VALIDATION OF THE ALFENTANIL CANONICAL UNIVARIATE PARAMETERS AS A MEASURE OF OPIOID EFFECT ON THE ELECTROENCEPHALOGRAM

Background: Several parameters derived from the multivariate electroen cephalographic (EEG) signal have been used to characterize the effects of opioids on the central nervous system. These parameters were formu lated on an empirical basis. A new statistical method, semilinear cano nical correlation, has been used to construct a new EEG parameter (a c ertain combination of the powers in the EEG power spectrum) that corre lates maximally with the concentration of alfentanil at the effect sit e. To date, this new canonical univariate parameter (CUP) has been tes ted only in a small sample of subjects receiving alfentanil. Methods: The CUP was tested on EEG data from prior studies of the effect of fiv e opioids: alfentanil (n = 5), fentanyl (n = 15), sufentanil (n = 11), trefentanil (n = 5), and remifentanil (n = 8). We compared the CUP to the commonly used EEG parameter spectral edge, SE(95%). The compariso n was based on the signal to noise ratio, obtained by fitting a nonlin ear pharmacodynamic model to both parameters. The pharmacodynamic para meter estimates obtained using both measurements were also compared. R esults: The values for signal-to-noise ratio were significantly greate r for the CUP than for SE(95%) when considering all the opioids at onc e. The pharmacodynamic estimates were similar between the two EEG para meters and with previously published results. Semilinear canonical cor relation coefficients estimated within each drug group showed patterns similar to each other and to the coefficients in the CUP, but differe nt from coefficients for propofol and midazolam. Conclusions: Although the CUP was originally designed and tested using alfentanil, we have proven it to be a general measure of opioid effect on the EEG.