The retrospective study was designed to examine the safety of propofol against other sedative agents when used by infusion for the sedation of children requiring mechanical ventilation. One-hundred-and-ninety-eight patients were recruited. One-hundred-and-six received propofol and 92 received other sedative agents for durations of 30 min to 156 days and 13 min to 11 days respectively. The mean infusion rate of propofol was 3.39 mg.kg-1.h-1. Sixty-one of the 92 patients in the nonpropofol group received midazolam at a mean infusion rate of 0.4 mg.kg-1.h-1. Forty-one patients developed clinical metabolic acidosis with five falling into the pathological range as defined. Seventeen received propofol and 24 another sedative agent. Seventy-eight percent of patients that became acidotic were under the age of three. No patients who became acidotic was noted to have lipaemic serum. Three of four patients were recorded as having lipaemic serum received propofol, however two of these patients along with the patient that received midazolam also received Intralipid. Overall mortality was similar in both sedation groups with 27 deaths being recorded. Thirteen patients received propofol. Five nonfatal adverse events occurred, three in patients that had received propofol. The findings of the survey suggest that propofol compares favourably with other sedative agents when used for sedating children in a paediatric intensive care unit.