A 54-year-old man with end-stage liver disease (ESLD) and no prior history of spontaneous bacterial peritonitis (SBP) presents with increasing shortness of breath and abdominal distention. He is admitted for worsening volume overload. The patient reveals that he has not been compliant with his diuretics. On the day of admission, a large-volume paracentesis is performed. Results are significant for a white blood cell count of 150 cells/mm3 and a total protein of 0.9 g/ul. The patient is started on furosemide and spironolactone, and his symptoms significantly improve throughout his hospitalization. His medications are reconciled on the day of discharge. He is not on any antibiotics for SBP prophylaxis; should he be? In general, which patients with ascites should receive SBP prophylaxis?
Spontaneous bacterial peritonitis is an infection of ascitic fluid that occurs in the absence of an indentified intra-abdominal source of infection or inflammation, i.e., perforation or abscess.1 It is diagnosed when the polymorphonuclear cell (PMN) count in the ascitic fluid is equal to or greater than 250 cells/mm3, with or without positive cultures.
SBP is a significant cause of morbidity and mortality in patients with cirrhosis, with the mortality rate approaching 20% to 40%.2 Of the 32% to 34% of cirrhotic patients who present with, or develop, a bacterial infection during their hospitalization, 25% are due to SBP.1 Changes in gut motility, mucosal defense, and microflora allow for translocation of bacteria into enteric lymph nodes and the bloodstream, resulting in seeding of the peritoneal fluid and SBP.1 Alterations in both systemic and localized immune defenses, both of which are reduced in patients with liver disease, also play a role in SBP pathogenesis (see Table 1, p. 41).
Current evidence supports the use of a third-generation cephalosporin or amoxicillin/clavulanate for initial treatment of SBP, as most infections are caused by gram-negative bacilli, in particular E. coli (see Table 2 on p. 41 and Table 3 on p. 42).1 Alternatively, an oral or intravenous fluoroquinolone could be used if the prevalence of fluoroquinolone-resistant organisms is low.1
Due to the frequency and morbidity associated with SBP, there is great interest in preventing it. However, the use of prophylactic antibiotics needs to be restricted to patients who are at highest risk of developing SBP. According to numerous studies, patients at high risk for SBP include:
- Patients with a prior SBP history;
- Patients admitted with a gastrointestinal bleed; and
- Patients with low total protein content in their ascitic fluid (defined as <1.5 g/ul).1
Spontaneous bacterial peritonitis portends bad outcomes. The one-year mortality rate after an episode of SBP is 30% to 50%.1 Furthermore, patients who have recovered from a previous episode of SBP have a 70% chance of developing another episode of SBP in that year.1,2 In one study, norfloxacin was shown to decrease the one-year risk of SBP to 20% from 68% in patients with a history of SBP.3 Additionally, the likelihood of developing SBP from gram-negative bacilli was reduced to 3% from 60%. In order to be efficacious, norfloxacin must be given daily. When fluoroquinolones are prescribed less than once daily, there is a higher rate of fluoroquinolone resistant organisms in the stool.1
Though once-daily dosing of norfloxacin is recommended to decrease the promotion of resistant organisms in prophylaxis against SBP, ciprofloxacin once weekly is acceptable. In a group of patients with low ascitic protein content, with or without a history of SBP, weekly ciprofloxacin has been shown to decrease SBP incidence to 4% from 22% at six months.4 In regard to length of treatment, recommendations are to continue prophylactic antibiotics until resolution of ascites, the patient receives a transplant, or the patient passes away.1
Saab et al studied the impact of oral antibiotic prophylaxis in patients with advanced liver disease on morbidity and mortality.5 The authors examined prospective, randomized, controlled trials that compared high-risk cirrhotic patients receiving oral antibiotic prophylaxis for SBP with groups receiving placebo or no intervention. Eight studies totaling 647 patients were included in the analysis.
The overall mortality rate for patients treated with SBP prophylaxis was 16%, compared with 25% for the control group. Groups treated with prophylactic antibiotics also had a lower incidence of all infections (6.2% vs. 22.2% in the control groups). Additionally, a survival benefit was seen at three months in the group that received prophylactic antibiotics.
The absolute risk reduction with prophylactic antibiotics for primary prevention of SBP was 8% with a number needed to treat of 13. The incidence of gastrointestinal (GI) bleeding, renal failure, and hepatic failure did not significantly differ between treatment and control groups. Thus, survival benefit is thought to be related to the reduced incidence of infections in the group receiving prophylactic antibiotics.5
History of GI Bleeding
The incidence of developing SBP in cirrhotics with an active GI bleed is anywhere from 20% to 45%.1,2 For those with ascites of any etiology and a GI bleed, the incidence can be as high as 60%.5 In general, bacterial infections are frequently diagnosed in patients with cirrhosis and GI bleeding, and have been documented in 22% of these patients within the first 48 hours after admission. According to several studies, that percentage can reach as high as 35% to 66% within seven to 14 days of admission.6 A seven-day course of antibiotics, or antibiotics until discharge, is generally acceptable for SBP prophylaxis in the setting of ascites and GI bleeding (see Table 2, right).1
Bernard et al performed a meta-analysis of five trials to assess the efficacy of antibiotic prophylaxis in the prevention of infections and effect on survival in patients with cirrhosis and GI bleeding. Out of 534 patients, 264 were treated with antibiotics between four and 10 days, and 270 did not receive any antibiotics.
The endpoints of the study were infection, bacteremia and/or SBP; incidence of SBP; and death. Antibiotic prophylaxis not only increased the mean survival rate by 9.1%, but also increased the mean percentage of patients free of infection (32% improvement); bacteremia and/or SBP (19% improvement); and SBP (7% improvement).7
Low Ascitic Fluid Protein
Of the three major risk factors for SBP, ascitic fluid protein content is the most debated. Guarner et al studied the risk of first community-acquired SBP in cirrhotics with low ascitic fluid protein.2 Patients were seen immediately after discharge from the hospital and at two- to three-month intervals. Of the 109 hospitalized patients, 23 (21%) developed SBP, nine of which developed SBP during their hospitalization. The one-year cumulative probability of SBP in these patients with low ascitic fluid protein levels was 35%.
During this study, the authors also looked at 20 different patient variables on admission and found that two parameters—high bilirubin (>3.2mg/dL) and low platelet count (<98,000 cells/ul)—were associated with an increased risk of SBP. This is consistent with studies showing that patients with higher Model for End-Stage Liver Disease (MELD) or Child-Pugh scores, indicating more severe liver disease, are at increased risk for SBP. This likely is the reason SBP prophylaxis is recommended for patients with an elevated bilirubin, and higher Child-Pugh scores, by the American Association for the Study of Liver Disease (see Table 2, p. 41).
Runyon et al showed that 15% of patients with low ascitic fluid protein developed SBP during their hospitalization, as compared with 2% of patients with ascitic fluid levels greater than 1 g/dl.8 A randomized, non-placebo-controlled trial by Navasa et al evaluating 70 cirrhotic patients with low ascitic ascitic protein levels showed a lower probability of developing SBP in the group placed on SBP prophylaxis with norfloxacin (5% vs. 31%).9 Six-month mortality rate was also lower (19% vs. 36%).
In contrast to the previous studies, Grothe et al found that the presence of SBP was not related to ascitic protein content.10 Given conflicting studies, controversy still remains on whether patients with low ascitic protein should receive long-term prophylactic antibiotics.
The consensus in the literature is that patients with ascites who are admitted with a GI bleed, or those with a history of SBP, should be placed on SBP prophylaxis. However, patients placed on long-term antibiotics are at risk for developing bacterial resistance. Bacterial resistance in cultures taken from cirrhotic patients with SBP has increased over the last decade, particularly in gram-negative bacteria.5 Patients who receive antibiotics in the pre-transplant setting also are at risk for post-transplant fungal infections.
Additionally, the antibiotic of choice for SBP prophylaxis is typically a fluoroquinolone, which can be expensive. However, numerous studies have shown that the cost of initiating prophylactic therapy for SBP in patients with a prior episode of SBP can be cheaper than treating SBP after diagnosis.2
Back to the Case
Our patient’s paracentesis was negative for SBP. Additionally, he does not have a history of SBP, nor does he have an active GI bleed. His only possible indication for SBP prophylaxis is low ascitic protein concentration. His electrolytes were all within normal limits. Additionally, total bilirubin was only slightly elevated at 2.3 mg/dL.
Based on the American Association for the Study of Liver Diseases guidelines, the patient was not started on SBP prophylaxis. Additionally, given his history of medication noncompliance, there is concern that he might not take the antibiotics as prescribed, thus leading to the development of bacterial resistance and more serious infections in the future.
Patients with ascites and a prior episode of SBP, and those admitted to the hospital for GI bleeding, should be placed on SBP prophylaxis. SBP prophylaxis for low protein ascitic fluid remains controversial but is recommended by the American Association for the Study of Liver Diseases. TH
Dr. del Pino Jones is a hospitalist at the University of Colorado Denver.
- Ghassemi S, Garcia-Tsao G. Prevention and treatment of infections in patients with cirrhosis. Best Pract Res Clin Gastroenterol. 2007;21(1):77-93.
- Guarner C, Solà R, Soriono G, et al. Risk of a first community-acquired spontaneous bacterial peritonitis in cirrhotics with low ascitic fluid protein levels. Gastroenterology. 1999;117(2):414-419.
- Ginés P, Rimola A, Planas R, et al. Norfloxacin prevents spontaneous bacterial peritonitis recurrence in cirrhosis: results of a double-blind, placebo-controlled trial. Hepatology. 1990;12(4 Pt 1):716-724.
- Rolachon A, Cordier L, Bacq Y, et al. Ciprofloxacin and long-term prevention of spontaneous bacterial peritonitis: results of a prospective controlled trial. Hepatology. 1995;22(4 Pt 1):1171-1174.
- Saab S, Hernandez J, Chi AC, Tong MJ. Oral antibiotic prophylaxis reduces spontaneous bacterial peritonitis occurrence and improves short-term survival in cirrhosis: a meta-analysis. Am J Gastroenterol. 2009;104(4):993-1001.
- Deschênes M, Villeneuve J. Risk factors for the development of bacterial infections in hospitalized patients with cirrhosis. Am J Gastroenterol. 1999;94(8):2193-2197.
- Bernard B, Grangé J, Khac EN, Amiot X, Opolon P, Poynard T. Antibiotic prophylaxis for the prevention of bacterial infections in cirrhotic patients with gastrointestinal bleeding: a meta-analysis. Hepatology. 1999;29(6):1655-1661.
- Runyon B. Low-protein-concentration ascitic fluid is predisposed to spontaneous bacterial peritonitis. Gastroenterology. 1986;91(6):1343-1346.
- Navasa M, Fernandez J, Montoliu S, et al. Randomized, double-blind, placebo-controlled trial evaluating norfloxacin in the primary prophylaxis of spontaneous bacterial peritonitis in cirrhotics with renal impairment, hyponatremia or severe liver failure. J Hepatol. 2006;44(Supp2):S51.
- Grothe W, Lottere E, Fleig W. Factors predictive for spontaneous bacterial peritonitis (SBP) under routine inpatient conditions in patients with cirrhosis: a prospective multicenter trial. J Hepatol. 1990;34(4):547.