Introduction and aim: Botulinum neurotoxin type A (BoNT/A) selectively enters neurons and inhibits synaptic neurotransmitter exocytosis by proteolytic cleavage of synaptosomal-associated protein 25 (SNAP-25). The aim of this experiment was to evaluate peripheral and central anti-inflammatory and antinociceptive effects of BoNT/A in rats with complete Freund's adjuvant (CFA)- induced knee joint inflammation. ----- Materials and methods: Botulinum toxin was administered intrathecally (i.t.) at the level of lumbar spine (2 U/kg/10μl), or intraarticularly (i.a.) into the rat knee (7 U/kg/20μl). After 7 days, to induce inflammation, animals were injected with CFA (50 μl). The knee diameter was measured for evaluation of edema, while nociceptive behavioural response was assessed by employing the Pressure Application Measurement (PAM) device . The effect on plasma protein extravasation was evaluated using spectrophotometry by measuring the removed joint tissue content of Evans-blue- plasma protein complex , two days after induction of the inflammation. ----- Results: BoNT/A injected directly into the joint (i.a.) significantly reduced edema, with the peak effect on the second day of the inflammation. BoNT/A administered intrathecally didn't reduce edema. On the other hand, the significant reduction of plasma protein extravasation was observed only in BoNT/A i.t- injected animals. ----- Conclusion: In present experiments, BoNT/A reduces edema only when delivered in the peripheral nervous system, while it reduces plasma protein extravasation when administered into the CNS. Dependently on the toxin application site, differences in observed BoNT/A effects might be caused by different inflammatory mediators affected. Further experiments to determine the molecular mechanisms of BoNT/A anti-inflammatory action are needed.