This study examined the incidence of hyperamylasaemia, in the absence of other plausible causes of pancreatic dysfunction, in intensive care unit (ICU) patients who received propofol. One-hundred-and-seventy-two consecutive patients of a general ICU who stayed for more than 24 hours were studied. Patients with a diagnosis consistent with elevated serum amylase levels at admission were excluded from the study, as were patients who had received medications known to raise serum amylase levels. Forty-four patients 53 +/- 20 years of age and median duration of ICU stay of five days (range two to 55) were eligible. Thirty of those, aged 54 +/- 21 years and median duration of ICU stay of five days (range two to 27) received continuous infusion of propofol for sedation (maximum dose 45 microg/kg/min). Of the 30 patients who received propofol, 16 (53%) developed hyperamylasaemia (125 to 466 IU/l) after two to nine days of continuous infusion. Liver and kidney function remained normal throughout the observation period. Of the 14 patients who did not receive propofol (aged 51 +/- 18 years), only two (14%) developed hyperamylasaemia, a significantly lower incidence (P = 0.021). Propofol infusion is associated with biochemical evidence of pancreatic injury. Amylase levels monitoring of propofol-sedated patients is warranted.