1. 
   

   How is spontaneous bacterial peritonitis diagnosed?

   
   
2. 
   

   What is the optimal duration of treatment of spontaneous bacterial peritonitis?

   
   
3. 
   

   How is secondary peritonitis diagnosed and treated?

   
   
4. 
   

   When should tuberculous peritonitis be suspected?

   
   
5. 
   

   How are intra-abdominal abscesses diagnosed and treated?

Peritonitis and intra-abdominal abscesses are much feared because of the high frequency of associated septic shock and multisystem organ failure. Both primary and secondary peritonitis have an in-hospital mortality rate of approximately 20%. Intra-abdominal infections are the second leading cause of infectious death in the intensive care unit. Although the precise incidence of peritonitis is unknown, it is encountered with some regularity by physicians caring for inpatients. Spontaneous bacterial peritonitis (SBP), or primary peritonitis, occurs in up to 30% of all patients with liver cirrhosis and ascites. Secondary peritonitis and intra-abdominal abscesses are important complications of conditions commonly seen on general medical wards, such as diverticulitis, peptic ulcer disease, pancreatitis, and cholecystitis. Classic symptoms and signs of intra-abdominal infection may be blunted in older or immunosuppressed patients, increasing the challenge for clinicians.

Peritonitis is inflammation of the peritoneal surface, caused by microorganisms or irritants such as foreign bodies, bile, and barium. Peritonitis is classified as primary (or spontaneous), secondary, or tertiary. In primary peritonitis, there is inflammation of the peritoneal surface without another intra-abdominal process. Secondary peritonitis develops as a result of inflammation of another structure within the abdomen. Tertiary peritonitis refers to persistent inflammation after treatment for secondary peritonitis.

Spontaneous (Primary) Peritonitis

Primary peritonitis, or spontaneous bacterial peritonitis (SBP), is most commonly seen in cirrhotic patients with ascites, and less often in patients with ascites from other causes, such as heart failure or systemic lupus erythematosus. Bacteria may gain access to the peritoneal fluid from hematogenous or lymphogenous spread, traversing the intact intestinal wall from the gut lumen or by passing through the fallopian tubes from the vagina in women.

Presentation

The classic signs and symptoms of peritonitis are fever, abdominal pain, and rebound tenderness. These may be mild or completely absent in cirrhotic patients. SBP should be suspected in all cirrhotic patients with clinical decompensation, such as the development of hepatic encephalopathy or hepatorenal syndrome.

Bacteriology

Most episodes of SBP are caused by enteric Gram-negative rods, such as E coli and Klebsiella species. An important minority of SBP episodes is caused by Gram-positive organisms, especially Streptococcus pneumoniae. Infections are rarely polymicrobial, and generally do not involve anaerobic bacteria. Table 187-1 shows the relative frequencies of common organisms isolated from ascitic fluid in patients with SBP.

Diagnosis

Fluid should be obtained by paracentesis and sent for Gram stain, culture, cell count with differential, and albumin. Paracentesis is generally safe, even in the setting of grossly abnormal coagulation tests and low platelets. Peritoneal fluid should be inoculated into blood culture bottles at the bedside, after changing the needle on the paracentesis syringe to a sterile needle. At least 10 cc of fluid should be inoculated into the bottle if possible. A separate syringe, tube, or container of fluid should be sent for immediate Gram stain. Fluid should also be injected into purple and red top tubes for cell count and chemistries, respectively.

Gram stain is positive in only a minority of cases. In SBP, the cell count is generally greater than 300 white blood cells/mm3, and is usually more than 1000 cells/mm3, with neutrophils predominating on differential. An absolute polymorphonuclear (PMN) leukocyte count of greater than or equal to 250 cells/mm3 in conjunction with a positive culture is diagnostic of peritonitis. A small percentage of patients, such as those with neutropenia, will have fewer cells with active infection. A traumatic procedure can result in entry of blood into the fluid being analyzed, artifactually raising red and white cell counts. To correct for extraneous blood, one can subtract one PMN for every 250 red blood cells present. In appropriately treated SBP, ascitic fluid cell count should start to decline within 48 to 72 hours. If a decline is not seen, an evaluation for secondary peritonitis should be undertaken, beginning with imaging of the abdomen. Repeat paracentesis is not necessary in a patient who is responding well clinically.

It can be difficult to distinguish secondary peritonitis from SBP. Some clues that the infection is secondary to an intra-abdominal process include failure to respond rapidly to antibiotic therapy, serum ascites-albumin gradient less than 1.1 g/dL, indicating the absence of portal hypertension, ascitic total protein greater than 1 g/dL, ascitic fluid glucose concentration less than 50 mg/dL, lactate dehydrogenase greater than the upper limit of normal for serum, and polymicrobial culture. If any of these findings are present, the patient should be imaged with flat and upright abdominal X-rays and an abdominal CT with oral and intravenous contrast.

Treatment

When SBP is suspected, treatment should not be delayed while awaiting the results of laboratory tests. Third-generation cephalosporins are preferred for initial therapy. It is not necessary to empirically cover Enterococcus or anaerobic bacteria. Historically, there has been reluctance to use ceftriaxone due to fear of worsening cholestasis in patients with liver failure, but this has not been supported by clinical experience. Due to increasing resistance, fluoroquinolones such as ciprofloxacin should not be used for empiric therapy. However, if culture results reveal a fluoroquinolone-susceptible organism, the high bioavailability of these drugs makes them a good option for oral therapy. Fluoroquinolones should never be used empirically in patients that have received fluoroquinolone prophylaxis for SBP. In patients with a good clinical response, five days of therapy is adequate. In patients with an incomplete response after five days (eg, persistent pain, fever, or altered mental status), paracentesis should be repeated. If the white blood cell count in ascitic fluid is less than 250 cells/mm3, treatment may be stopped. If the white blood cell count is in excess of 250 cells/mm3, but improved compared to the pretreatment value, treatment should continue, and repeat paracentesis should be considered in a few days. If the white blood cell count remains high, evaluation for secondary peritonitis is indicated. Bacteremia associated with SBP does not require a longer course of therapy, unless it is due to an organism with a robust association with endocarditis, such as Staphylococcus aureus or Enterococcus.

Prevention

Oral antibiotics are recommended as prophylaxis against SBP in patients with an ascitic fluid protein concentration less than 1.0 g/dL, any patient with a history of SBP, and patients with variceal bleeding. Ciprofloxacin, norfloxacin, and trimethoprim/sulfamethoxazole are both effective and cost-effective. Daily dosing is preferred to intermittent dosing, due to the lower likelihood for the development of antibiotic resistance.

Tuberculous peritonitis is often confused with other conditions. In alcoholic patients, it may mimic SBP, and it may be mistaken for ovarian carcinoma in women. It results from hematogenous dissemination of tuberculosis from remote foci of disease such as the lung, or from direct inoculation of bacilli into the peritoneum from intra-abdominal lymph nodes. The peritoneum becomes studded with tubercles that exude fluid, and ascites develops. Tuberculous peritonitis should be suspected in patients with the subacute onset of ascites, abdominal pain, and risk factors for TB exposure. Fever is variable. On paracentesis, there will generally be greater than 150 white cells/mm3, with lymphocytic predominance and elevated protein. If the patient is not cirrhotic, the serum ascites-albumin gradient will be less than 1.1 g/dL, indicating the absence of portal hypertension. Ascites fluid culture is rarely positive for Mycobacteria tuberculosis, and peritoneal biopsy is usually necessary to make the diagnosis. Biopsy under direct visualization, ideally via laparoscopy, has a much higher yield than blind percutaneous biopsy. Therapy for confirmed TB peritonitis is the same as that given for pulmonary TB.

Secondary peritonitis is classified as perforation peritonitis and nonperforation peritonitis. Any ruptured gastrointestinal or genitourinary organ may result in perforation peritonitis. Causes of perforation include trauma, infection, peptic ulcer, ischemia, neoplasm, and surgery (see Table 187-2).

Presentation

Abdominal pain is almost universal in secondary peritonitis. Nausea, vomiting, and anorexia are common. Fever and leukocytosis are frequently present. Alteration in bowel function may also be present. The specific character, location, and progression of the abdominal pain depend on the etiology of the peritonitis. Ruptured peptic ulcer pain has a sudden onset and is worse in the epigastric region, whereas appendicitis pain has a gradual onset and classically begins as a diffuse pain that localizes to the right lower quadrant, sometimes with periappendiceal abscess formation (Figure 187-1).