| Soul 2021 |
Low risk of bias |
Triple‐masked, sequence generation random and baseline characteristics does not reveal any imbalance between the two groups. |
Low risk of bias |
Double‐blinded study and neither the participants nor the treating physicians were aware of the allocation |
Low risk of bias |
Data reasonably complete for all the included patients |
Low risk of bias |
aEEG was utilised to detect seizures which is a reliable method when compared to clinical diagnosis of seizures |
Low risk of bias |
Trial analysed as per a priori registered protocol. |
Some concerns |
Some concerns in one domain |
