Background: S-ketamine (the S-isomer of ketamine) is twice as potent as the racemic mixture of this agent and carries fewer side effects when administered to humans. Information regarding the use of S-ketamine for the prevention of emergence delirium (ED) is limited. Thus, we evaluated the effect of S-ketamine administered at the end of anesthesia on ED in preschool children undergoing tonsillectomy and/or adenoidectomy.Methods: We investigated 108 children aged 3–7 years, who were scheduled for elective tonsillectomy and/or adenoidectomy under general anesthesia. They were randomly assigned to receive either S-ketamine 0.2 mg/kg or an equal volume of normal saline at the end of anesthesia. The primary outcome was the highest score on the pediatric anesthesia ED (PAED) scale during the first 30 min post-surgery. The secondary outcomes included the incidence of ED (defined as a score of ≥ 3 on Aono scale), pain score, time to extubation, and incidences of adverse events. Multivariate analyses were also performed using logistic regression to evaluate the independent factors predictive of ED.Results: The median (interquartile range) PAED score of the S-ketamine group (0 [0, 3]) was significantly lower than that in the control group (1 [0, 7]) (estimate median difference = 0, 95% confidence interval −2 to 0, p = 0.040). Significantly fewer patients in the S-ketamine group had an Aono scale score ≥ 3 (4 [7%] vs. 12 [22%], p = 0.030). Patients in the S-ketamine group also had a lower median pain score than did control subjects (4 [4, 6] vs. 6 [5, 8], p = 0.002). The time to extubation and incidences of adverse events were comparable between the two groups. However, multivariate analyses indicated that except S-ketamine use, pain scores, age and duration of anesthesia were independent factors predictive of ED.Conclusion: S-ketamine (0.2 mg/kg) administered at the end of anesthesia effectively reduced the incidence and severity of ED in preschool children undergoing tonsillectomy and/or adenoidectomy without prolonging the time to extubation or increasing adverse events. However, S-ketamine use was not an independent factor predictive of ED.