To the Editor:
—Olsen and Dahan describe an analysis of the dynamic electroencephalographic (EEG) response to step changes in end-tidal concentration of isoflurane or sevoflurane. Understanding the dynamic and steady state responses of a system to a changing input is a prerequisite to designing a robust automatic control system. Unfortunately, the use of a single, fixed-size step change in concentration is suboptimal as a “forcing” function for several technical reasons, including (1) the absence within this function of many frequencies in the range of interest, and (2) the possible blinding to nonlinearities. The discipline of control systems engineering provides many better alternatives to the development of a dynamic response measurement of a complex “black box” system similar to an EEG response in a patient. We reported the use of one such technique (pseudorandom binary sequence testing) to measure the dynamic (impulse) response of canine EEG (spectral edge frequency) to volatile anesthetics. 1This work was later reported in a Ph.D. thesis. 2Historians of our specialty will also appreciate that Dr. N. T. Smith was administering sine-wave concentrations of agents at various frequencies to human volunteers and measuring EEG response in 1976. 3
