The peritoneal cavity is lined by the peritoneal membrane that consists of a single layer of mesothelial cells supported by loose connective tissue. The membrane comprises two components in continuity: the visceral peritoneum covers the abdominal organs and the parietal peritoneum lines the abdominal wall (Fig. 1) 1189. The cavity is thus a closed sac except for the fimbriated ends of the fallopian tubes. The mesothelial cells secrete serous fluid which provides a lubricant to allow gliding movements of the viscera. In the normal healthy adult the peritoneal cavity contains about 100 ml of serous fluid, secreted from a total peritoneal surface area of approximately 2 m². This large surface area of thin semipermeable peritoneal membrane can be utilized for peritoneal dialysis.

The technique of peritoneal dialysis has been used in the treatment of renal failure since the late 1940s but was of limited intermittent application until the development of the implanted Tenckhoff silicon rubber catheter in 1968 (see Section 7.16 61). This allowed peritoneal dialysis to be performed continuously and safely for longer periods commonly of several years' duration. Dialysis fluid is stored in sterile PVC bags and during an ‘exchange’ the bag is connected to the Tenckhoff catheter by PVC tubing. The dialysate is encouraged to enter the abdominal cavity by gravity, through elevation of the bag. Between three and five exchanges of 1 to 2 litres of peritoneal dialysate are performed daily. Clearance of particular electrolytes is achieved by reducing their concentration in the dialysate. Fluid balance is monitored by the measurement of daily weights and measured by altering the proportion of ‘hypertonic’ (3.8 per cent glucose solution) and ‘isotonic’ (2.36 per cent glucose solution) exchanges. While the dialysis fluid is in the peritoneal cavity, the uraemic toxins diffuse from the blood supply of the mesentery and peritoneum. At the end of each dialysis cycle the dialysate is drained by gravity and returned to the bag. This simple technique of continuous ambulatory peritoneal dialysis (CAPD) has been adopted in many dialysis centres as initial therapy for renal failure in diabetics, in young patients with the prospects of any early transplant, and in those patients with poor vascular access.

The pattern of innervation of the visceral and parietal peritoneum determines the pattern of abdominal pain that occurs when the peritoneum is inflamed. As the visceral peritoneum receives only afferent innervation from the autonomic nervous system, stimuli from the visceral peritoneum are localized only to the foregut, midgut, or hindgut in the midline. The visceral nerves do not mediate pain or temperature but respond to distension, traction, or pressure. In contrast, the parietal peritoneum is innervated by both somatic and visceral afferent fibres which allows the accurate localization of pain when the parietal peritoneum is inflamed.

The anatomy of the abdominal cavity determines the pattern of fluid movement within the abdomen. Functionally, four compartments exist within the peritoneal cavity: the pelvis, the right and left paracolic gutters, and the infradiaphragmatic space (Fig. 1) 1189. As the paracolic gutters slope into the infradiaphragmatic space superiorly and over the pelvic brim inferiorly, fluid in the recumbent patient collects under the diaphragm and in the pelvis, the common sites for abscess formation (see Section 21.2 153).

Peritonitis is defined as inflammation of the peritoneal cavity, where the peritoneal fluid increases in volume with the passage of a transudate rich in leucocyte polymorphs and fibrin. Initially, peritoneal inflammation is often localized and the affected area contained by a wrapping of greater omentum, adjacent bowel, and fibrinous adhesions. If the inflammatory focus is part of an ongoing process, or if host defences are lowered, localized peritonitis may progress to life-threatening generalized peritonitis. There is massive exudation of inflammatory fluid into the peritoneal cavity causing hypovolaemia, often compounded by toxaemia from absorbed products and septicaemia if infection is present. Diffuse peritoneal irritation causes peristaltic paralysis with the cessation of bowel motility.

The clinical features of peritonitis are dependent on both the aetiology and the progression of the inflammation. The early manifestations of peritonitis following disease of an abdominal viscus are characterized by the primary disease process itself and a detailed discussion can be found in the appropriate chapter. An overview of the signs and symptoms of the acute abdomen are detailed in Chapter 28 154. In summary, the signs and symptoms of peritonitis secondary to a diseased abdominal viscus are as follows: initial inflammation of the visceral peritoneum overlying the damaged organ results in poorly localized abdominal pain, through stimulation of the visceral autonomic nerves. Irritation of the nearby parietal peritoneum results in localization of the pain. Associated symptoms include malaise, nausea and vomiting, and an associated low-grade fever. The four cardinal signs of peritonitis, consisting of tenderness, guarding, rigidity, and rebound tenderness may also be elicited. When the peritonitis is generalized the patient is clearly unwell, with marked fever and dehydration and absent bowel sounds. Pain is diffuse throughout the abdomen and is exacerbated by even the slightest movement. Shoulder-tip pain is diagnostic of diaphragmatic inflammation.

Acute peritonitis may be classified as (a) primary (spontaneous), where an infection has arisen de novo within the peritoneum or (b) secondary where the inflammatory process involving the peritoneum is the result of an identifiable primary process.

Idiopathic peritonitis is uncommon, constituting about 1 per cent of all cases of peritonitis and occurs when no obvious source for the peritoneal infection can be demonstrated. It was classically described in young girls where the port of entry was presumed to be through the fallopian tube or via haematogenous spread. Formerly, pneumococci were implicated but in recent years haemolytic streptococci, Escherichia coli, and Klebsiella spp. are now more frequently cultured.

Other causes of spontaneous peritonitis are associated with patients suffering intercurrent disease. Pneumococcal infection is well recognized in the postsplenectomy child and in those who have nephrotic syndrome. Adults with cirrhosis are prone to a wide variety of spontaneous bacterial infections because of the proteinaceous nature of the ascitic fluid which provides an excellent culture medium for blood-borne bacteria to proliferate. Successful treatment of primary peritonitis, without resorting to laparotomy is rare. Only when peritoneal tap and culture of peritoneal fluid reveals a non-enteric organism can conservative antibiotic therapy be instituted with caution.

This is the most common form of peritonitis encountered by the surgeon and results from the perforation of a viscus (e.g. appendix, peptic ulcer, colonic diverticulum, or gallbladder), ischaemia of an intra-abdominal organ (e.g. strangulated hernia, volvulus, mesenteric artery occlusion), or extension of an existing infection of an abdominal organ (e.g. appendix abscess, liver abscess, pyosalpinx). Surgical intervention is the treatment of choice and in most cases is mandatory, being aimed at the primary disease process. Supportive therapy before laparotomy includes opiate analgesia, antibiotic therapy (against both aerobes and anaerobes), correction of hypovolaemia and electrolyte imbalance by intravenous fluid replacement, and nasogastric aspiration to reduce abdominal distension and prevent aspiration pneumonia through inhalation of vomit.

The presence of chronic granulomata within the abdominal cavity in association with peritonitis is the result of infection, most commonly with tuberculosis but occasionally with fungi such as Candida albicans. Tuberculous peritonitis is usually associated with a tuberculous focus within the abdomen (e.g. intestine, mesenteric lymph node, fallopian tubes) but occasionally can develop by haematogenous spread from a focus outside the abdominal cavity, such as in the lung. The causative agent is Mycobacterium tuberculosis, nowadays usually of the human rather than the bovine strain, following widespread pasteurization of milk. Although the incidence of the disease is low in Western countries, occurring in the chronically ill or malnourished patient, the disease is still prevalent in the Indian subcontinent and in the Far East. Immigrants from these locations remain susceptible to the condition. Females are twice as commonly affected as males but all age groups are involved.

Pathologically, tuberculous peritonitis may present with ascites or with intra-abdominal adhesions but there is often considerable overlap between the two forms of the disease. Clinical presentation is insidious with vague generalized abdominal pain, low-grade fever, anorexia, and weight loss. The ascitic patient may also present with gross abdominal distension while those with intra-abdominal adhesions may present with pain and distension or bowel obstruction. Preoperative diagnosis is uncommon. Tuberculous ascites is difficult to differentiate from other forms of ascites. The leucocyte count may demonstrate a lymphocytosis and lymphocytes also predominate in the ascitic fluid. Centrifugation and direct microscopy of the ascitic fluid rarely demonstrate the tubercle bacillus. Culture of ascitic fluid is positive for tuberculosis in less than one-half of all cases. At laparotomy, the entire peritoneal cavity is usually studded with tuberculous nodules and the initial differential diagnoses include carcinomatosis, starch peritonitis, and widespread fat necrosis. In the patient who presents with subacute or acute bowel obstruction, widespread fibrinous adhesions with matting of intestines and omentum predominate. Here the differential diagnoses includes other forms of sclerosing peritonitis (see under Chronic Peritonitis). Confirmation of tuberculous peritonitis is made by histopathology and treatment consists of combination chemotherapy (rifampicin, isoniazid, and ethambutol) for up to 12 months.

Aseptic peritonitis refers to the peritoneal inflammation from substances other than bacteria but bacterial contamination and overgrowth soon follow. A perforated peptic ulcer provides the most severe and common form of chemical peritonitis with gastric juice and bile contaminating the peritoneal cavity. Biliary peritonitis alone may follow gangrene and perforation of the gallbladder, or, after cholecystectomy, may be the result of unrecognized division of an accessory hepatic duct, an insecure ligature on the cystic duct remnant or displacement of a T-tube following exploration of the common bile duct. Blood in the peritoneum is also a cause of peritoneal irritation after slow bleeding (e.g. a ruptured graafian follicle or following splenic injury) rather than from a catastrophic haemorrhagic event such as a ruptured aneurysm where the primary pathology itself overshadows the peritoneal irritation. Meconium and urine may also precipitate chemical peritonitis.

Endoscopy of the gastrointestinal tract may precipitate acute peritonitis through colonoscopic perforation of a diverticulum or inadvertent perforation of the oesophagogastric junction during oesophageal dilatation. Similarly, the urinary bladder may be perforated during diagnostic or therapeutic cystoscopy leading to a urinary leak within the abdomen. The recent expansion of interventional radiological procedures has precipitated a multitude of assaults on the abdominal cavity, such as CT guided biopsy and drainage of abscesses, mesenteric angiography and therapeutic embolization, and percutaneous transhepatic cholangiography and stenting, all providing further potential for peritonitis.

Peritonitis may follow abdominal surgery where bowel and gastric contents, blood, and urine escape into the abdominal cavity following anastomotic dehiscence. In patients with renal failure treated by continuous ambulatory peritoneal dialysis, a permanent indwelling catheter in the abdominal cavity provides a portal of entry for exogenous bacteria despite the use of stringent aseptic techniques during dialysis exchanges (see Chapter 8 155).

Abdominal trauma may produce acute peritonitis in several ways. Penetrating wounds of the abdomen without visceral injury may provide a route for exogenous bacterial contamination. Penetration of a visceral organ may precipitate the spillage of visceral contents into the peritoneal cavity. Severe blunt trauma may disrupt intra-abdominal organs directly or indirectly through disruption of their vascular supply.

Warfarin anticoagulation can cause peritoneal irritation and peritonitis through leakage from a spontaneous retroperitoneal haematoma. The symptoms of acute peritonitis have also been described during treatment with the antituberculous agent, isoniazid.

Chronic or sclerosing peritonitis is the end result of a heterogenous group of conditions (Table 1) 357. Classically, sclerosing peritonitis is characterized by dense adhesions, especially between loops of small bowel, and in the most extreme cases the entire small bowel and even the large intestine and liver is cocooned in a dense adhesive membrane of fibrous tissue (Fig. 2) 1190. The classical description of sclerosing peritonitis relates to the &bgr;-adrenoceptor blocker practolol. This drug was first introduced in 1970 but the first reports of thickening of serosal membranes appeared 4 years later and the drug was withdrawn from general use in 1976. During this 6-year period, approximately 200 patients worldwide were diagnosed as having sclerosing peritonitis. Patients presented with subacute small bowel obstruction or acute-on-chronic small bowel obstruction. Treatment consisted of surgical stripping of the cocoon of fibrous tissue from the underlying intestine and was a long and tedious procedure. Intra-abdominal fistula formation and death was not uncommon. In the early 1980s sclerosing peritonitis made a reappearance in patients undergoing continuous ambulatory peritoneal dialysis. Patients presented with impaired ultrafiltration capacity due to the deposition of the fibrous membrane on the peritoneum, and the clinical features of bowel obstruction, either acute or chronic. Epidemiological studies demonstrated that the sclerosis was due to the chlorhexidine and alcohol solution used to sterilize the connectors of the CAPD catheters at the time of peritoneal dialysis ‘exchanges’. Approximately 1 ml of antiseptic solution was able to enter the peritoneal cavity at the time of each exchange causing irritation to the peritoneal membrane. Other mechanisms which may precipitate sclerosing peritonitis in CAPD patients include the presence of acetate in the dialysate and recurrent CAPD peritonitis with fibrin deposition over the peritoneal membrane.

Foreign-body granulomatous peritonitis was not uncommon in patients who had undergone laparotomy, in the days before the introduction of ‘talc-free’ surgical gloves. The main component of surgical talc is starch and starch peritonitis was first described in 1956 and is due to starch sensitivity. Some 2 to 6 weeks after an initial laparotomy from which the patient usually makes an uneventful recovery, the patient would represent with either ascites or signs and symptoms of bowel obstruction. At laparotomy, multiple nodules are seen over the parietal and visceral peritoneum, mimicking malignant deposits with dense adhesions between loops of bowel. Treatment consists of lysis of adhesions and the diagnosis is confirmed by histology where birefringent granules are observed. Administration of steroids is ineffective in preventing intestinal adhesions.

Other forms of chronic peritonitis which may form dense adhesions include the adhesive form of tuberculous peritonitis (see under acute peritonitis) and carcinomatosis.

Brown P, Baddeley H, Read AE, Davies JD, McGarry J. Sclerosing peritonitis, an unusual reaction to a &bgr;-adrenergic-blocking drug. Lancet, 1974; ii: 1477–81.

Ellis H. The hazards of surgical glove dusting powders. Surg Gynecol Obstet, 1990; 171: 521–7.

Kittur DS, Korpe SW, Raytch RE, Smith GW. Surgical aspects of sclerosing encapsulating peritonitis. Arch Surg, 1990; 125: 1626–8.

Maddaus MA, Ahrenholz D, Simmons RL. The biology of peritonitis and implications for treatment. Surg Clin N Am, 1988; 68: 431–43.

Saklayen MG. CAPD peritonitis. Incidence, pathogens, diagnosis and management. Med Clin N Am, 1990; 74: 997–1010.

Willcox CM, Dismukes WE. Spontaneous bacterial peritonitis. A review of pathogenesis, diagnosis and treatment. Medicine (Baltimore) 1987; 66: 447–56.