The ProSeal laryngeal mask airway (PLMA) is a new laryngeal mask device with a larger, wedge-shaped cuff and a drainage tube. We tested the hypothesis that directly measured mucosal pressure and oropharyngeal leak pressure (OLP) are higher for the PLMA compared with the laryngeal mask airway (LMA). We also assess the mechanism of seal, and the reliability of cuff volume, in vivo intracuff pressure and calculated mucosal pressure (in vivo minus in vitro intracuff pressure) to predict directly measured mucosal pressure. Thirty-two anaesthetized, paralysed adult patients were randomly allocated to receive either a size 4 LMA or PLMA. Microchip sensors were attached at locations corresponding to: (a) base of tongue; (b) distal oropharynx; (c) hypopharynx; (d) lateral pharynx; (e) posterior pharynx; and (f) pyriform fossa. In vitro and in vivo intracuff pressures, OLP and directly measured mucosal pressure were documented at zero volume and after each 10 ml up to 40 ml. Directly measured mucosal pressure was similar between devices for a given cuff volume, but was lower for the PLMA for a given OLP. Directly measured mucosal pressure was highest in the distal oropharynx for both devices, but rarely (< 5%) exceeded 35 cm H2O. OLP was higher for the PLMA at all cuff volumes. Directly measured mucosal pressure was usually lower than OLP for both devices, and there was a positive correlation between directly measured mucosal pressure and OLP. Cuff volume, in vivo intracuff pressure and calculated mucosal pressure were poor to moderate predictors of directly measured mucosal pressure for the LMA and PLMA. We conclude that the PLMA forms a better seal than the LMA without an increase in directly measured mucosal pressure.