In a randomized double-blind study, 60 children, aged 4–7 years, undergoing dental extractions of six or more teeth under day-case general anaesthesia, were assigned to receive either tramadol drops 1.5 mg kg−1 (n=31), or placebo (normal saline) (n=29), 30 min before surgery. In addition, all received anxiolytic premedication of oral midazolam 0.5 mg kg−1 (max7.5 mg) at the same time. No differences were seen in behaviour, respiratory or cardiovascular assessments. In both groups, 93% were drowsy preanaesthetic, 3% were asleep but rousable and less than 4% exhibited minor distress. At induction, mild weeping occurred in 9.7% of the tramadol group and 6.9% of the placebo group (p>0.05). Active awake recovery took 48.8 min, SD 32.6 in the tramadol group and 36.4 min, SD 29.6 in the placebo group (p>0.05). Post-operative analgesia (paracetamol 120 mg) was given to 19.4% of the tramadol group compared with 82.8% of the placebo group (p>0.05), after which the Hannalah objective pain scale scores were comparable. Analysis of the Oucher six faces pain scale showed significantly better analgesia in the tramadol group at all time points, the pain score being half that of the placebo group at 60 min and one third from 60 to 120 min (p>0.05). No adverse respiratory or cardiovascular effects were seen. For children undergoing multiple extractions, 10.7, SD 3.0, effective postextraction analgesia was provided.