Comparison between cerebral state index and bispectral index as measures of electroencephalographic effects of sevoflurane using combined sigmoidal E(max) model

Dong Woo Han, Olinto-Jose Linares-Perdomo, Jong Seok Lee, Jun Ho Kim, Steven E Kern, Dong Woo Han, Olinto-Jose Linares-Perdomo, Jong Seok Lee, Jun Ho Kim, Steven E Kern

Abstract

Aim: The cerebral state index (CSI) was recently introduced as an electroencephalographic monitor for measuring the depth of anesthesia. We compared the performance of CSI to the bispectral index (BIS) as electroencephalographic measures of sevoflurane effect using two combined sigmoidal E(max) models.

Methods: Twenty adult patients scheduled for laparotomy were studied. After induction of general anesthesia, sevoflurane concentrations were progressively increased and then decreased over 70 min. An analysis of the BIS and CSI with the sevoflurane effect-site concentration was conducted using two combined sigmoidal E(max) models.

Results: The BIS and CSI decreased over the initial concentration range of sevoflurane and then reached a plateau in most patients. A further increase in sevoflurane concentration produced a secondary plateau in the pharmacodynamic response. The CSI was more strongly correlated with effect-site sevoflurane concentration (R(2)=0.95±0.04) than the BIS was (R(2)=0.87±0.07) (P<0.05). The individual E(max) and C(eff50) (effect-site concentration associated with 50% decrease from baseline to plateau) values for the upper and lower plateaus were significantly greater for BIS (12.7±7.3, 1.6±0.4, and 4.2±0.5, respectively) than for CSI (3.4±2.2, 1.2±0.4, and 3.8±0.5, respectively) (P<0.05). The remaining pharmacodynamic parameters for the BIS and CSI were similar.

Conclusion: The overall performance of the BIS and CSI during sevoflurane anesthesia was similar despite major differences in their algorithms. However, the CSI was more consistent and more sensitive to changes in sevoflurane concentration, whereas the measured BIS seemed to respond faster. The newly developed combined E(max) model adequately described the clinical data, including the pharmacodynamic plateau.

Figures

Figure 1
Figure 1
Observed value and predicted value of CSI vs time (A) and effect-site sevoflurane concentration (B) from one patient. The individual prediction value (E) is the sum of the two sigmoidal Emax models (E1 and E2). E1 represents the effect of sevoflurane concentrations from the baseline value to plateau value, while E2 remains constant. E2 represents the effect at sevoflurane concentrations from the plateau value to the maximal value while E1 remains constant.
Figure 2
Figure 2
Time course of the measured BIS (A) and CSI (B) obtained from each patient.
Figure 3
Figure 3
Relationship between the individually measured BIS and CSI values vs the measured sevoflurane end-tidal concentrations. The hysteresis in the relationship is shown in both plots.
Figure 4
Figure 4
Observed BIS values (A) and individually predicted values (B) for all patients vs the calculated sevoflurane effect-site concentrations. Observed CSI values (C) and individually predicted values (D) for all patients vs the calculated sevoflurane effect-site concentrations.
Figure 5
Figure 5
Relationship between the observed and individually predicted values of the BIS (A) and CSI (B). The solid line represents the line of identity.

Source: PubMed

3
Předplatit