, spinal fractures and
skeletal fracture; each of them was
found in 1 case. CT showed evidence of
pancreatic injury at admission on 11
cases 73.3% (Fig.1-3).
A
|
Fig. 1. A Transcaudal
fracture of the pancreatic
neck with blurred
surrounding fat planes yet
with no obvious
peripancreatic collection
B ERP shows leakage e of
contrast via the ruptured
main pancreatic duct
|
A
|
Fig. 2. Full thickness
fracture of the body of
pancreas with peripancreatic
collection . B ERP
demonstarted contrast
leakage the main pancreatic
duct
|
A
|
Fig. 3.A Ultrasound on
admission revealed
hypoechoic contusion or
hematoma on the pancreatic
nec.
B spiral CT doneimmediately
after the ultrasound showed
transection of the pacreatic
neck with pancreatic and
peripancreatic collection .
C Repeated CT after 3 days
showed formation of
posttraumatic pseudocyst
|
The diagnosis of pancreatic contusion was missed
on the initial CT on 4 cases (26.7%).
Two of them showed no specific findings
on the spiral CT done at admission. One
case showed evidence of right renal
injury. On the repeated CT done on the
next day, a fluid collection related to
the pancreatic tail appeared (Fig.4).
A
B
|
Fig.4. A CT at admission
revealed no definite
evidence of pancreatic tail
injury. The attention was
paid to the right renal
injury. B Repeat CT revealed
appearance of fluid
collection related to the
tail of pancreas and extends
to the pararenal space.
Pancreatic tail injury was
confirmed at laparotomy.
|
Laparotomy done on the second day of
admission revealed evidence of
pancreatic tail rupture.
The second one showed evidence of splenic,
hepatic, both renals, left suprarenal
injuries with retroperitoneal hematoma.
Since the serum amylase was high in this
patient, CT was repeated on the next
day, which revealed pancreatic tail
laceration that was confirmed at
laparotomy.
The third case showed suspected heterogenous hypodense
texture on the CT at admission but no
definite injury was established.
Laparotomy was also negative at the day
of admission. A second CT reading found
a trace of fluid related to the
pancreatic tail extended to the
pararenal space. Repeated multiple CT
were performed showing increased amount
of this fluid with development of
pseudopancreatic cyst then abscess.
The initial CT on the fourth case revealed pancreatic
body swelling (edema) with no definite
injury. ERP on the same day of admission
revealed middle third main pancreatic
duct rupture. Repeat CT after one day
showed evidence of pancreatic body tear
with pancreatic and peripancreatic
collections.
& In our study, an elevated serum amylase level was
found at admission in 8 cases (53.3%).
Ultrasound was performed at admission in 7 cases
(46.7%). It was negative in four
patients. Three of them showed evidence
of suspicion of pancreatic injury on the
CT done at admission shortly after the
ultrasound. ERP was done on 7 cases in
whom main pancreatic duct rupture with
leakage was found in 6 patients and one
patient was normal..
As a treatment, sump drainage alone was done in 5
cases, CT-guided drainage was performed
in two patients, caudal pancreatectomy
alone was performed in 2 cases. The
combined caudal pancreatectomy and sump
drainage was done in two patients.
Cystogastrostomy was done for the case
of pseudopancreatic cyst. No specific
treatment was given for the remaining 3
cases. Two cases died from brain injury.
Two patients had pseudopancreatic cyst, one of them
transformed into abscess. Pancreatic
fistula occurred also in two, and one
patient experienced delayed pancreatitis,
while the remaining eight patients
survived without sequelae.
Discussion
Pancreatic injuries are relatively uncommon,
accounting for 3 to 12% of abdominal
trauma. These injuries appear after a
sudden force compresses the pancreas
against the lumbar spine. In the adult,
they usually are seen after motor
vehicle accidents whereas in children
after bicycle accidents (21).
Most traffic accidents occur in unrestrained
drivers. Blunt pancreatic trauma in a
passenger or a restrained driver is rare
(3). In our study, 46.7% of patients
were involved on motor vehicle
accidents, while 33.3% of patients were
bicycle riders. In our locality, riding
accident was the causative injury in 20%
of cases.
Pancreatic injury is more common in young adults,
possibly because less retroperitoneal
fat to act as a protective buffer (22).
Because children are smaller than
adults, traumatic forces in children
usually affect a larger region of the
body than in adults. Thus multisystem
injuries are more common among cases of
pediatric trauma (19).
Pancreatic injuries occurring to the right of the mesenteric
vessels are defined as proximal, while
those occurring to the left of the
mesenteric vessels were defined as
distal (10).
The pancreatic fracture is demonstrated in abdominal CT
by a clear separation or fracture line
across the long axis of the pancreas.
The fracture line is generally present
across the neck (23). All portions of
the pancreas except the pancreatic head
were involved in our study.
Most of the studies described that the associated injuries
are more common than the isolated
pancreatic injuries. Figures of 50-98%
are widely reported (2,15,17). However,
isolated pancreatic injuries were more
common in our study accounting 53.3% of
cases. 46.7% of our patients had
associated injuries including liver,
spleen, kidneys, duodenum and
suprarenals. In blunt abdominal trauma,
the morbidity and mortality is mainly
due to associated injuries (12) and the
isolated injuries usually indicate good
prognosis (8).
The typical clinical triad of pancreatic trauma
includes upper abdominal pain.
Leukocytosis and hyperamylasemia are
uncommon early on and are nonspecific
findings that can result from bowel
injury (22,23). This triad may be
delayed for 24 hours and even for
several days following injury (2, 21).
The serum amylase level was normal at admission in
46.7% of cases and thus is an unreliable
early predictor of pancreatic trauma.
Injuries of the pancreas can remain unrecognized during
laparotomy in 8% of the cases with
potentially disastrous consequences (2).
In our study, laparotomy was negative in
1 case (25%) of the preoperatively
missed diagnosis. The indications of
laparotomy in patients with injury to
the pancreas have been described as
being dependent on whether the main
pancreatic duct has been injured (20).
In our study, ERP was obtained in 7 cases (46.7%)
and confirmed rupture of the main
pancreatic duct, which further confirmed
at surgery; however, the pancreatography
whether preoperative or intraoperative
to diagnosis ductal damage remains
controversial (7,11). ERP is even
thought to be inappropriate in an acute
posttraumatic case because of its
invasive nature (11).
Ultrasound is suitable for diagnosing focal or
diffuse pancreatitis or pseudocyst, but
ultrasound generally does not depict
pancreatic fracture (22). In our study,
ultrasound was performed at admission in
7 cases. It was only positive in 3 cases
(42.9%).
CT is the most effective diagnostic
modality to diagnose pancreatic fracture
(2). In our study, the diagnosis was
missed in 26.7% of cases at admission.
This is partially due to an observer
error. In 2 cases, the attention was
focused on the other abdominal visceral
injury or potential active bleeding. A
10-mm slice thickness section and
absence of oral contrast (despite the
clinical stability of the patients) as
well as the lack of getting better
parenchymal enhancement with arterial
phase imaging were also considerable
factors.
Associated splenic injury, which focused the
attention, particularly when associated
with massive hemoperitoneum should not
be misleading. The possibility of a
pancreatic contusion should always be
considered when the mechanism of injury
was a direct frontal upper abdominal
impact (2).
The pancreatic fracture is clearly demonstrated with a
perpendicular orientation on the long
axis of pancreas. Other helpful signs of
suspicion are pancreatic or
peripancreatic hematomas, periduodenal
hematoma, retroperitoneal fluid, edema
of the peripancreatic fat or around the
superior mesenteric vessels or
thickening of the anterior perirenal
fascia (6).
A high index of suspicion is required for CT diagnosis
of pancreatic injury. The spectrum of CT
findings also include focal or diffuse
pancreatic enlargement, irregularity of
the pancreatic contour, fluid in the
transverse mesocolon or lesser sac and
fluid between the pancreas and splenic
vein (22).
In most cases of pancreatic injuries, the attention is
focused on the injury of the main
pancreatic duct (20). The presence of
retroperitoneal fluid collection
suggests pancreatic duct rupture, which
may require emergent ERP (22). A patient
with posttraumatic pseudocyst is
considered to have ductal leak until
proven otherwise (17).
CT is not adequate in demonstrating pancreatic duct
rupture, but ERP is credited with 100%
sensitivity (2, 24). However, ERP could
be undertaken only in stable patients.
It also requires experienced
endoscopists who generally are not
rapidly available in emergency settings
(24). This technique is currently most
useful in preoperative delineation of
the ductal anatomy in patients with
missed injuries (4) .
As a treatment, sump drainage alone was applied in 5
(33.4%) of our patients. Three of them
had grade II injury according to Lucas
classification. Caudal pancreatectomy
alone or with guided drainage was
applied on 13.3% of our patients. All of
them were grade II. CT-guided drainage
was performed in two cases. In our
study, the duration of hospital stay was
neither influenced by the type of injury
or type of management.
Regarding complications, most of the literature
reported that the formation of
pancreatic fistula is the most common
complication of pancreatic injury with
an incidence ranging from 7 to 20%
(16,18). In our study, pancreatic
fistula occurred in 13.3% of cases. The
literature reported a variable mortality
rate from 3-40%, which increases when
associated injuries are present (11,18).
In our study, we had two deaths (13.3%)
from brain injury, 5 patients (33.3%)
had delayed complications while the
majority of the patients (53.4%)
survived without sequelae.
Summary
Pancreatic trauma is an uncommon and
frequently overlooked sequel of major
blunt abdominal trauma. CT is the most
effective modality for diagnosis of
acute abdominal trauma; however, the CT
diagnosis of pancreatic injury may be
missed at admission. Proper examination
technique including oral contrast in
stable patients, I.V. contrast
enhancement and spiral acquisition with
thin slice thickness sections are
mandatory. Accurate film reading should
pick up indirect signs of pancreatic
injury, particularly when associated
with contusions of the other organs. A
patient with pancreatic injury and
associated collections or posttraumatic
pseudocyst should be considered to have
a ductal leak until proven otherwise.
Also ductal disruption is likely to be
present if the pancreas shows
transection or deep laceration on CT. In
doubtful cases or patients with
sustained blunt abdominal trauma have
persistent unexplained abdominal pain or
elevated serum amylase level, CT should
be repeated after 24-48 hours.
References
Other
Topics:
Original Study # 1
- Autonomic Neuropathy and Silent Myocardial Ischemia in
Type II Diabetic
Patients
Original Study # 3
- Risk Factors for Duodenal Ulcer Perforation.
|
