In Response: We appreciate the opportunity to reply to Dr. Behringer and would like to comment on each of the points raised. First, we disagree that the results between the single-institute trial [1] and the multicenter trial [2] are in "sharp contrast." In terms of ease of placement, time taken to achieve an adequate airway, first time placement success rates, airway sealing pressure, number of airway interventions required, and postoperative jaw/neck pain, both trials showed that the laryngeal mask airway (LMA) was the superior device. The main contrasting result was that the multicenter trial [2] showed that the cuffed oropharyngeal airway (COPA) had a less frequent incidence of sore throat and the single-center trial [1] showed a less frequent incidence with the LMA. We consider the postoperative data from the single-institute trial more reliable because data collection was double-blinded and the severity was graded, unlike the multicenter trial. Second, we did not make "protestations" that clinical expertise was comparable. We accurately described the experience of the investigators with both devices. Furthermore, in an attempt to measure the performance, we scored the quality of use of each device on a six-point scale using video recordings assessed by an independent investigator. We found that the score was high and similar for both devices, which suggests that both devices were being used according to the manufacturer's recommendations. We apologize for the lack of availability of the "in press" reference, which was accepted for publication by Anaesthesia in September 1997. We made no statement that "the plateau of the COPA's learning curve is reached with as few as 50 insertions." What we found was that there was no evidence of skill acquisition over the first 50 uses when comparing 25-use epochs. We therefore considered that inclusion of investigators with a minimum of 15-use experience [average >40 uses, as chosen in a recent trial [3]] to be reasonable. In the Discussion, we did state, "it is likely that performance of the COPA may improve with increasing individual experience, understanding, and application of the anatomical issues involved." Third, we accept that the use of different drug doses may favor one device or the other, but the optimal regimens are unknown for both devices, although more information is available for the LMA. From a methodological viewpoint, when comparing two devices, it is important to use the same drug regimen; otherwise, the outcome may be drug-related rather than device-related. Two recent studies have compared the response to insertion between the LMA and COPA. Nakata et al. [4] showed that the COPA required a lower minimum alveolar anesthetic concentration of sevoflurane for placement, but Versichelen et al. [3] showed similar hemodynamic and bispectral index changes after induction with propofol. Fourth, we do not consider that the maximal cuff volumes used (50 mL size 10; 60 mL size 11) contributed to the poor performance of the COPA. These volumes were recommended to us by the manufacturer for an earlier study and have since been revised. In our experience, higher volumes do not provide a better seal and are associated with unacceptably high in vivo intracuff pressures. Finally, we agree that the COPA deserves more careful scrutiny to fully define its appropriate clinical usage. However, a finding common to all published reports is that the COPA is not a consistently hands-free device. J. R. Brimacombe, MB ChB, FRCA, MD J. C. Brimacombe, BSc Department of Anaesthesia and Intensive Care; Cairns Base Hospital; Cairns, Australia A. M. Berry, MB ChB, FRCA Department of Anaesthesia and Intensive Care; Nambour General Hospital; Nambour, Australia