ED50 of desflurane for LMA removal: is the estimate reliable?

摘要

We read with interest Hui et al.'s recent article on the 50% and 95% effective doses (ED50 and ED95) of desflurane for removal of the classic laryngeal mask airway (LMA) in spontaneously breathing anaesthetised adults [1]. We would like to make some points regarding the methodology of the study and analysis of the data. The authors chose 6.0% as the initial concentration of desflurane. The choice of starting dose was probably inappropriate as shown by the sequence of repeated identical outcomes. Dixon and Massey's up-and-down methodology demands that the starting dose should be the minimum dose expected to result in a positive response, i.e. close to the expected ED50 [2]. Starting concentrations larger than the population ED50 yield average values slightly greater than population ED50 and vice versa [3]. Also, if the starting concentration deviates markedly from the population ED50, a larger sample size is required to obtain the desired number of crossovers. Reduction in the sample size by 30-40%, which is the primary advantage of Dixon method, is thereby lost [2].