To compare intravenous, buccal midazolam & rectal diazepam in abating seizures in an emergency setting. Study design: Prospective, non-blinded, Randomized Controlled research.

Children aged 1 month -18 years who presented and developed active seizures in the emergency room in a tertiary hospital from May 1, 2006 -April 30, 2007 were included. Those who have received rescue treatment at home or pre- hospital, with known allergic reaction to either midazolam or diazepam were excluded. Daily randomization sequence was done by drawlots. The dose of buccal/ intravenous midazolam and rectal diazepam administered was designed to give about 0.5mg/kg. Standardized data form was used to record demographic data, adverse effects, seizure type, duration, time to terminate and seizure recurrence. The primary outcome measure was the cessation of clinical seizures within 10 minutes of administration of the randomized drug. If the child was still seizing despite the 1^st treatment line, additional medication was based upon the seizure management protocol of the section.

Pearson correlation coefficient, Chi square, Kruskal Wallis Test.

A total of 34 patients were enrolled in the study. Six were randomized under the rectal gtoup while 15 & 13 children were assigned to the intravenous & buccal Midazolam group respectively. There were l O (29%) females & 24 (71 %) males with age range from 1 month up to 15 years. The top 3 etiologies of the seizures included epilepsy 35%, CNS infections 23% & hyperi:ensive encephalopathy 1 8%. With rectal Diazepam, the mean time to control seizure was 5 minutes & was effective in all (1 00%) patients. With intravenous Midazolam, the mean rime to control seizures was recorded at 7 minutes & 33 seconds & was effective in 14 of 1 5 (93%) subjects. The patients under the buccal route responded similarly with the intravenous Midazolam group with cessation of seizures in 12 of 13 (92%) children. Comparing the three-treatment group using Kruskall Wallis, KW value of0.619 is not significant at 0.05 levels. The seizures of32 (94%) of the subjects were controlled within 10 minutes. Comparing seizure duration with rime to control the attack using the Pearson Correlation, this yielded a coefficient of 0.203 and found to be not significant at 0.05 levels. Comparing the seizure recurrence of the three treatment arms using Chi-square, it showed x2 value of 24.824, which is significant at 0.05 levels. There were no adverse effects reported in chis study.

Buccal Midazolam is safe & effective in cessation of seizures in all age groups irrespective of the type and etiology of seizures. There was no statistical significance in time to terminate seizures between the three treatment groups. It has the additional advantage of ease in administration and less seizure recurrence.