Laboratory quality
Endotoxins
What endotoxins are, why they are dangerous even in sterile fluids, and how the LAL assay reports the number on a CoA.
- Last updated:
- 2026-07-22
Overview
Endotoxins are lipopolysaccharide molecules from the outer membrane of Gram-negative bacteria. They are stable to autoclaving and to most chemical treatments, and they trigger a strong inflammatory response even in nanogram amounts.
Because sterilisation kills bacteria but does not always destroy their released endotoxin, a sterile product can still contain enough endotoxin to be pyrogenic.
Why it matters
Endotoxin exposure produces fevers, chills, hypotension, and in severe cases septic-shock-like reactions. Injectable products in the pharmaceutical world therefore have specific endotoxin limits (typically expressed in Endotoxin Units per kg body weight per hour, EU/kg/hr).
Key concepts
LAL assay
The Limulus Amebocyte Lysate (LAL) assay is the standard endotoxin test. It uses an extract from horseshoe-crab blood cells that clots in the presence of endotoxin. Modern recombinant equivalents (rFC) exist and produce comparable results.
Units and thresholds
Endotoxin is reported in Endotoxin Units per volume or per mass (EU/mL or EU/mg). A CoA reporting "<0.5 EU/mg" is saying the assay's detection limit was not exceeded, not that endotoxin is truly absent.
Sterile does not mean low-endotoxin
Sterility and endotoxin are separate specifications tested by separate assays.
Common misconceptions
- "Sterile" implies low endotoxin. It does not.
- Endotoxin is destroyed by refrigeration. It is not.
- The absence of a fever after a first exposure means the vial is endotoxin-free. Endotoxin reactions vary by dose and individual, and a single tolerated exposure is not evidence of safety.
Practical interpretation
Where a CoA reports an endotoxin number under LAL or rFC, that is meaningful evidence about that batch. Where it does not, endotoxin is an unknown.
Limitations
Endotoxin testing has a detection limit and a method. Comparing endotoxin numbers across CoAs from different labs and methods can mislead if the underlying assays differ.