Paucis Verbis: Strength of diagnostic tests for cholecystitis
Posted Mar 18 2011 12:00am
You have a 40 year-old man who presents to the ED for persistent right upper quadrant abdominal pain for 12 hours after eating a fatty meal. He has no fevers, nausea, flank pain, or dysuria. His physical exam shows no fever and only moderate tenderness in the RUQ without guarding. He has a Murphy's sign which is improved after a total of 8 mg of IV morphine. His laboratory results, which include a WBC, liver function tests, lipase, and urinalysis, are normal.
Can you safely say that the patient doesn't have cholecystitis? Can you discharge him for outpatient ultrasonography to assess for symptomatic cholelithiasis?
As bedside ultrasonography becomes more of a staple in Emergency Departments, it is easy to just perform the ultrasound yourself if such a patient presents. If you do not have an ED ultrasound available, however, you need to send this patient for a formal ultrasound because he is still very much at risk for cholecystitis despite having unremarkable lab tests and no fever.
JAMA published a meta-analysis of 17 studies on the test characteristics for cholecystitis. I found it odd that they defined a fever as temperature >35 Celsius. The best performing characteristic was a Murphy's sign, although the positive likelihood ratio (LR) slightly crossed 1.0 (0.8-8.6).