The Dept of Anaesthesia & Intensive Care, CUHK thanks

for an unrestricted education grant
BASIC instructor/provider course, Hong Kong, July 2nd-4th
Other upcoming courses
Home Feedback Contents

Spont bacterial peritonitis

Up Cholangitis Cholecystitis Spont bacterial peritonitis

Updated February 2007 by Charles Gomersall

Ascitic fluid infections


  • Spontaneous bacterial peritonitis (SBP). Most common infection in patients with ascites and cirrihosis. Ascitic fluid infection with positive bacterial culture, neutrophil count 250/ml and no surgically treatable source of infection. Bacterial cultures almost invariably yield a single growth. The presence of >1 organism suggests secondary peritionitis.
  • Culture negative neutrocytic ascites. Diagnostic criteria as for SBP but cultures are negative. Other causes of raised ascitic fluid neutrophil count need to be excluded (eg peritoneal carcinomatosis, pancreatitis, TB peritonitis). Clinical, prognostic and therapeutic characteristics are similar to those of SBP and this condition should be treated in a similar fashion to SBP.
  • Mononmicrobial non-neutrocytic bacterascites. Positive culture but neutrophils <250/ml. Clinical course is dependent on the presence/absence of clinical features. Those with clinical features of ascitic fluid infection have similar morbidity/mortality to those with SBP.
  • Secondary bacterial peritonitis. Positive cultures (usually polymicrobial), neutrophils  ≥250/ml, and surgically treatable source of infection. Clinical features do not distinguish SBP from secondary bacterial peritonitis. Ascites usually meets ≥2 of the following: total protein >1g/dL, glucose <50 mg/dL, LDH >225 U/ml (or higher than the upper limit of normal for serum)
  • Polymicrobial bacterascites. Gram stain or culture reveals multiple organisms but neutrophils <250/ml. Usually due to inadvertent puncture of intestine during paracentesis (risk» 1/1000). If ascitic fluid protein >1g/dL colonization usually resolves spontaneously
  • First three occur most commonly in patients with cirrhosis and ascites although they may occur in patients with other causes of ascites

Aetiology & Pathogenesis

- seemingly innocuous procedures can lead to transient bacteraemia and SBP in cirrhotics (eg NG tube placement, endoscopy, IV catheter insertion, bladder catheterization)
- usually develops when the volume of ascites is sizeable but can occur even when fluid is difficult to detect on physical examination
- E.coli, Strep (usually pneumococcus) and Klebsiella account for 3/4 of cases. Gram negative infection more common than gram positive. Anaerobes rarely isolated
- current view is that SBP is the result of bacterial overgrowth in the small intestine followed by bacterial translocation across the intestinal wall. Because of impaired immune response in cirrhotics and portosystemic shunts the bacteria then spread into the systemic circulation and seed in ascites.

Predisposing factors

  • Severity of liver disease:
    70% of cases occur in patients with Child’s grade C with most of the rest occuring in patients with Child’s grade B.
    Total ascites protein <1 g/dL predisposes to SBP because it correlates with complement levels and opsonic activity
  • GI bleeding
    20% of patients with cirrhosis and ascites presenting with a GI bleed have SBP
  • Bacteriuria
  • Previous SBP
    Recurrence rates: 43% by 6 months, 69% by 1 year and 74% by 2 years

Clinical features

  • signs may be absent in up to 1/3
  • fever/hypothermia
  • abdominal pain and tenderness
  • rigidity is not a feature in patients with infected ascites, even if there is a free perforation of the intestine. This is a result of the large volume of ascites preventing contact between the visceral and parietal peritoneal surfaces
  • hepatic encephalopathy
  • diarrhoea
  • ileus
  • shock
  • important to have high index of suspicion and low threshold for diagnostic paracentesis. Unexplained deterioration in a patient with cirrhosis and ascites should lead to diagnostic paracentesis


  • diagnostic tap of ascites. Innoculate some fluid directly into blood culture bottle as well as sending fluid for gram stain (often not helpful) and cell and differential count. Suspect SBP if neutrophil count > 250/ml. Other tests which may be helpful include total protein, glucose (­ in malignancy or gut perforation), LDH in spontaneous bacterial peritonitis) , amylase (­ in pancreatitis). Risk of paracentesis is small despite almost invariable impairment of clotting in these patients. Approx. 1% risk of significant abdominal wall haematoma, 0.01% risk of haemoperitoneum and 0.01% risk of iatrogenic infection
  • follow up paracentesis probably not necessary if response to treatment is dramatic, setting is typical and infection is monomicrobial. However if there are features to suggest secondary peritonitis paracentesis should be repeated (usually after 48 h)
  • peritonitis secondary to abscess or perforated viscus should be suspected if WCC >10,000/ml, ascitic protein is elevated, cultures of fluid grow anaerobes or multiple organisms, or follow-up paracentesis after 48 h of treatment reveals persistently positive cultures or a rising WCC


Spontaneous bacterial peritonitis, culture negative neutrocytic ascites, symptomatic monomicrobial bacterascites

3rd generation cephalosporin (cefotaxime 2g 8 hrly) until sensitivity known. 5 day course satisfactory even for patients who are bacteraemic.

Asymptomatic monomicrobial bacterascites

  • repeat paracentesis after 48 h.
  • treat with cefotaxime if symptoms develop or neutrophil count in repeat sample 250/ml


Consider for:

  • patients with cirrhosis and ascites presenting with a GI bleed. (Norfloxacin 400mg PO/NG bd for 7 days)
  • patients with ascitic fluid protein <1g/dL (Norfloxacin 400 mg od PO/NG during hospitalization)
  • patients with a previous episode of SBP (Norfloxacin 400 mg od indefinitely)


  • Good if detected and treated before the onset of renal failure or shock
  • Long term prognosis is poor due to the association with severe liver impairment. Approximately 30% 1 year and 20% 2 year survival

Further reading

Such, J. and Runyon, B.A. Spontaneous bacterial peritonitis. Clinical Infectious Diseases 27:669-676, 1998.

MeiLan King Han, Robert Hyzy. Advances in critical care management of hepatic failure and insufficiency. Crit Care Med, 34(9 Suppl):S225-S231

© Charles Gomersall December 1999, February 2007

©Charles Gomersall, April, 2014 unless otherwise stated. The author, editor and The Chinese University of Hong Kong take no responsibility for any adverse event resulting from the use of this webpage.
Copyright policy    Contributors