This study was designed to determine the optimal dose of remifentanil required for the successful insertion of Streamlined Liner of the Pharynx Airway (SLIPA) and to compare it to that required for laryngeal mask airway (LMA) insertion in patients receiving a propofol infusion at a standard effect-site concentration. Fifty-eight patients scheduled to undergo general anaesthesia were randomly assigned to either the SLIPA (n = 29) or LMA (n = 29) group. All patients were premedicated with midazolam 0.05 mg x kg(-1) and glycopyrrolate 0.004 mg x kg(-1) intramuscularly. After the administration of lignocaine 1 mg x kg(-1) intravenously, a propofol infusion was started at an effect-site concentration of 3.5 microg x ml(-1) with a remifentanil infusion without a neuromuscular blocking agent. The remifentanil dose was determined by the modified up-and-down method starting in each group at 4 ng x ml(-1). Six minutes after induction, the airway device was inserted. Airway device insertion was classified as 'success' or 'failure' based on patient response. From the isotonic regression analysis and bootstrap distribution, the EC50 of remifentanil for SLIPA and LMA were 0.93 ng x ml(-1) (95% confidence interval [CI] 0.81 to 1.50 ng x ml(-1) and 1.36 ng x ml(-1) (95% CI 1.19 to 2.06 ng x ml(-1)) respectively, and the EC95 for SLIPA and LMA insertions were 1.90 ng x ml(-1) (95% CI 1.39 to 1.95) and 2.43 ng x ml(-1) (95% CI 1.80 to 2.46 ng x ml(-1)) respectively. Using the 83% CIs from the bootstrap distribution, EC50 for SLIPA was significantly less than that of LMA (0.83 to 1.23 vs 1.26 to 2.00, respectively) (P < 0.05). These findings suggest that the insertion of SLIPA needs about a 32% lower depth of anaesthesia than LMA insertion.