Comparison of SNAP™ II and BIS Vista indices during normothermic cardiopulmonary bypass under isoflurane anesthesia

Abstract

Processed EEG monitoring during cardiopulmonary bypass (CPB) may help determine loss of consciousness and depth of anesthesia. This study compared the SNAP(™) II and BIS Vista monitors in patients undergoing isoflurane anesthesia with normothermic CPB. 40 subjects undergoing CPB with isoflurane anesthesia were enrolled. Subjects were premedicated with 1-2 mg midazolam approximately 5 min prior to acquisition of baseline index values and anesthesia induced with midazolam and fentanyl. Anesthesia was maintained with isoflurane, midazolam, and fentanyl and a cis-atracurium infusion. SNAP(™) II (version 1.2.9 algorithm 1.88) and BIS Vista (application version 3.00 platform version 2.03) indices were recorded at baseline, pre-induction, post-intubation, incision, start of CPB, every 15 min during CPB, end of CPB, and end of case. Agreement between methods was determined using Pearson correlation and the Bland-Altman method with repeated observa- tions. Twenty-four male and 12 female subjects completed the analysis. The correlation between SNAP(™) II and BIS Vista index values was 0.61 (P < 0.005). A linear relationship between the difference in the indices and the average index values was observed following the induction of anesthesia. In awake subjects, the bias between the SNAP(™) II and BIS Vista was 5 (95% CI 3-7). The limits of agreement were 23 (95% CI 19-26) and -13 (95% CI -9--16). During anesthesia, the mean difference on a log scale was 0.11 (95% CI 0.09-0.12). The limits of agreement were 0.43 (95% CI 0.40-0.45) and -0.21 (95% CI -0.18--0.24). The antilog of the mean difference demonstrated that the SNAP(™) II value was 28% (95% CI 24-33%) higher than the BIS Vista value following induction of anesthesia. The SNAP(™) II monitor demonstrates a consistently positive bias during cardiopulmonary bypass under isoflurane anesthesia compared with the BIS Vista.