Two hundred and nineteen children (boys: 56%, girls: 44%) were included in a randomized, double-blind, multicentre (4 centres) controlled trial designed to assess the efficacy and safety of ibuprofen (IBU) in the treatment of 1 to 6 year-old children with otoscopically proven acute otitis media (AOM), either unilateral or bilateral. They randomly received 10 mg/kg IBU (n = 71), or acetaminophen (PARA) (n = 73) or placebo (PLA) (n = 75), orally, tid, for 48 hours. All received oral cefaclor (Alfatil, Lilly, France) for seven days. They were evaluated before (D0) and at the end of treatment (D2). The main criterion of response was the aspect (landmarks and color) of the tympanic membrane assessed on a semi-quantitative scale from 0 to 6. Other criteria, assessed on semi-quantitative scales, included relief of pain (0 or 1), rectal temperature (0 to 2), and overall evaluation by parents of the improvement of quality of life on three items: appetite (0 to 2), sleep (0 to 2), and playing activity (0 to 2). The results at D2 were as follows: there was no significant difference between treatment groups as to the main criterion, but only a trend for IBU and PARA to do better than PLA but not for IBU to do better than PARA. From these data there is no argument to emphasize the utility of non-steroidal anti-inflammatory drugs (NSAIDs) in treating the inflammatory signs of the tympanic membrane in otitis. There was a statistically significant difference between treatment groups at D2 for pain, IBU being superior to PLA (P < 0.01): 7%, 10% and 25% of the children were still suffering at D2 in the IBU, PARA and PLA treatment groups, respectively. The difference between PARA and PLA for pain was not statistically significant. There was no significant difference between treatment groups for the other criteria. All treatments were well and equally tolerated. Although no significant difference was found between the treatment groups on the aspect of the tympanic membrane, the efficacy of IBU was evidenced on the relief of pain, the symptom that most disturbs the child.