0
Introduction with a case
0
A patient with advanced alcoholic cirrhosis presents to the hospital with fever and altered mental status. Examination is notable for abdominal distention with rebound tenderness. Bedside ultrasound reveals a large amount of ascites, which is carefully sampled revealing a cloudy fluid with 15,000 neutrophils/uL and a differential of 90% neutrophils (a video of the left lower quadrant abdominal ultrasound is shown above). The diagnosis of spontaneous bacterial peritonitis is made, and the patient is treated. However, ten hours later the patient develops worsening hypotension and tachycardia requiring transfer to the ICU. What went wrong?
0
Secondary bacterial peritonitis
0
We’re all very familiar with spontaneous bacterial peritonitis: a patient with cirrhosis and ascites gets a tiny inoculum of bacteria into their peritoneum and a low-grade infection of this space. This diagnosis can be hard to make, because patients may not have much abdominal pain and usually don’t appear toxic.
0
Secondary bacterial peritonitis is more sinister and rare (twenty times less common than spontaneous bacterial peritonitis). This is infected ascites fluid due to an underlying surgically-treatable source of infection (i.e., perforated viscus, appendicitis, cholecystitis, mesenteric ischemia, etc.). It is critical to distinguish this from spontaneous bacterial peritonitis, because secondary bacterial peritonitis usually requires surgery.
0
The diagnosis of secondary bacterial peritontis is often delayed or missed due to low awareness of this condition. The possibility of secondary bacterial peritonitis should always be considered carefully prior to concluding that the patient has spontaneous bacterial peritonitis.
0
Red flags for secondary bacterial peritonitis
0
When evaluating a patient with ascites and probable spontaneous bacterial peritonitis, the following features may raise suspicion for an alternative diagnosis:
0
(1) Prominent peritoneal irritation: In our experience the abdominal examination in spontaneous bacterial peritonitis is often unimpressive, and an abdominal examination suggestive of frank peritonitis should raise suspicion of an alternative diagnosis. However, this is only weakly supported by the evidence: series report rebound tenderness in 17% of patients with secondary bacterial peritonitis versus 10% of patients with spontaneous bacterial peritonitis (Koulaouzidis 2007). Nonetheless, cirrhosis doesn’t protect against other intra-abdominal disorders. Therefore, it may be prudent to scan a patient with findings suggestive of frank peritonitis even if they have coexisting ascites.
0
(2) Free air on point-of-care bedside ultrasound: Experts achieve a sensitivity of 85-90% and specificity near 100%. In practice, performance varies depending on the operator, equipment, the amount of free air, and the precise set of findings visualized. This examination takes minutes and is may be incorporated into routine practice when performing ultrasound to evaluate ascites or abdominal pain (Hoffmann 2012).
0
(3) Gram stain with multiple organisms: Spontaneous bacterial peritonitis is a monomicrobial process. If multiple different morphologies of bacteria are seen on the gram stain this is nearly diagnostic of secondary bacterial peritonitis. When concerned about secondary bacterial peritonitis, a stat gram stain is occasionally very helpful.
0
(4) Treatment failure: Alternative diagnoses should always be entertained when a patient fails to respond to appropriate therapy.
0
Many sources suggest that an ascitic fluid leukocyte count above 10,000 neutrophils/uL is suggestive of secondary bacterial peritonitis. However Runyon 1990 found significant overlap between spontaneous and secondary bacterial peritonitis (8,500 +/- 12,000 vs. 10,000 +/- 13,000, respectively).
0
Diagnostic Algorithms
0
0
In 1990 Runyon published the above algorithm for differentiating spontaneous vs. secondary bacterial peritonitis. Textbooks and articles have faithfully reproduced versions of this algorithm ever since. The crux of this algorithm is Runyon’s Criteria for secondary bacterial peritonitis, which requires two of these three features: total protein >1 g/dL, glucose <50 mg/dL (2.8 mM), and lactate dehydrodgenase above the upper limit of normal for serum. These criteria were designed to detect patients with perforation. Therefore, in the original publication Runyon used a repeat paracentesis at 48 hours to detect non-perforated secondary bacterial peritonitis. This strategy has the obvious drawback that it requires a 48-hour diagnostic delay for any patient with secondary bacterial peritonitis without perforation.
0
In 2010 Soriano performed a retrospective study describing 24 patients diagnosed with secondary peritonitis compared to 106 patients with spontaneous bacterial peritonitis. The Runyon criteria had 67% sensitivity and 90% specificity among all patients with secondary bacterial peritonitis, and only 64% sensitivity among patients with perforation. The average delay from the paracentesis to diagnosis was 2.5 days. This study suggests that the Runyon criteria are insufficiently sensitive.
0
0
Our diagnostic approach is shown above. Although secondary bacterial peritonitis only constitutes ~5% of cirrhotic patients with ascites, this diagnosis has major implications for management. Adding red flags to Runyon’s criteria may improve sensitivity and thereby reduce the time delay to diagnosis. Overall there should be a low threshold to obtain imaging (CT scan and possibly an ultrasound if concerned regarding biliary pathology).
0
Significance of ultrasound showing septated ascitic fluid.
0
Sterile ascitic fluid from cirrhosis should have a simple, anechoic (black) appearance on bedside ultrasonography. The presence of septations and debris indicates significant inflammation with a differential diagnosis including malignancy, tuberculosis, and bacterial infection. Septations do not differentiate between spontaneous bacterial peritonitis versus secondary bacterial peritonitis.
0
Treatment
0
Treatment centers around broad-spectrum antibiotics and source control. Albumin has been shown to reduce renal dysfunction in spontaneous bacterial peritonitis, suggesting a potential utility in related infections. A recent randomized controlled trial showed improved renal function when albumin was administered to cirrhotic patients with infections other than spontaneous bacterial peritonitis (Guevara 2012).
0
Unfortunately this disease is hard to treat. Patients with advanced cirrhosis have a high mortality with intra-abdominal surgery. For patients with end-stage cirrhosis who are not transplant candidates, a frank discussion should occur between the patient, surgeon, and primary team. Many of these patients will benefit most from palliation. It is ideal to have this discussion early, while the patient is stable and lucid.
0
Conclusion of the case
0
The patient was stabilized in the ICU with intubation, vasopressors, and central line placement. CT scan revealed duodenal perforation. Due to worsening mutiorgan failure, the patient was not deemed to be a surgical candidate. Despite aggressive care the patient did not survive.
0
Take-home points
- Secondary bacterial peritonitis should always be considered before making a diagnosis of spontaneous bacterial peritonitis.
- Red flags for secondary bacterial peritonitis:
- Frank peritoneal signs on physical examination
- Point-of-care ultrasonography showing free air
- Gram stain with multiple organisms
- Treatment failure
- Septations should not be seen in sterile ascitic fluid from cirrhosis. The presence of septations suggests infection, either primary or secondary bacterial peritonitis.
- When concerned regarding the possibility of secondary bacterial peritonitis, there should be a low threshold to obtain a CT scan.
Coauthored with Paul Farkas, my father & senior consultant in Gastroenterology.
Latest posts by Josh Farkas (see all)
- PulmCrit Blogitorial – Use of ECGs for management of (sub)massive PE - March 24, 2024
- PulmCrit Wee: Propofol induced eyelid opening apraxia – the struggle is real - March 20, 2024
- PulmCrit wee: Why I like central lines for GI bleed resuscitation - March 13, 2024
not hopeful for a response but worth a shot. thoughts on leaving paracentesis catheter in place for recurrent drainage, frequent cell counts and following cultures till “negative” in a patient with polymicrobial secondary peritonitis from a perforation + cirrhosis, that is not a surgical candidate. thanks