Physician clinical impression does not rule out spontaneous bacterial peritonitis in patients undergoing emergency department paracentesis. Ann Emerg Med. 2008 Sep;52(3):268-73. doi:10.1016/j.annemergmed.2008.02.016. Epub 2008 Apr 23
Clinician assessment and clinical characteristics are not sufficient to r/o or diagnose SBP. Patients presenting to the ED presenting for paracentesis cannot be ruled out from SBP from clinical judgment (sensitivity 76.5%) or any clinical characteristsics. Pts should get routine fluid analysis even when clinical suspicions are low.
Why is this important for EM
An increasingly large number of patients without primary or tertiary care to the ED for therapeutic paracentesis. The mortality of SBP even in treated patients is 20% so it is a can’t miss diagnosis in the ED. Understanding the limitations of clinical judgment for safe disposition with patients presenting with ascites can prevent poor outcomes.
Summary of test characteristics from H&P
|Subjective or measured Fevers in last 24 hrs||35.3% (6 of 17 patients)|
|Nausea/Vomiting||29.4% (5 of 17 patients)|
|Any abdominal pain*||94.1% (16 of 17 patients)|
*However, specificity is very low at 15.1%. 85% (107 of 126 patients) who did not have SBP also presented with some type of abdominal pain
Clinician Gestalt on Dx SBP
|All physicians||76.5% (26 of 44 patients)|
|Faculty||87.5% (14 of 16 patients)|
|Residents||66.7% (12 of 18 patients)|
Design & Results
Prospective, Observational Study of 3 urban ED centers serving a largely underserved population without access to tertiary hepatology services. Enrolled from April 2005-August 2006. Pts enrolled were those with ascites and were thought to require paracentesis/fluid analysis by physicians. The following clinical characteristics were collected 1) Sx within 24hrs of ED visit: N/V, hematochezia, melena, hematemesis. 2) ED Vitals and physician clinical impression: AMS, abd pain or SBP likelihood. A second physician repeated the clinical impression without knowledge of the first physician’s evaluation. All centers had a residency program and enrolling physicians had to be at least a PGY2 provider.
Primary outcomes were test characteristics of clinical characteristics and physician clinical assessment in making the diagnosis of SBP. SBP was defined as peritoneal fluid of ANC >250 cells/mm3 or pathologic bacteria growth on cultures.
The sample size of the study is small (n=144). Physicians’ clinical impression of SBP was made before viewing the ascitic fluid appearance which might have impacted impressions. Vital signs were only recorded up to the point when the paracentesis occurred and the physician made his impressions without vitals from the entire ED course. Also, serum laboratory results such as WBC and BMP were not available to many of the physicians. Both of these factors could have changed physicians’ impressions. The study was limited to patients who physicians thought needed fluid analysis which may have biased the data to those who were more likely to have SBP.