OBJECTIVE: Each supraglottic airway requires different anesthetic depth because it has a specific structure and different compressive force in the oropharyngeal cavity. We designed the study to compare the effect-site concentration (Ce) of remifentanil in 50 % of patients (EC50) for successful insertion of the i-gel second-generation supraglottic airway device with that for laryngeal mask airway (LMA) insertion during target-controlled infusion (TCI) of propofol. METHODS: Forty-one female patients were randomized to the i-gel group (n = 20) or the LMA group (n = 21). Anesthesia was induced with propofol Ce of 5 mug/ml and the predetermined remifentanil Ce, and the i-gel or LMA was inserted 5 min later. The remifentanil Ce was estimated by modified Dixon's up-and-down method (initial concentration: 3.0 ng/ml, step size: 0.5 ng/ml). The patient's response to device insertion was classified as either "success (no movement)" or "failure (movement)". RESULTS: Using the Dixon's up-and-down method, EC50 of remifentanil Ce for the i-gel (1.58 +/- 0.41 ng/ml) was significantly lower than that for LMA (2.25 +/- 0.55 ng/ml) (p = 0.038). Using isotonic regression, EC50 (83 % CI) of remifentanil in the i-gel group [1.50 (1.37-1.80) ng/ml] was statistically lower than that in the LMA group [2.00 (1.82-2.34) ng/ml]. EC95 (95 % CI) of remifentanil in the i-gel group [2.38 (1.48-2.50) ng/ml] was statistically lower than that in the LMA group [3.35 (2.58-3.48) ng/ml]. CONCLUSIONS: We found that EC50 of remifentanil Ce for i-gel insertion (1.58 ng/ml) was significantly lower than that for LMA insertion (2.25 ng/ml) in female patients during propofol TCI without neuromuscular blockade.