We hypothesized that the concentration of propofol in the aqueous phase may be the most important variable responsible for the pain experienced during injection of the drug. The concentration of propofol in the aqueous phase (18.57 micrograms/mL) can be decreased by increasing the fat content of the solvent. To test this hypothesis, 36 patients were randomly allocated to one of three groups, each receiving a different formulation of propofol. Group A received 20 mL of propofol alone in a commercial preparation (Diprivan(R) with 10 mL of saline); Group B, 20 mL of propofol to which 5 mL of long-chain triglyceride (LCT) fat emulsion and 5 mL of saline and been added; and Group C, 20 mL of propofol and 10 mL of LCT fat emulsion. The propofol emulsion was injected over 30-60 s into a dorsal vein of the hand. Patients reported pain during injection as none, mild, moderate, or severe (almost intolerable). In Group A, 8 of 12 patients reported moderate or severe pain upon injection whereas in Group C only mild pain was reported by 6 of 12 patients. Our results suggest that a smaller concentration of propofol in the aqueous phase of the emulsion reduces pain on injection. With the addition of more lipid (10 mL), a higher percentage of propofol is absorbed by fat particles. If solvents that permit a smaller concentration of the drug in the aqueous phase of oil-in-water emulsions were used for propofol and other drugs that cause pain on injection, pain would be reduced and patient satisfaction may be increased.