Recognizing Postoperative Intra-Abdominal Sepsis 
Richard Davis, Winnie Mutunga 
 
OPEN MANUAL OF SURGERY IN RESOURCE-LIMITED SETTINGS 
www.vumc.org/global-surgical-atlas 
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License  
 
Background:  
 
Postoperative intra-abdominal complications 
can be challenging to diagnose and treat for even the 
experienced surgeon. Many of the signs and 
symptoms that can signal infection in the abdomen, 
such as pain, tachycardia, and fever, will already be 
present or can be “blamed on” the postoperative 
state. The examination is also compromised; extreme 
tenderness and even peritoneal irritation are always 
present after surgery. And other, non-surgical 
conditions such as pulmonary embolism, pneumonia, 
wound infection, and urinary tract infection can 
cause signs and symptoms that are similar to intra-
abdominal infection.  
 
In this chapter we present our overall 
approach to patients after laparotomy. The surgeon 
must be very vigilant during the first 5-10 days after 
laparotomy, especially if a bowel anastomosis was 
done. During this time period, to borrow a quote 
from the author Chinua Achebe, “things fall apart.”  
 
Anatomy and Physiology: 
The Inflammatory Reaction: 
 
After surgery, the patient will undergo an 
inflammatory reaction. The greater the insult, the 
greater the reaction. This reaction is associated with 
surges of catecholamines, glucagon, and cortisol, the 
body’s stress reaction. Although we do not usually 
measure these substances, we can follow the 
inflammatory reaction. It is signaled by:  
● Tachycardia 
● Hypotension 
● Tachypnea 
● Low oxygen saturation 
● Low urine output 
● Peripheral edema 
● Ileus 
The most important concept in this chapter is 
the following one: It may be normal to have all of 
these signs for a certain period after the surgery, 
but they should not persist. The amount of time that 
these signs will “normally” be present varies 
depending on the magnitude of the surgical or 
traumatic insult. The surgeon must therefore watch 
all of these signs very carefully during the patient’s 
convalescence. 
This 
is 
because, 
after 
an 
uncomplicated 
surgery, 
the 
inflammatory 
reaction will be self-limited. When complications 
set in, signs of inflammation will either reappear, 
or fail to resolve.  
 
While we focus here on intra-abdominal 
complications, keep in mind that most forms of 
surgical 
illness 
will 
follow 
this 
course. 
Complications 
become 
apparent 
when 
the 
inflammatory reaction fails to resolve- this occurs 
with soft tissue infections, ischemia, and abscesses.  
 
Healing of Intestinal Anastomoses: 
 
When bowel edges are sutured together, the 
tissue follows a predictable course of healing. The 
surgeon must have an understanding of this process, 
especially regarding ways in which it can go wrong. 
There are two “windows” of time after creation of an 
anastomosis, when it is prone to leakage. 
In the presence of adequate blood supply, the 
bowel tissue becomes weakest between days 3-5. 
This is the inflammatory phase of wound healing 
and the main activity in the wound is lysis of 
collagen. During this time, macrophages and 
neutrophils invade the wound and remove dead 
tissue. Technical errors will manifest themselves 
during this time. These include failing to suture 
adequate tissue, including devitalized tissue in the 
anastomosis, or making the lumen of the anastomosis 
too small. Assuring that the serosa is approximated 
(when present) is an excellent way to assure that the 
anastomosis is “waterproof” and will last through 
this phase. This is why anastomoses to the rectum or 
esophagus, without serosa, are more prone to 
leakage.   
Another technical error is failing to 
incorporate all of the bowel in the anastomosis, in 
effect leaving a hole in the bowel. The best way to 
avoid this mistake is to pay strict attention to 
technical details while creating the anastomosis. This 
type of error will manifest itself around day 3-5 as 
well; enteric contents will start passing through the 
anastomosis at this time as ileus temporarily 
resolves. (Ileus will return, however, in the presence 
of leakage.) Therefore, days 3-5 constitute the first 
“window” for anastomotic leakage.  
The second “window” occurs around day 7-
12. An anastomosis that is performed technically 
well will be in the proliferative stage at this time. 
Problems occur if blood supply is inadequate for this 
stage; collagen deposition will be inadequate and the 
Recognizing Postoperative Intra-Abdominal Sepsis 
Richard Davis, Winnie Mutunga 
 
OPEN MANUAL OF SURGERY IN RESOURCE-LIMITED SETTINGS 
www.vumc.org/global-surgical-atlas 
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License  
 
healing process will stop, resulting in leakage. These 
“late leaks” are more rare, but they occur when 
anastomosis is done on bowel that is not adequately 
perfused. These patients may have return of bowel 
function before this happens. They might even be 
discharged from the hospital.  
The classic scenario where poor blood supply 
can lead to a late leak is a left colon resection and 
anastomosis in the left transverse colon, in the 
“watershed” area between the superior and inferior 
mesenteric arteries. In this area we usually resect 
further, preferring to remove more bowel until the 
anastomosis can be made using colon that is adjacent 
to two pulsatile vessels.  
For more details, see the Chapter on 
Intestinal Anastomosis. 
 
Principles:  
In large surgeries, the inflammatory reaction 
will involve the whole body, but should follow a self-
limited course. The careful surgeon will watch for 
signs that it is decreasing. The experienced careful 
surgeon, having cared for many patients after 
surgery, will know how long inflammation should 
last, based on the magnitude of the surgery.  
 
Diagram showing the normal pattern of inflammation after a 
large surgery. Of note, the amount of the inflammation (height 
of the curve) and the time at which it peaks will depend on the 
magnitude of the operation. An experienced surgeon will 
understand how this curve “should” look for a given operation, 
and when a patient is deviating from it.  
 
 
Diagram showing inflammation when there is unrecognized 
pathology, such as an intra-abdominal abscess or anastomotic 
leakage. In the patient represented by the Green arrow, the 
complication began before the inflammation could resolve. The 
patient represented by the Purple arrow developed a late 
complication, deviating from what would have otherwise been 
a normal postoperative course.  
 
 
The heart rate is the main and earliest signal 
of things falling apart. If the patient has been 
adequately resuscitated, tachycardia has usually 
resolved by the first postoperative day. Exceptions 
include patients who had sepsis before surgery (such 
as perforated duodenal ulcer) or patients who had a 
massive traumatic insult. But after a large elective 
surgery such as gastrectomy or colectomy, 
tachycardia should not be present on the first 
postoperative day. If it is, determine whether the 
patient’s pain is adequately controlled. If pain is 
controlled, give a fluid bolus and monitor the heart 
rate response. Check the hemoglobin level as the 
patient may be anemic. It is unlikely that anastomotic 
leak will manifest on the first day after surgery.  
 
Tachycardia which appears after the 3rd or 4th 
day in a patient who was previously well is more 
worrisome. Carefully assess the patient’s other 
parameters: measure their radial pulse yourself and 
assess the respiratory rate. If both are elevated, or if 
there is fever, this warrants further investigation. The 
first step is to talk to the patient. Have they passed 
flatus yet? Are they having difficulty urinating? 
Have they ambulated? Do they have a cough or 
shortness of breath? Then, do a thorough physical 
exam checking for crackles in the lungs or any 
tenderness anywhere, including the calves. Examine 
the wound for induration or erythema: sometimes the 
problem is a wound infection. Do not hesitate to open 
part of the wound if you suspect this.  
Recognizing Postoperative Intra-Abdominal Sepsis 
Richard Davis, Winnie Mutunga 
 
OPEN MANUAL OF SURGERY IN RESOURCE-LIMITED SETTINGS 
www.vumc.org/global-surgical-atlas 
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License  
 
If foul-smelling fluid comes out of the wound 
when opened, it may be a simple wound infection, or 
something worse. If bilious or feculent fluid come 
out, or if purulent fluid continues to drain, or if you 
feel a defect in the fascia in a patient with purulent 
wound discharge, return to the operating room for a 
careful exploration. See “Closure of Laparotomy 
Wound Dehiscence.”  
 
Unfortunately, in a postoperative patient the 
abdominal examination will not be very helpful. 
Nevertheless, palpate all of the abdomen gently, 
understanding that it will be painful to the patient. If 
the patient is cooperative, you can repeat this 
examination later to assess their progress; be 
reassured if there is some improvement.  
 
Resolution of ileus is a very reassuring sign 
that things are going well inside the abdomen. The 
colon and rectum recover from ileus last, so if the 
patient has flatus you can be sure that the stomach 
and small bowel are active. An exception to this rule 
is diarrhea: if the patient has diarrhea you have not 
ruled out anastomotic leakage or intra-abdominal 
abscess. Remember that if you give a patient water-
soluble oral contrast, they will often have diarrhea on 
the following day- this is normal. 
 
Plain abdominal or chest x-rays can be very 
useful. If the patient can stand or sit upright, have 
them do so for 3-5 minutes before taking a chest x-
ray, to allow any intraperitoneal air to rise into the 
subdiaphragmatic space. If the patient cannot stand, 
have them lie with their left side down for 3-5 
minutes before taking a lateral decubitus x-ray in that 
position (not supine!) The lateral decubitus x-ray 
takes advantage of the lack of bowel gas in the area 
of the liver- any gas seen in the left upper quadrant is 
likely outside the bowel.  
In general, the free air from an operation will 
resolve after 24 hours, 48 maximum. If you see free 
air in the abdomen after postoperative day 2, don’t 
fool yourself that it is left over from your operation: 
your patient’s anastomosis is leaking.  
 
Air under the right hemidiaphragm (Red arrow) in an upright 
chest x-ray. Sometimes the air may be a very small amount, but 
any air in this position is abnormal (unless it is clearly part of 
the transverse colon, which can rarely migrate above the liver.) 
Case courtesy of Dr Varun Babu, from the case 
https://radiopaedia.org/cases/19474?lang=us 
 
 
Sometimes the pneumoperitoneum is so large that the 
diaphragm can be confused for a lung marking, as shown by the 
Red arrow. One clue that this is the case is the unnatural 
appearance of the “false right hemidiaphragm,” which is 
actually the liver. Note also the clearly seen air under the left 
hemidiaphragm. Case courtesy of Dr Jeremy Jones, from the 
case https://radiopaedia.org/cases/6129?lang=us  
 
Recognizing Postoperative Intra-Abdominal Sepsis 
Richard Davis, Winnie Mutunga 
 
OPEN MANUAL OF SURGERY IN RESOURCE-LIMITED SETTINGS 
www.vumc.org/global-surgical-atlas 
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License  
 
 
The anteroposterior abdominal x-ray taken in the left lateral 
decubitus position takes advantage of the liver, which displaces 
any bowel. The area within the Red triangle is usually gasless 
on abdominal x-ray. In this case, a large collection of free 
intraperitoneal air is seen within this space, shown by the Red 
arrow. Case courtesy of Dr Prashant Mudgal, from the case 
https://radiopaedia.org/cases/33973?lang=us  
 
Decision Making: 
 
There are many things that could give your 
patient 
tachycardia 
or 
fever 
after 
surgery: 
pneumonia, urinary tract infection, pulmonary 
embolus, wound infection, and others. Although you 
would not want to miss any of these diagnoses, you 
should 
always 
take 
the 
following 
attitude: 
Increasing inflammatory markers in a patient 
after abdominal surgery is intra-abdominal sepsis 
until proven otherwise!  
 
A classic pitfall is blaming the patient’s 
deviation from “normal” on some other condition. 
You may feel tenderness in the calf and confirm a 
deep venous thrombosis by doppler ultrasound, 
leading you to believe that the tachycardia and 
hypoxia are due to pulmonary embolism. You may 
check a chest x-ray and see an infiltrate, leading you 
to believe that tachycardia and hypoxia are due to 
pneumonia. And yes, postoperative patients do get 
these problems. But they could just as easily be 
manifestations of a more serious intra-abdominal 
problem. The systemic inflammation caused by 
intra-abdominal 
sepsis 
leads 
to 
disordered 
coagulation and venous thrombosis. The abdominal 
pain from peritonitis leads to poor pulmonary 
hygiene and pneumonia.  
This is not to say that every postoperative 
complication is automatically caused by intra-
abdominal sepsis. Rather, we want you to be vigilant. 
After any operation that leaves a suture line in the 
abdomen, anastomotic leakage should always be at 
or near the top of your list of suspects.  
When a postoperative patient has new onset 
tachycardia, fever, elevated respiratory rate, or 
elevated oxygen requirement, check a white blood 
cell count and differential. An elevated WBC with a 
“left shift” (increased neutrophil percentage) is 
diagnostic for untreated infection, unless the values 
are decreased compared to the admission values. At 
this point you may either do imaging, if this is 
available to you, or explore the abdomen if no other 
diagnosis is apparent.  
Overall, we are looking for early signs of 
inflammation, so we can intervene before the 
inflammation becomes life-threatening. Bear in mind 
that these signs we watch for are even more subtle in 
patients who can not mount an immune response. 
Such 
patients 
include 
the 
elderly, 
the 
immunosuppressed, and those who will not become 
tachycardic because of beta blockade or heart block. 
Conversely, patients on systemic steroids will have 
an elevated white blood cell count and left shift, 
which may not reflect infection.  
To make the diagnosis, if possible obtain an 
x-ray study that shows enteric contrast passing 
through the anastomosis. This is possible with 
anastomoses in the rectum, rectosigmoid junction, 
esophagus, stomach and duodenum. Use water-
soluble contrast such as gastrografin (meglumine-
diatrizoate.) Barium will cause mediastinitis or 
peritonitis if it comes in contact with these surfaces. 
If you have fluoroscopy, these studies can be done 
under your supervision and you can make sure you 
are satisfied with the quality of the images. 
Otherwise, instruct the x-ray staff on what part of the 
gastrointestinal tract you are hoping to see, and they 
will time their images accordingly.  
If you have a CT scan available, these images 
will be more useful than plain x-rays.  If you can give 
IV contrast, you can distinguish abscesses from non-
infected fluid collections by the presence of rim 
enhancement in an abscess. But you do not need to 
give IV contrast, as the most important thing you are 
looking for is active leakage of enteric contrast, or 
any of it outside the bowel. 
 
Recognizing Postoperative Intra-Abdominal Sepsis 
Richard Davis, Winnie Mutunga 
 
OPEN MANUAL OF SURGERY IN RESOURCE-LIMITED SETTINGS 
www.vumc.org/global-surgical-atlas 
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License  
 
Case Studies:  
Case #1: JA, 60 year old otherwise healthy woman 
underwent 
an 
Antrectomy 
and 
Bilroth 
1 
Reconstruction for obstructive peptic ulcer disease. 
Surgery was uneventful. She has not had flatus 
postoperatively but has been afebrile, with heart rate 
consistently <90 beats per minute. The surgical team 
which performed her operation has seen her every 
day.  
 
On postoperative day 5, her heart rate is noted 
to be 115 beats per minute. Her blood pressure is 
110/80. She is not nauseated but has not yet passed 
flatus. Her urine output has not been recorded, as her 
foley catheter was removed several days ago.  
 
On 
examination, 
her 
abdomen 
is 
nondistended. There is mild generalized tenderness. 
Her wound is clean and dry without erythema or 
discharge, and her abdominal drain output is 
minimal, clear yellow non-bubbly fluid. Her sclerae 
are noted to be normal in color and not pale. Lung 
auscultation does not reveal crackles or diminished 
lung sounds. She is given a bolus of 500mL of 
Ringer’s Lactate. At this time, the following 
differential diagnoses are considered:  
● Intra-abdominal abscess 
● Anastomotic leakage 
 
 
The team orders a complete blood count and 
a gastrografin (water soluble contrast) swallow 
study. They discuss with the radiography team and 
ask them to obtain images immediately after 
administering 50mL of gastrografin, as the area of 
interest is the stomach.  
The images clearly show leakage of contrast, 
emptying into a small area that is evacuated by the 
drain:  
 
Swallow study with water-soluble contrast shows the site of 
leakage from the anastomosis (Black arrow) and the collection 
of contrast adjacent to the tip of the drain, with contrast within 
the drain (Red arrow.) Source: Dong Woo Hyun, Ki Hyun Kim 
et al, https://doi.org/10.7602/jmis.2018.21.1.13 
 
The complete blood count returns the 
following results:  
White blood cells: 22.26 
Neutrophils: 92% 
Hemoglobin 8.8g/dL 
Platelets 667 
 
 
At this time the patient is explained the 
complication and she agrees to return to the operating 
room. Excessive tension on the gastroduodenostomy 
is found, with an anastomotic dehiscence of about 
20% of the circumference. The entire anastomosis is 
taken down and converted to a Roux-en-Y 
reconstruction, and a feeding jejunostomy is placed 
distally. 
The 
abdomen 
is 
washed 
out. 
Postoperatively, she is started on IV antibiotics. 
 
 
 
Comment: This team acted well, seeing the 
patient daily and recognizing an early sign of intra-
abdominal sepsis. Note also that they did not add 
“Pneumonia” or “Urinary Tract Infection” to their 
differential diagnosis- they correctly assumed that 
the cause of sepsis was the site of their operation. 
They did not start the patient on antibiotics and hope 
she would improve, a move that would have only 
delayed recognition and treatment of this problem.   
Recognizing Postoperative Intra-Abdominal Sepsis 
Richard Davis, Winnie Mutunga 
 
OPEN MANUAL OF SURGERY IN RESOURCE-LIMITED SETTINGS 
www.vumc.org/global-surgical-atlas 
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License  
 
 
Some learning points from this case: 
● The earliest sign of infection after intra-
abdominal surgery will be tachycardia. An 
elevated  WBC with tachycardia must be 
taken seriously.  
● The diagnosis is an intra-abdominal process 
that 
needs 
intervention 
until 
proven 
otherwise. Do not assume it’s pneumonia. 
● If you do not have a CT scanner, it is better 
to re-operate early when you suspect sepsis. 
If you wait to intervene, surgery will be much 
more dangerous for the patient, as the 
inflammatory cycle will have been ongoing. 
In this patient, all other signs pointed to an 
intra-abdominal process. If the x-ray 
swallow study did not show an anastomotic 
leak, it would have been reasonable to re-
explore the patient anyway.  
● The lack of purulent discharge from the drain 
does not rule out anastomotic leakage.  
 
Case #2: BD, a 70 year old man with a past history 
of asthma and bronchitis, undergoes elective left 
hemicolectomy for non-obstructing colon cancer. A 
stapled anastomosis is performed. His morning and 
afternoon vital signs are as follows:  
 
Postop. Day 
Heart 
Rate 
Resp. 
Rate 
Temp(C) 
3 (6AM) 
81 
19-24 
37 
3 (4PM) 
87 
22-24 
37 
4 (6AM) 
112 
21-24 
38.1 
 
On the morning of Postoperative Day 4, as 
above, the patient’s heart rate is noted to be 112 and 
temperature elevated to 38.1. The blood pressure is 
within normal limits. He is not nauseated and has not 
yet passed flatus. Urine output is unknown, as the 
foley catheter was removed the previous day.  
On 
examination, 
his 
abdomen 
is 
nondistended. There is mild generalized tenderness. 
The wound is clean and dry without erythema or 
discharge. His sclerae are noted to be normal in color 
and not pale. Lung auscultation reveals decreased 
breath sounds on the left side. A plain chest x-ray 
shows no sub-diaphragmatic air and a small 
consolidation of the left lower lobe. A diagnosis of 
pneumonia is made and the patient is started on 
piperacillin-tazobactam.  
Morning and afternoon vital signs continue 
as follows:  
 
Postop. Day 
(Time) 
Heart 
Rate 
Resp. 
Rate 
Temp(C) 
4 (4PM) 
89 
18-22 
37.4 
5 (6AM) 
110 
17-27 
37.7 
5 (4PM) 
115 
18-26 
37.4 
6 (6AM) 
118 
22-30 
38 
 
 
On Postoperative Day 6 as above, the patient 
is clearly unwell with tachypnea and hypotension. 
His abdomen is distended and peritonitic. He has not 
yet passed flatus. The team obtains a chest x-ray, 
which shows:  
 
Plain upright chest x-ray shows free air under both the right 
and left hemidiaphragm. Case courtesy of Knipe, H, from the 
case https://radiopaedia.org/cases/25402  
 
 
A diagnosis of anastomotic leakage is 
considered. The team obtains a CT scan with 
intravenous and rectal water-soluble contrast, which 
shows:  
Recognizing Postoperative Intra-Abdominal Sepsis 
Richard Davis, Winnie Mutunga 
 
OPEN MANUAL OF SURGERY IN RESOURCE-LIMITED SETTINGS 
www.vumc.org/global-surgical-atlas 
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License  
 
 
CT scan of the upper abdomen shows free air and fluid anterior 
to the liver (Black arrow) and posterior to the spleen (Red 
arrow.) Case courtesy of Knipe, H, from the case 
https://radiopaedia.org/cases/25402  
 
 
CT scan of the lower abdomen shows a small amount of 
contrast extravasation (Black arrow) adjacent to the 
anastomosis, whose location can be clearly seen because of the 
staples (Red arrow.) Case courtesy of Knipe, H, from the case 
https://radiopaedia.org/cases/25402  
 
The patient is taken for laparotomy and 
washout. He is noted to be hypotensive and 
tachycardic 
throughout 
the 
operation. 
The 
anastomosis is taken down and a Hartmann 
colostomy is fashioned. He requires ventilator 
support for 1 week after surgery and his 
hospitalization lasts 5 weeks in total. He eventually 
recovers.  His hospital bill is far in excess of what his 
family is able to pay at the time of discharge.  
 
Comment: It is quite clear from a review of 
the vital signs that this patient’s anastomotic leak 
began on the night between Postoperative Day 3 and 
4. The signs were there for the team to recognize but 
instead they made a diagnosis of “Pneumonia.” 
Once this was done, they were perhaps reassured by 
a transient decrease in heart rate on the afternoon of 
Postoperative Day 4. The overall effect of this 
misdiagnosis was a delay of 48h. During this time, 
the anastomosis continued to leak and the sepsis got 
worse.  
If the team had chosen to investigate on 
Postoperative Day 4 with a water-soluble contrast 
enema x-ray series, or a CT scan with rectal 
contrast, it is possible that they could have repaired 
the anastomotic leak with an omental patch or 
otherwise avoided a colostomy. Also, if they had 
operated on Day 4 instead of 6, it is certain that the 
patient would have been more fit for operation, and 
would not have needed prolonged intubation 
afterwards.  
 
Some learning points from this case:  
● The earliest sign of infection after intra-
abdominal surgery will be tachycardia. A 
fever with tachycardia must be taken 
seriously.  
● The diagnosis is an intra-abdominal process 
that 
needs 
intervention 
until 
proven 
otherwise. Do not assume it’s pneumonia. 
● A plain chest or abdominal x-ray that shows 
free intra-abdominal air >48 hours after 
surgery is all the proof you need of 
anastomotic leakage. No other imaging 
studies 
are 
necessary 
at 
this 
point. 
Conversely, lack of free air on plain x-ray 
does not rule out anastomotic leak. 
● Delaying intervention for abdominal sepsis 
can lead to a catastrophic result, including 
sepsis and prolonged ICU stay, prolonged 
overall hospitalization, elevated hospital 
expenses, and often the patient’s death.  
 
 
Recognizing Postoperative Intra-Abdominal Sepsis 
Richard Davis, Winnie Mutunga 
 
OPEN MANUAL OF SURGERY IN RESOURCE-LIMITED SETTINGS 
www.vumc.org/global-surgical-atlas 
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License  
 
Richard Davis MD FACS FCS(ECSA) 
AIC Kijabe Hospital 
Kenya 
 
Winnie Mutunga 
AIC Kijabe Hospital  
Kenya 
 
October 2022 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Resource-Rich Settings 
CT scan of the neck, chest, or abdomen is extremely helpful 
for investigating postoperative patients. Oral contrast can be 
given and then the scan is timed according to the suspected 
area of leakage:  
For esophagus, stomach or duodenum the 
images are taken within minutes of oral 
administration.  
For small or proximal large bowel, the images 
are taken 30-60 minutes after oral administration.  
For rectal or left sided colon, contrast is 
administered by enema.  
If the patient’s renal function is adequate, IV 
contrast can be administered as well, which helps 
diagnose abscesses and plan percutaneous drainage. 
IV contrast should be avoided in patients with 
sepsis.  
 
Image-guided percutaneous drainage can avoid an operation 
for an abscess. Even in the case of anastomotic leakage, if the 
fluid collection is small, this is standard treatment in places 
where it is available. In such cases, postoperative nutrition is 
crucially important. 
 
Endoscopic clip application has revolutionized the 
management of anastomotic leakage: a skilled endoscopist 
with the right equipment can often help the patient avoid an 
operation altogether. Any intra-abdominal fluid can then be 
managed by percutaneous drainage if necessary. In the cases 
discussed, this modality would have been effective on Case 
#1 and possibly on Case #2 if the diagnosis had been made 
l
h
