OBJECTIVE: To test whether anti-inflammatory and antioxidant drugs that inhibit the nuclear factor kappa light chain enhancer of activated B cells (NF-kB), celastrol and edaravone, suppress spasms and improve developmental outcomes in the multiple-hit rat model of refractory infantile spasms (IS) due to structural lesions. METHODS: Postnatal day 3 (PN3) Sprague-Dawley rats were treated according to the multiple-hit IS model protocol. Using a randomized, blinded, vehicle-controlled, dose- and time-response study design, we tested the effects of single celastrol [1, 2, or 4 mg/kg intraperitoneally (i.p.), 10-14 rats/group] or edaravone (1, 10 or 30 mg/kg i.p., 14-17 rats/group) injections vs their vehicles on behavioral and electroclinical spasms and developmental milestones. Video-EEG monitoring was done on PN6-7 (n = 11-12 rats/group). Pulse celastrol treatment effects (PN4: 4 mg/kg, PN5-6: 2 mg/kg/day i.p.) were determined on spasms, developmental milestones and Barnes maze. Celastrol and edaravone pharmacokinetics in plasma and neocortex were assessed. Linear mixed model analysis of raw or normalized log-transformed spasm frequencies, considering repeated observations was used. RESULTS: Single (2-4 mg/kg i.p) or pulse celastrol, but not edaravone, reduced behavioral and electroclinical spasms frequencies within 5hrs. Pulse celastrol did not affect spasm-freedom, survival, developmental milestones or Barnes maze performance. Celastrol had erratic i.p. absorption with maximum concentrations observed between 2-4 h, when effects on spasms were seen. Edaravone had low blood-to-brain permeability. CONCLUSIONS: Celastrol's efficacy on spasms is partially explained by its better brain penetration than edaravone's. NFkB inhibitors may be useful in treating drug-resistant IS but delivery methods with improved bioavailability and brain permeability are needed.