Drug interaction models are better predictors of tolerance/ response to noxious stimuli compared to individual measured parameters

Abstract

Background and Goal of Study: This study continues the pursuit of the parameter with the best correlation to the probability of response to noxious stimuli of different intensity. We used data from a previous study by Heyse et al.1 on the interaction of sevoflurane and remifentanil to compare several parameters. Materials and Methods: After institutional review board approval, 40 adult patients were randomised to receive different combinations of sevoflurane (Sevo) and remifentanil (Remi) according to a criss-cross design. After reaching pseudo-steady state, the patients were assessed for tolerance of 'shake and shout' (SAS), tetanic stimulation (TET), insertion of laryngeal mask airway (LMA) and laryngoscopy (LAR). Bispectral index (BIS), state and response entropy (SE, RE), composite variability index (CVI) and surgical pleth index (SPI) were either recorded or computed from raw electroencephalographic and plethysmographic data retrospectively. Sevo and Remi concentrations were recorded. The combined potency of Sevo and Remi according to the fixed C50O hierarchical interaction model (U) and the noxious stimulation response index (NSRI) were the population-based predictors. We used the prediction probability (PK) to assess the performance of these parameters on the probability of response. Bootstrapping (n=1000) was used to produce 84%-confidence intervals of the PKs, with significance being achieved if the confidence intervals did not overlap (p <0.05). Results and Discussion: The parameter PKs per stimulus are summarised in Table 1.(Table Presented) The PK for U and NSRI were highest for all stimuli. Effect site concentrations of either Sevo or Remi alone were significantly worse predictors. BIS, SE, RE and CVI were significantly worse at predicting tolerance to the three painful stimuli, but similar to U and NSRI for SAS. SPI performed poorly overall. Conclusion: U and NSRI perform significantly better than EEG-derived parameters and single drug effect site concentrations in predicting tolerance to noxious stimuli. Therefore both U and NSRI could be useful parameters in anaesthetic practice.