Peritonitis--Reducing Mortality and Morbidity
World Small Animal Veterinary Association World Congress Proceedings, 2006
Richard A. Read, BVSc(Hons), PhD, FACVSc
Professor, Small Animal Surgery, Murdoch University, Murdoch, Western Australia, Australia


Peritonitis in dogs and cats can be classified in a number of ways: primary vs secondary, septic vs aseptic, acute vs chronic or diffuse vs localised. Most cases of peritonitis are secondary to gastrointestinal leakage and are septic. Other sources of sepsis in septic peritonitis include the urogenital tract (ruptured pyometra, ruptured prostatic abscess, ruptured bladder or renal abscess), hepatobiliary tract (necrotizing cholecystitis, liver lobe torsion or liver abscess), or from penetrating wounds, pancreatic or splenic abscesses. Reviews cases of generalised peritonitis in dogs showed an overall mortality rate of between 30 and 68%.

Clinical Signs and Pathophysiology

Clinical signs of septic peritonitis vary significantly between patients depending on the duration and extent of the peritoneal contamination and the type and number of bacteria present. However, it can be assumed that all these patients are somewhere on a spectrum the end point of which is septic shock and multiple organ dysfunction. The peritoneum has enormous absorptive capacity but consequently has equally enormous capacity to deliver fluid into the abdominal cavity when the normal balance is disturbed. The following sequence can rapidly develop:

 Sepsis results in vasodilation and increased capillary permeability, which delivers fluid and plasma proteins into the abdominal cavity.

 Further fluid and protein losses may occur due to anorexia, vomiting and diarrhoea.

 The resulting hypovolaemia results in decreased cardiac output, poor perfusion, hypoxia and cell death.

 Ischaemia and inflammation of intestinal mucosa compromises its integrity and translocation of bacteria occurs.

 Influx of neutrophils and macrophages which release cytokines and other active substances, along with bacterial toxins, further exacerbate peripheral vasodilation and capillary permeability.

 Cardiac function is compromised, the coagulation cascade is activated and the systemic inflammatory response syndrome (SIRS) is fully activated.

Blood samples reveal leucocytosis with left shift, hypoproteinaemia and hypoglycaemia, along with other specific signs related to the organ systems involved.


There is usually a high suspicion of septic peritonitis based on clinical signs and history, and once an abdominal effusion is suspected, obtaining a sample of this fluid for cytology, culture and chemical analysis is essential. Abdominocentesis should be performed first and if initially unsuccessful, ultrasound guidance and repeated sampling in different areas may improve yield rates. If still negative, diagnostic peritoneal lavage can be performed: 15-20 ml/kg of warmed 0.9% NaCl is infused into the abdominal cavity and the patient is rolled from side to side. Fluid is then collected by gravity and submitted for culture, cytology and chemistry. The presence of degenerate neutrophils along with intracellular bacteria is grounds for exploratory surgery.

Various chemical analyses of abdominal effusion fluid have been evaluated in dogs to try to find the definitive diagnostic test for septic peritonitis. The most accurate measure appears to be the differential between blood glucose and abdominal fluid glucose--100% of dogs with abdominal fluid glucose more than 20mg/dl lower than the blood glucose had septic peritonitis.

Emergency Treatment

Aggressive fluid therapy with crystalloids and colloids is indicated in most patients, with vasopressors and inotropes used where hypotension persists. Blood or plasma transfusions may be indicated, especially if coagulation abnormalities are present.

Surgical Management

There is a small window of opportunity for timing surgical intervention in many patients with septic peritonitis. Some patients will respond dramatically to resuscitatory therapy but if the diagnosis has been made, surgery should proceed as soon as possible to find the source of contamination. As stated earlier, the source of contamination in most cases is the gastrointestinal tract, so this should be the first focus of attention, starting at the large bowel and working proximally.

Once the site of leakage has been definitively dealt with, large volume lavage with isotonic warmed fluid is essential to dilute the bacteria and remove particulate contaminating matter. How much lavage is a matter of judgment--the aim is to clear all particulate matter but there is a down side to peritoneal lavage--mesothelial lining cells are damaged, the ability of the peritoneal defense mechanisms to clear bacteria may be further reduced, and in some cases the infection can be further spread in cases where the peritonitis was being somewhat contained in one area of the abdomen. What is clear is that as much of the lavage fluid as possible should be removed as possible to minimise these issues. The addition of antibiotics or antiseptics to the lavage fluid has not been shown to be beneficial.

To Drain or not to Drain?

The need to provide ongoing drainage of the abdominal cavity in cases of septic peritonitis is a very controversial topic. Following the trend in human medicine in the 1980's, open peritoneal drainage became the preferred method for some surgeons wishing to provide ongoing postoperative drainage of septic peritonitis in dogs and cats. This was partly fuelled by studies in normal dogs showing that drainage of the peritoneal cavity with sump penrose drains was rapidly compromised by adhesion formation around the drains. However, there is some evidence to suggest that these adhesions may not obstruct the drainage of fluid. In an experimental study in normal dogs, sump-penrose drainage and open peritoneal drainage were established in two separate groups via laparotomy. Fluid was then infused into the abdomen over 72 hours and the volume of drainage was estimated by weighing the bandage covering the ventral abdomen. It was shown that despite developing significant adhesions around them, the sump-penrose drains were just as efficient at allowing drainage as the open peritoneal technique.

In open drainage, the ventral abdominal incision is left partly open at the end of exploratory surgery and covered with sterile dressings to allow continued drainage by gravity. Regular bandage changes are required which may require sedation or anaesthesia, but these changes also provide the opportunity to either infuse fluid into the abdomen for ongoing lavage, or introduce a sterile gloved finger through the incision to break down any adhesions. The duration of open drainage can vary between 2-14 days. Complications can include further contamination of the abdominal cavity and serious loss of protein and electrolytes; disadvantages of the technique focus around the need for intensive monitoring and the cost.

Early studies of open peritoneal drainage showed promising results, but the ensuing published studies varied both in duration of drainage and technique used. Overall, the mortality rate for cases treated by open drainage does not appear to be any different from that of cases in which the abdomen is closed at the end of surgery. A recent study evaluated postoperative suction drainage using Jackson Pratt drains placed between the liver and the diaphragm and in this group of 40 patients the mortality rate was 30%. Those advocating closed drainage or no drainage suggest that since the outcome is possibly no different, the complications of open peritoneal drainage are sufficient reason not to use it.

The decision about whether to leave an abdomen open at the end of surgery for septic peritonitis still rests with the surgeon and will be governed by:

 The severity of the peritonitis

 The surgeon's personal experience and preference

 The finances of the owner

 The availability of postoperative intensive monitoring

 The serum albumen of the patient

 The temperament of the patient

Other Supportive Treatment

Providing adequate nutritional support for patients with septic peritonitis is essential. Consideration should be given to the placement of enteral feeding tubes at the time of definitive surgery. These should be placed so as to maximise the use of whatever normal gastrointestinal tract is available. Alternatively, total parenteral nutrition may be preferred in some patients.

Antimicrobial therapy will be based on results of culture and sensitivity but in most cases combination therapy will be indicated initially to provide cover against both gram positive and gram negative aerobes and against anaerobic bacteria.


As stated earlier, the published mortality rates for septic peritonitis in dogs and cats ranges from 30-68%, with no clear evidence for any particular post-operative management system offering superior results. The cause/source of the contamination does not appear to have a significant impact on prognosis, although there is some evidence that septic bile peritonitis carries a poor prognosis--in one study of 26 dogs and cats with bile peritonitis, only 27% of patients with septic bile peritonitis survived whereas 100% of patients from which no bacteria could be detected in the abdominal fluid survived. It has been postulated that the presence of bacteria and bile salts impairs peritoneal phagocytic activity and increases lipopolysaccharide levels.

There are no studies that have identified any other clinical or diagnostic parameters that offer a clear guide to prognosis in cases of septic peritonitis. The development of a standardized diagnostic and therapeutic protocol in a prospectively designed study is possibly necessary to identify any potential prognostic indicators.


1.  Hosgood and Salisbury. JAVMA 193: 1448-1450, 1988.

2.  Hosgood, Salisbury & Denicola JAAHA 27: 115-121, 1991.

3.  Lanz, Ellison, Bellah et al. JAAHA 37:87-92, 2001.

4.  Ludwig, McLoughlin, Graves & Crisp. Vet Surg 26:90-98, 1997.

5.  Mueller, Ludwig & Barton. JAVMA 219: 789-794, 2001.

6.  Staatz, Monet & Seim. Vet Surg 31:174-180, 2002.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Richard A. Read, BVSc(Hons), PhD, FACVSc
Murdoch University
Western Australia, Australia