We have studied the sedation achieved with a mixture of midazolam (0.56 mg/kg-1) and ketamine (5 mg/kg-1) administered nasally in 30 children weighing less than 16 kg undergoing computerised tomography. Assessment was two fold using a visual analogue scale; the radiologist/radiographer rated the exam from "failed examination" to "perfect working conditions" while the anesthetist's assessment ranged from "poor sedation" to "perfect sedation with clinical well being". This new method proved to be effective alone in 83% of the cases and there were no complications. The rapid onset obtained after intranasal midazolam and ketamine offers advantages over orally or rectally administered drugs. The absence of respiratory depression and oxygen desaturation suggests that this technique is safe and efficient in the CT room with its particular working conditions.