13/10/2026
Comment 'CRITERIA' and I'll send you the PDF.
Your 72-year-old with heart failure has bilateral effusions.
Three days of IV furosemide, then you tap it. Protein ratio
0.53. LDH ratio 0.64.
Light's criteria say exudate. So now you're thinking infection,
malignancy, PE — and you're about to work up a patient who has
plain old heart failure.
Do the subtraction instead.
Serum albumin minus pleural albumin: 3.6 − 1.9 = 1.7. Anything
above 1.2 g/dL says transudate.
Serum protein minus pleural protein: 7.2 − 3.8 = 3.4. Anything
above 3.1 g/dL says the same.
Both gradients say transudate. Light's says exudate. The
gradients are right.
Diuretics pull water out of the pleural space and leave the
protein and LDH behind. The fluid concentrates, the ratios
climb, and a transudate starts reading as an exudate. The
mechanism never changed. That's a pseudoexudate.
If you still want confirmation, pleural NT-proBNP above
1,500 pg/mL points hard at a cardiac origin.
One caveat worth holding onto: not every effusion in a heart
failure patient is cardiac. If the picture doesn't fit —
unilateral, febrile, pleuritic, not settling — investigate it
properly.
Light's criteria are sensitive. They are not specific. Treat the
patient, not the criteria.
Comment CRITERIA and the full PDF lands in your DMs.
