Spontaneous bacterial peritonitis (SBP) is a well-recognized, severe complication in cirrhotic patients presenting to the emergency department (ED) with ascites. The prevalence of this disease varies widely among study populations, with a range of 10% to 30% in hospitalized patients with ascites, compared to 0% to 3.5% among asymptomatic outpatient clinic populations.
Spontaneous bacterial peritonitis (SBP) is a well-recognized, severe complication in cirrhotic patients presenting to the emergency department (ED) with ascites. The prevalence of this disease varies widely among study populations, with a range of 10% to 30% in hospitalized patients with ascites, compared to 0% to 3.5% among asymptomatic outpatient clinic populations. The most recent prospective ED-based study found an SBP prevalence of 12% in all patients undergoing paracentesis. Mortality rates from treated SBP range from 20% to 30% in-hospital to nearly 50% at 1 year.
This significant condition is defined as an infection of ascitic fluid with no apparent intra-abdominal source (i.e., abscess, perforated viscus). Most sources postulate that this infection arises from translocation of bacteria from the intestines to the systemic circulation in the context of an impaired immune system, secondary to hepatic disease. The diagnostic evaluation for SBP centers on the analysis of ascitic fluid. Although this analysis remains somewhat controversial, the diagnosis of SBP and its variants is generally agreed upon:
■ SBP: polymorphonuclear (PMN) count in ascites >250 cells/mm3 with the growth of pathogenic bacteria from ascites fluid (however high false-negative rate of the culture).
■ Culture-negative neutrocytic ascites (CNNA): the PMN count is >250 cells/mm3, but there is no culture growth (should be treated as SBP).
■ Bacterascites: positive culture result, but with a PMN count <250 cells/mm3 (most sources regard this as transient intestinal colonization or skin flora contamination).
While the definitions are important to know, it is important to realize that you could be missing the diagnosis in patients with SBP who “look good.” Most patients will NOT display classic signs and symptoms of SBP. Likewise, there is no single clinical characteristic that we can rely upon to effectively rule in or rule out SBP. In fact, clinical impression has a sensitivity of only 76% and specificity of 34% for the detection of SBP. Fever was present in only 50% to 68% of patients, while abdominal pain was noted in only 49% to 60% of patients. Confusion or encephalopathy was seen in 50% to 60% of patients. Rebound tenderness was present in only 10% to 42% of patients.
Because there is no reliable clinical finding that can rule in or rule out the condition, the ED physician must maintain a HIGH index of suspicion for SBP in all patients with ascites. Certainly, during a busy shift, you would prefer to call the medicine team and let them worry about tapping the belly, but you MUST PERFORM a paracentesis in any patient with new-onset ascites or with ascites and associated fever, abdominal pain, peritoneal signs, leukocytosis, unexplained encephalopathy, or worsening liver and renal function.
Once you have come to the realization that you must consider SBP and that the only way to truly investigate the condition is by performing a paracentesis, it is incumbent upon you to make sure the results are as valid as possible.
The standard for diagnosis of SBP is a PMN count of >250 cells/mm3 in ascitic fluid, with or without a subsequent positive culture. Interestingly, ascitic cultures yield a pathogenic bacterium in only 40% to 50% of cases. In an effort to maximize your yield and improve patient care, the following are evidence-based considerations to help you clinch an early and accurate SBP diagnosis:
■ Use sterile approach, with patient positioned at 30 to 45 degrees. Use a 20- or 22-gauge, 2.5-in. needle.
■ Use ultrasound guidance. Left lower quadrant (LLQ) is preferred to the infraumbilical approach (less bleeding and greater depth of ascites).
■ Note the clarity of fluid (sensitivity of 98.1%, specificity 22.7% for SBP if hazy, cloudy) and send for cell count.
■ Inoculate 10 mL of ascitic fluid directly into aerobic and anaerobic blood culture bottles to increase the yield of cultures to 50% to 80%.
■ Use a urine dipstick test on the ascites at bedside, as this can provide rapid diagnosis if leukocyte esterase positive (varies by manufacturer, but generally >stage 2 on the strip). A sensitivity of 89% to 100% with a specificity of 98% to 100%.
Certainly, there are risks to the procedure that need to be discussed but do not talk yourself out of performing the paracentesis because of the possible complications. You should counsel patients and their families concerning the benefits of diagnostic paracentesis and the relatively low rate of risks. These are the facts that should encourage you and patients:
■ Bleeding risk: 0.2% to 1% even in mild-moderate coagulopathy and thrombocytopenia; risk is increased with renal failure
■ Perforation of bowel or bladder: <1% (empty bladder, use ultrasound)
■ Infection: theoretical risk, no statistics available
■ Persistent leak at puncture site: <1% (using Z-Track technique)
■ Death: <1.2% After successfully performing a safe and efficacious procedure, you should initiate broad-spectrum antibiotics immediately in all patients with an ascitic PMN count >250 cells/mm3. Intravenous third-generation cephalosporins are considered first-line treatment at present by some authors, with reported 75% to 90% treatment success; however, recent data show an increasing incidence of SBP caused by Gram-positive and drug-resistant bacteria. In a small prospective validation study, there was treatment failure in 41% of cases treated with cefotaxime. The microbial etiology of SBP is changing, and you should anticipate this and initiate the therapy with broad-spectrum antibiotics. Choosing imipenem or piperacillin-tazobactam, for any patient with an ascitic PMN count >250 cells/mm3, will not only prove that you are well read on the subject but also improve your patients’ care.
This article is an excerpt from the book “Avoiding Common Errors in the Emergency Department” (Lippincott Williams & Wilkins, 2010)