Spontaneous Bacterial Peritonitis

Etiology

SBP is most commonly (75%) caused by gram-negative aerobic organisms, with Klebsiella pneumoniae accounting for 50% of these. Gram-positive aerobic microorganisms are responsible for the remainder of cases; the most common of these are Streptococcus pneumoniae or Viridans group streptococci.[3][4]

The ascitic fluid typically has a high oxygen tension. Therefore anaerobic organisms are not commonly seen. In the majority of cases of SBP, only one infecting organism is involved (92%), though a small number of cases have been reported as polymicrobial.

Typically, patients who experience SBP have chronic liver disease with a Child-Pugh classification, which assesses the prognosis of liver disease, of C. This ranking involves a high to a maximum score of 10 to 15 points (on the Child-Pugh scale) and measures 1-year patient survival at 45% and 2-year survival at 35%. Decompensated cirrhotic patients are those at the highest risk of developing SBP. Infecting organisms typically originate from the intestinal lumen, from where they pass via translocation to mesenteric lymph nodes.

Additional risk factors for SBP include a previous history of SBP, low complement levels, and reduced hepatic synthesis of proteins with (1) an associated prolonged PT time and reduced protein levels in ascitic fluid (less than 1 g/dL), and (2) long-term proton pump inhibitor (PPI) therapy such as increased gastric pH with PPI use, which promotes gut bacterial growth and translocation.

Though in adults, SBP is typically seen in patients with abdominal ascites, most children with SBP do not have ascites. The reason for this has not yet been elucidated.[5]

Epidemiology

SBP can occur in adults and children. In children, it most commonly occurs in neonates and those around five years of age. It is most common in patients with cirrhosis. However, it can occur as a complication of any disease that results in the accumulation of ascitic fluid, such as liver disease, Budd-Chiari syndrome, congestive heart failure, systemic lupus erythematosus, renal failure, or cancers, and has a poor prognosis. Approximately 10 to 25% of patients with ascites will develop SBP, and the condition is associated with a 20% in-hospital rate of mortality.

Patients with a prior incidence of SBP are more likely to encounter a subsequent infection with a drug-resistant organism. Additionally, the risk of developing SBP increases with age, the use of proton-pump inhibitors (PPIs), and when undergoing SBP prophylaxis, such as selective intestinal decontamination.[6]

Pathophysiology

The majority of the isolated organisms in SBP (90%) are gram-negative enteric organisms (e.g., Escherichia coli or Klebsiella pneumoniae), which suggests that the primary source of contamination is the gastrointestinal (GI) tract. Enterotoxin is also frequently isolated from ascitic fluid, further supporting the theory that bacteria involved in SBP migrate transmurally from the intestinal lumen (i.e., bacterial translocation).

An alternatively proposed mechanism of contamination involves hematogenous spread, from a distant source, such as a urinary tract infection, in individuals predisposed to the disease by a weakened immune system (i.e., immunocompromised).  Patients with cirrhosis typically have an elevated level of bacterial overgrowth in the GI tract, mainly due to a prolonged intestinal transit time. This, coupled with reduced protein production by a cirrhotic liver (e.g., low complement levels in both the serum and ascites) and poor phagocytic and reticuloendothelial system function, results in a decreased ability to clear microorganisms from the system, thereby further contributing to bacterial overgrowth, migration, and expansion within the ascites fluid.[7]

History and Physical

One should have a high index of suspicion for SBP in all patients presenting with ascites, and this is especially true if the patient has an acute history of clinical deterioration. The majority of patients with SBP will present with fever, chills, and abdominal pain, although some patients may be asymptomatic, and SBP is an incidental finding. Temperature is the most common symptom encountered in patients with SBP, which is a particularly useful clinical symptom as patients with cirrhosis are typically hypothermic. Additional signs and symptoms include diarrhea, paralytic ileus, new-onset or worsening encephalopathy (e.g., altered mental status) without any other identifiable cause, new-onset or worsening renal failure, or presence of ascites that does not improve with the use of diuretic medications.

On physical examination, most patients will have a tender abdomen, although patient response can vary from mild discomfort to the presence of guarding and rebound tenderness. The absence of fever does not rule out SBP.

Evaluation

There is a short window of opportunity for treating SBP before it progresses to septic shock or multisystem organ failure; therefore, rapid assessment and diagnosis are critical (i.e., treatment is much more successful if antibiotics are started before the development of shock).[2][8][9]

Peritoneal fluid analysis (e.g., cell count, differential, culture, lactate level, pH) should be performed in all patients with suspected SBP. This may be achieved either by diagnostic paracentesis or withdrawal of fluid through a peritoneal catheter, which is sometimes present in patients undergoing peritoneal dialysis. In patients with only a small amount of ascites, ultrasonography should help guide the paracentesis procedure.

Additionally, blood and urine cultures should be obtained before initiation of antibiotic therapy as the results of these may help point towards a source of infection and guide antibiotic therapy.

The most accurate predictor of SBP is a polymorphonuclear leukocyte (PMN or granulocytes such as neutrophils, basophils, and eosinophils) count of greater than 500 cells/uL in a sample of ascitic fluid with a sensitivity and specificity of 86% and 98%, respectively. At a PMN count of higher than 250 cells/microliter, sensitivity increases to 93%, while specificity declines to 94%. This is the widely accepted number of PMNs needed to form a possible diagnosis of SBP before beginning empiric antibiotic therapy.

If the perforation is suspected within the abdomen, imaging is warranted, and computed tomography (CT) should be strongly considered, as it is more sensitive for detecting smaller perforations than a plain radiograph.

More recently, proposed tests for SBP include a rapid reagent strip that evaluates for the presence of leukocyte esterase in ascitic fluid, which is shown to have a sensitivity of 100% in the diagnosis of SBP when compared to manual PMN counting. Though this may prove to be a much more efficient method of SBP diagnosis than those previously used, the test still needs to undergo a large-scale comprehensive evaluation. 

Patients who are at high risk for SBP include: 

• Patients with gastrointestinal bleed and cirrhosis.
• Patients who already had SBP one or more times in the past.
• Cirrhotic patients with ascites in which ascitic fluid protein is < 1.5 g/dl along with renal failure (creatinine > 1.2 mg/dl).
• Cirrhotic patients hospitalized for a cause other than SBP have an ascitic protein concentration of < 1 g/dl (10 g/L).

Treatment / Management

Empiric antibiotic therapy, such as intravenous third-generation cephalosporin, should be started in all patients with suspected SBP and a PMN count of greater than 250 cells/microliter on ascitic fluid analysis. Exceptions to this rule include patients with recent beta-lactam antibiotic exposure or diagnosis of SBP in a nosocomial setting. Antibiotics should be chosen based on the results of susceptibility testing in these cases.[10][11][12]

Patients with an ascitic fluid PMN count of greater than 500cells/uL should be admitted and treated with empiric antibiotic therapy as soon as possible, and the antibiotics adjusted later based on the results of susceptibility testing. A follow-up analysis of ascitic fluid to assess PMN count should be performed to ensure the patient receives adequate and appropriate antibiotic therapy. This is important, as lack of improvement after 48 hours of therapy may indicate an underlying perforation or abscess formation, such as secondary bacterial peritonitis, that may require surgery. In general, ascitic fluid PMN count should be reduced by at least 25% after 48 hours of antibiotic therapy.

Some cirrhotic patients with SBP and either a serum creatinine higher than 1 mg/dL, a blood urea nitrogen (BUN) higher than 30mg/dL, or total bilirubin greater than 4 mg/dL should be given adjunctive (i.e., in addition to antibiotics) albumin intravenously. This has been shown to reduce both in-hospital mortality and renal damage compared to the use of antibiotic therapy alone. 

Certain high-risk patients may benefit from outpatient antibiotic prophylactic therapy, including norfloxacin, ciprofloxacin, or trimethoprim-sulfamethoxazole. These individuals have had a prior episode of SBP, or ascitic fluid with either a low protein count (< 1 g/dL) or is associated with a GI bleed.

A common related concern to antibiotic prophylaxis is antibiotic resistance, especially in those centers which particularly use fluoroquinolone as SBP prophylaxis. In these patients, cefotaxime is the treatment of choice. Piperacillin-tazobactam or a carbapenem can also be used as an alternative empirically if cefotaxime can not be utilized.[13](A1)

Renal failure is the major cause of death in patients with SBP and develops in 30 to 40 percent of the patients. The risk can be minimized by giving IV albumin. IV albumin should be given when the creatinine is > 1 mg/dl, the blood urea nitrogen is > 30 mg/dl, or the total bilirubin is > 4 mg/dl. Treatment with octreotide or midodrine is helpful if renal failure develops.[14]

Differential Diagnosis

• Perforated viscus
• Perinephric abscess
• Pyelonephritis
• Diverticulitis
• Appendicitis
• Mesenteric ischemia

Prognosis

The infection-related mortality in SBP is very low with appropriate treatment. However, the mortality rate is quite high in those patients who develop sepsis. But with appropriate antibiotic therapy, better outcomes can be achieved. In hospital settings, the non-infection-related mortality in SBP patients can be as high as 20-40 percent, and one to two years of mortality rates are 70 and 80 percent, respectively.[15] Regardless of the short-term outcome of SBP, patients who have liver disease severe enough to develop SBP usually have a terrible long-term prognosis. Moreover, liver transplantation should be considered for the survivors of SBP, who are otherwise good transplantation candidates.

Complications

• Renal failure 
• Sepsis
• Liver failure/insufficiency
• Tense ascites
• Bleeding after paracentesis 
• Bowel perforation after paracentesis 
• Spontaneous fungal peritonitis