Spontaneous Bacterial Peritonitis

• Published in: Hepatology (Baltimore, Md.) Md.), 1982-07, Vol.2 (4), p.399S-407S
• Main Authors: Hoefs, John C., Canawati, Hanna N., Sapico, Francisco L., Hopkins, R. Randy, Weiner, John, Montgomerie, John Z.
• Format: Article
• Language: English
• Subjects: Anti-Bacterial Agents - therapeutic use; Ascitic Fluid - cytology; Ascitic Fluid - microbiology; Autopsy; Bacterial Infections - blood; Bacterial Infections - diagnosis; Bacterial Infections - drug therapy; Bacterial Infections - microbiology; Bacterial Infections - mortality; Female; Humans; Leukocyte Count; Male; Middle Aged; Peritonitis - blood; Peritonitis - diagnosis; Peritonitis - drug therapy; Peritonitis - microbiology; Peritonitis - mortality; Prognosis
• ISSN: 0270-9139; 1527-3350
• DOI: 10.1002/hep.1840020402

Forty‐three patients with spontaneous bacterial peritonitis (SBP) between 1973 and 1978 were identified. Criteria for SBP included a positive ascites culture and polymorphonuclear cell concentration greater than 250 cells per mm3. Chronic liver disease was documented by varices in 91%, severe histologic fibrosis or cirrhosis in 94%, splenomegaly in 91%, and past hospitalization for liver disease in 57% of the patients. SBP was detected within 7 days of admission in 17 patients (40%) and within 35 days in 38 patients. Single organisms were isolated from 38 patients and multiple organisms from 5 patients. Twenty‐six of 43 patients survived the episode of SBP, but only 13 survived the hospitalization. Analysis of the survival curve from the onset of SBP revealed a rapid death rate and a slow death rate set of patients. Rapid death (< 7 days from SBP onset) correlated with a lack of prior hospitalization for liver disease (p < 0.001), hepatomegaly (p < 0.001), increased serum bilirubin (p < 0.005), serum creatinine (p < 0.05), and peripheral white blood cell concentrations (p < 0.05). Survival during hospitalization was associated with prior hospitalization with liver disease (p < 0.001) and chills during the episode of SBP (p < 0.001). The 43 patients were divided into Group 1 patients on the basis of a serum bilirubin > 8 mg% and/or serum creatinine > 2.1 mg%; Group 2 patients had lower values. Survival was greater in Group 2 patients with advanced, relatively quiescent liver disease compared to Group 1 patients for both the episode of SBP (91 vs. 29%; p < 0.001) and for hospitalization (50 vs. 9%; p < 0.05). Death in Group 2 patients was related to inadequate antibiotic therapy (p < 0.05), nonhepatic factors, and new onset of renal failure. Although SBP in the setting of severe acute liver injury has a dismal prognosis, SBP with minimal acute liver injury has a relatively good prognosis for hospital survival even with advanced chronic liver disease. Long‐term survival is also possible since 4 of 9 patients with prolonged follow‐up have survived 3 years.