Case reports

Surg Endosc (1999) 13: 523525

Springer-Verlag New York Inc. 1999

Duodenal perforations after laparoscopic cholecystectomy

E. Croce, M. Golia, R. Russo, M. Azzola, S. Olmi, G. De Murtas
General and Thoracic Surgery Department, Minimally Invasive Surgery Center, Fatebenefratelli ed Oftalmico Hospital, Corso di Porta Nuova 23,
20121 Milan, Italy
Received: 27 May 1998/Accepted: 14 September 1998

ing diagnostic aspects, as will be shown by our cases and

some others that have been reported in the literature.

Abstract. Duodenal perforations after laparoscopic cholecystectomies are rarely reported. The aim of this study is to
focus on this complication and to suggest ways to reduce its
occurrence and avoid diagnostic mistakes and therapeutical
delays that could be fatal. We reviewed four personal cases
and a number of others reported in the literature. Duodenal
perforations are caused by improper use of the irrigatoraspirator device when retracting the duodenum, or by electrosurgical and laser burns. A duodenal perforation should
be suspected in cases of bile leakage, peritonitis, intraabdominal or retroperitoneal collections, high serum or drainage amylase concentration, absence of bile leakage from the
biliary tree, and the existence of a retroduodenal mass. Diagnosis requires a gastrografin upper GI series. Differential
diagnosis is mainly with biliary lesions and other causes of
peritonitis. Relaparoscopy may require intraoperative upper
GI endoscopy or Kochers duodenal mobilization to detect
the perforation. Early diagnosis allows primary repair, usually by laparoscopy. Perforations of the duodenal cap are
easier to diagnose and have a better prognosis than those of
the descending duodenum. A lumbar abscess is a frequent
complication.

A 45-year-old man presented with immediate severe and diffuse abdominal

pain and muscle contraction after laparoscopic cholecystectomy. Ultrasound examination showed free peritoneal fluid, and needle paracentesis
revealed bile, so percutaneous drainage was immediately obtained. On the
1st postoperative day, the patient underwent relaparoscopy, which confirmed choleperitoneum, but no bile leakage was noticed from the gallbladder fossa, cystic stump, or the main biliary duct. After a while, a large
perforation of the descending duodenum became evident. Laparotomy was
performed, the perforation was sutured, and a drain was placed beneath the
liver.
The postoperative course was still complicated by fever, even though
clinical examination of the abdomen was normal, the patient ate regularly,
and his bowel functions were normal. Abdominal ultrasonography was
normal, but CT revealed fluid collection behind the duodenum. Percutaneous drainage was unsuccessful but a right lumbar incision drained a
quantity of necrotic and purulent fluid. Thereafter, the patient healed successfully.

Key words: Laparoscopic cholecystectomy Duodenal

perforation Biliary lesions Retroperitoneal infection

Case 2

In our experience, duodenal perforations have never complicated the postoperative course of open cholecystectomies. It occurred only four times in 2,100 laparoscopic cholecystectomies (0.2%). Thus, it is as frequent as major biliary complications (three of 2,100, or 0.15%), while the
total complication rate was 1.85%. It is a rare, potentially
fatal complication, [3, 6] that is almost never reported.
However, we believe it needs to be better understood, in
order to reduce its occurrence and avoid dangerous diagnostic delays. In fact, this complication has certain interest-

A 48-year-old woman presented with immediate postoperative pain in her

upper right abdominal quadrant, vomiting, fever, hypotension, and tachycardia. Ultrasound showed subhepatic fluid collection, so she had relaparoscopy. Turbid bile was then aspirated from below the liver and the right
paracolic gutter. No bile leaked from the liver, cystic stump, or main bile
duct, and no other lesions were found. The procedure was concluded by
placing drains, one of them beneath the liver. The patient did not recover.
Sepsis continued, and nothing came out of the drains. One week after
relaparoscopy CT scan showed right lumbar fluid and gas collection, so we
drained the abscess via a right lumbotomy.
Her fever decreased, but the postoperative course was still troublesome.
The patient vomited what she ate, and finally bile appeared from the
subhepatic drain. Amylase concentration in that bile was 4,000 U/L. A
gastrografin upper gastrointestinal series showed slow gastric emptying
and a small fistula of the descending duodenum. The contrast medium
leakage out of the duodenum was almost missed, since it showed only for
a very brief time during the radiological examination. The patient was
treated conservatively, by total parenteral nutrition and nasogastric suction.
After somatostatin infusion was started, amylase concentration in the drain-

Correspondence to: M. Golia, Via dei Prati 7/5, 22060 Arosio, Italy

Case reports
Case 1

524
age fell to 4 U/L. The duodenal fistula healed after 15 more days, and the
patient was finally dismissed 2 months after cholecystectomy.

Case 3
A 50-year-old woman complained of postoperative abdominal pain, lowgrade fever, slight abdominal tenderness but no muscle contraction. Ultrasound examination revealed intraperitoneal fluid. She had relaparoscopy 2
days after cholecystectomy. Gastric juice mixed with bile was found in the
peritoneal cavity, and there was a perforation of the apex of the duodenal
cap. Once laparoscopic sutures and an omental patch were fixed, the patient recovered quickly.

Case 4
A 56-year-old woman had a scleroatrophic gallbladder that was detached
from the duodenum. She had immediate postoperative pain in her right
upper abdominal quadrant and fever. White blood cells and LFT were
normal, and amylasemia was 3,000 U/L. Plain abdominal X-rays showed
relevant pnumoperitoneum, and a gastrografin upper GI series revealed a
small perforation at the apex of her duodenal cap. She had relaparoscopy;
the site of the perforation was discovered only after irrigating the subhepatic region and performing intraoperative gastroduodenoscopy with air
insuflation. The small lesion was sutured by laparoscopy, and the patient
healed promptly.

Discussion
Iatrogenic duodenal perforations may have several causes.
In one of our cases, adhesions between a scleroatrophic
gallbladder and the duodenum may have played a role. We
believe that improper use of the 5-mm suction/irrigation
device caused the other three lesions. We also use that instrument to retract the duodenum caudally and to the left, to
facilitate dissection of Calots triangle. The distal extremity
of the device has a cut edge, so it must be used gently, by
its side, when retracting the duodenum. An associated perforation of the peritoneal membrane of the posterior abdominal wall close to the duodenum might lead to lumbar
abscess.
Other authors have reported perforations due to monopolar electrocautery [4] and laser dissection [1]. Electric
burns may occur without any contact between the active
electrode and the duodenal wall, due to stray currents [2].
During laparoscopic surgery, electric burns are likely to
occur, because the electrosurgical instruments are partially
out of the surgeons visual field.
The symptoms usually arise soon after surgery, except
in case of duodenal burns, when they occur abruptly 13
weeks after cholecystectomy [4]. Abdominal pain and tenderness either diffuse or at the right upper quadrant, were
always present in our cases. If a subphrenic abscess develops, pain at the right shoulder may be the only symptom for
a few days. Pain in that site may be overlooked since it is
common after laparoscopy, but its persistence should alert
the surgeon [1]. Right lumbar pain and tenderness should
prompt an examination for lumbar abscess. Abdominal palpation may or may not reveal signs of peritonitis. This is
especially true for perforations of the posterior wall of the
descending duodenum.
Abdominal ultrasound or CT may show diffuse intraperitoneal fluid or right upper quadrant collections. Paracentesis usually reveals bile. A biliary fistula may also show

Fig. 1. Abdominal CT showing liquid and gaseous retroduodenal collection.

Fig. 2. Gastrographin upper gastrointestinal radiograph showing duodenal
fistula draining through a subhepatic tube.

if subhepatic drainage was done after cholecystectomy. In

case 2, bile came out of the drain at a later time; this occurrence indicates that posterior perforations may in fact be
temporarily walled off. Any time bile is found, a biliary
lesion should usually be suspected, but duodenal perforation
must also be considered; and if there is any doubt about a
biliary lesion, a gastrografin upper GI series should be obtained before ERCP.
Although patients with posterior perforations may have
no intraperitoneal fluid, they may have liquid and gaseous
retroduodenal collections that can be seen on abdominal CT
(Fig. 1), which is more sensitive than ultrasound. Plain thorax or abdominal radiographs may show pneumoperitoneum, which is common after laparoscopy. However, its
magnitude or persistence should alert the physician to the
possibility of gastrointestinal perforation [1, 5]. Obliteration
of the right psoas muscle and evidence of retroperitoneal
gas are other signs that can be found on plain x-rays, but
they are difficult to appreciate. Amylase concentration may
be very elevated in the serum and in the fluid collected from
surgical or percutaneous drains. However, the main diagnostic tool is a gastrografin upper GI series (Fig. 2), which
needs to be accurate since small perforations can be difficult
to demonstrate. Indeed, false negative upper GI series have
been reported [1]. Perforation is a relative contraindication
for upper GI endoscopy, which nevertheless may be needed
in selected cases [1].

525

A therapeutic approach to duodenal perforation must be

taken as early as possible. Relaparoscopy allows the recognition and repair of the complication, but conversion to
laparotomy may be required. At reoperation, perforation
should be suspected whenever bile is found in the peritoneal
cavity and no leakage from the gallbladder fossa, the cystic
stump, or the main bile duct is found. Another suggestive
finding is anterior displacement of the duodenum by a posterior mass [4]. Duodenal perforation may require a meticulous research, by means of intraoperative upper GI endoscopy or duodenal mobilization by Kochers maneuver.
Thus, if you do not suspect it, you may miss duodenal
perforation, as in case 2. Usually duodenal sutures, an
omental patch, and drainage are sufficient. In cases of late
diagnosis or late perforations, more complex procedures are
needed, including duodenal drainage, pyloric temporary exclusion, gastrojejunostomy, and feeding jejunostomy. In
case 2, a small duodenal fistula with a delayed diagnosis
was treated successfully by conservative means. A right
lumbar abscess may complicate duodenal perforation, but
this has occurred only in cases of descending duodenum in
our experience. It might be the result of disruption of the
posterior peritoneal membrane during cholecystectomy or
reoperation for duodenal repair. It is probably also favored
by a late diagnosis of perforation, but in case 1 it occurred
after early duodenal repair. This condition usually does not
interfere with bowel functions and feeding. In this situation,
percutaneous drainage is not sufficient and lumbar incision
is required.
Conclusions
The first rule is that you must think about this complication;
otherwise, the consequences may be fatal. Even during reoperation, the diagnosis may be missed. Suspicion should

arise in cases of postoperative peritonitis (either diffuse or

localized), bile leakage, subphrenic abscess, lumbar pain
and tenderness with unexplained fever, elevated serum or
drainage amylase concentrations, retroperitoneal gas, retroperitoneal collections, retroduodenal mass, or unexplained
intraabdominal collections.
Free perforations of the duodenal cap are easier to diagnose and usually have an early and successful primary
repair by laparoscopy. On the other hand, perforations of the
descending duodenum may be walled off, and have poor
abdominal signs (pain, tenderness, muscle contraction, fluid
collection at ultrasound, pneumoperitoneum at radiograph).
They also may be complicated with a lumbar abscess and
end up with a late diagnosis and long secondary healing
processes. Last but not least, awareness of this complication
can help to reduce its occurrence by encouraging a more
appropriate use of mechanical and electrosurgical instruments. Specifically, sharp-pointed suction/irrigation devices
should never be used to retract the duodenum.
References
1. Berry SM, Ose KJ, Bell RH, Fink SA (1994) Thermal injury of the
posterior duodenum during laparoscopic cholecystectomy. Surg Endosc 8: 197200
2. Draper K, Jefson R, Jongeward R, McLeod M (1997) Duration of
postlaparoscopic pneumoperitoneum. Surg Endosc 11: 809811
3. Eden CG, Williams TG (1992) Duodenal perforation after laparoscopic cholecystectomy. Endoscopy 24: 790792
4. Peters JH, Gibbons GD, Innes JT, Nichols KE, Front ME, Roby SR,
Ellison CE (1991) Complications of laparoscopic cholecystectomy.
Surgery 110: 769778
5. Ress AM, Sarr MG, Nagorney DM, Farnell MB, Donohue JH,
Mellrath DC (1993) Spectrum and management of major complications of laparoscopic cholecystectomy. Am J Surg 165: 655662
6. Voyles CR, Tucker RD (1994) Unrecognized hazards of surgical electrodes passed through metal suction-irrigator devices. Surg Endosc 8:
185187