ObjectivesTo record the number of children with withdrawal symptoms after the administration of sedatives for mechanical ventilation, and to discuss the possible connection with the administration of midazolam.DesignRetrospective data collection from case records and charts.SettingMedical and surgical intensive care unit (ICU) in a university hospital.PatientsChildren 6 months to 14 yrs of age who required sedation for mechanical ventilation (n = 40).InterventionsNone.Measurements and Main Results60 mg/kg was strongly significantly associated with occurrence of withdrawal. Statistical analysis to determine the occurrence of withdrawal associated with the administration of morphine was not possible.ConclusionsSigns and symptoms of a withdrawal reaction were observed in several children. The occurrence of withdrawal was statistically related to high doses of midazolam, but it was not possible to determine the influence of morphine. If large doses of midazolam and opioids have been administered, there may be justification for reducing the dose gradually instead of abruptly, or using longer-acting benzodiazepines or opiods on discontinuation of sedation. (Crit Care Med 1999; 27:196-199)