Summary Objectives/Aim This randomized control trial was designed to evaluate the incidence of emergence delirium ( ED ) in preschool children receiving sevoflurane or desflurane anesthesia combined with an effective caudal block. Background While ED has been described in children receiving sevoflurane or desflurane anesthesia, a direct comparison between the two agents using a validated ED assessment tool has not been reported previously. Methods/Materials Two hundred and sixty preschool children scheduled for elective sub‐umbilical surgery were randomized to receive sevoflurane or desflurane anesthesia combined with a caudal block. ED was defined as a Paediatric Anesthesia Emergence Delirium scale ( PAED ) ≥10 points. A delirium‐specific score ( ED I) was calculated from the first three items of the PAED score (eye contact, purposeful actions, awareness of the surroundings) and a nonspecific score ( ED II ) from the last two items on the PAED score (restlessness and inconsolability) to test the hypothesis that some items of the PAED scale may better reflect clinical ED than others. Results Thirty‐one (25%) children in each group demonstrated ED after awakening with ED being of shorter duration in the desflurane group than the sevoflurane group. An ED I score of nine points had a sensitivity of 0.93 and a specificity of 0.94 for ED . An ED II score of five points had a sensitivity of 0.34 and specificity of 0.95 for ED . Conclusion Sevoflurane and desflurane anesthesia were associated with similar incidences of ED in children undergoing sub‐umbilical surgery and receiving effective regional anesthesia. High scores on the first three items of the PAED scale were highly correlated with ED . The items restlessness and inconsolability had lower sensitivity for the diagnosis of ED .