Background: Epileptic encephalopathy with spike-wave activation in sleep (EE-SWAS) is a rare syndrome associated with cognitive and behavioural regression. On the basis of mostly small observational and retrospective studies, corticosteroids and clobazam are often considered the most effective treatments for this syndrome. We aimed to compare cognitive outcomes of children with EE-SWAS 6 months after starting treatment with either corticosteroids or clobazam. Methods: We did a multicentre, randomised controlled trial at eight tertiary referral centres for rare epilepsies in seven European countries. Children were eligible to participate if they were aged 2–12 years, were diagnosed with EE-SWAS within 6 months before inclusion, and had not been treated with corticosteroids or clobazam previously. Participants were randomly assigned (1:1) to treatment with corticosteroids (either continuous treatment with 1–2 mg/kg per day of prednisolone orally or pulse treatment with 20 mg/kg per day of methylprednisolone intravenously for 3 days every 4 weeks) or clobazam (0·5–1·2 mg/kg per day orally). The primary outcome was cognitive functioning after 6 months of treatment, which was assessed by either the intelligence quotient (IQ) responder rate (defined as improvement of ≥11·25 IQ points) or the cognitive sum score responder rate (defined as improvement of ≥0·75 points). Safety was assessed by number of adverse events and serious adverse events. Data were analysed in the intention-to-treat population, which included all children as randomised who had primary outcome data available at 6 months. The trial is registered with the Dutch Trial Register, Toetsingonline, NL43510.041.13, and the ISRCTN registry, ISRCTN42686094. The trial was terminated prematurely because enrolment of the predefined number of 130 participants was deemed not feasible. Findings: Between July 22, 2014, and Sept 3, 2022, 45 children were randomly assigned to either corticosteroids (n=22) or clobazam (n=23); two children assigned clobazam dropped out before 6 months and were excluded from the intention-to-treat analysis. At the 6-month assessment, an improvement of 11·25 IQ points or greater was reported for five (25%) of 20 children assigned corticosteroids versus zero (0%) of 18 assigned clobazam (risk ratio [RR] 10·0, 95% CI 1·2–1310·4; p=0·025). An improvement of 0·75 points or more in the cognitive sum score was recorded for one (5%) of 22 children assigned corticosteroids versus one (5%) of 21 children assigned clobazam (RR 1·0, 95% CI 0·1–11·7, p=0·97). Adverse events occurred in ten (45%) of 22 children who received corticosteroids, most frequently weight gain, and in 11 (52%) of 21 children who received clobazam, most often fatigue and behavioural disturbances. Occurrence of adverse events did not differ between groups (RR 0·8, 95% CI 0·4–1·4; p=0·65). Serious adverse events occurred in one child in the corticosteroid group (hospitalisation due to laryngitis) and in two children in the clobazam group (hospitalisation due to seizure aggravation, and respiratory tract infection). No deaths were reported. Interpretation: The trial was terminated prematurely, and the target sample size was not met, so our findings must be interpreted with caution. Our data indicated an improvement in IQ outcomes with corticosteroids compared with clobazam treatment, but no difference was seen in cognitive sum score. Our findings strengthen those from previous uncontrolled studies that support the early use of corticosteroids for children with EE-SWAS. Funding: EpilepsieNL, WKZ fund, European Clinical Research Infrastructure Network, and Ming fund.