03/06/2026
I just finished teaching a webinar on septic peritonitis for vet-webinar and as always, going back to the literature before teaching reminded me why we should never stop questioning what we think we know.
🔬 That number we all quote? It’s more complicated than we thought.
If you work in small animal surgery or emergency medicine, you have probably quoted this at some point:
Blood-to-fluid glucose difference >20 mg/dL = septic peritonitis. 100% sensitive. 100% specific.
It comes from a small landmark 2003 study by Bonczynski et al. — and it stuck. Hard.
But when is something 100 percent certain in nature? Here is what we now know:
❌ When you use a point-of-care glucometer instead of a laboratory analyser, sensitivity drops to just 41%. Not 100%. 41%. Use plasma instead and raise cutoff to >2.1 mmol/L: sensitivity 88%, specificity 100% (Koenig & Verlander 2015).
❌ Similar gradients have been found in non-septic effusions — including cancer, pancreatitis and bile leakage.
So the test is still useful. But it is not the cutoff many of us were taught.
The research is changing. Our teaching needs to keep up.
Have you ever caught yourself quoting a number you later discovered was more complicated? 👇
