The purpose was to study the agreement between cardiac output measurements with electrical velocimetry vs. intermittent thermodilution before and after coronary artery bypass graft surgery.

Cardiac output was measured simultaneously with electrical velocimetry and intermittent thermodilution before and immediately after coronary artery bypass graft surgery, and in the intensive care unit. Measurements were performed in three different body positions. The results were analysed according to Bland and Altman.

The mean bias of all 150 paired measurements in 16 patients was 0.21 ± 0.78 L min−1, and the mean error was 40%. Before skin incision the mean bias was 0.04 ± 0.41 L min−1, and the mean error was 25%. After skin closure the mean bias was 0.57 ± 0.92 L min−1, and the mean error was 42%. In the intensive care unit the mean bias was 0.26 ± 0.68 L min−1, and the mean error was 32%.

The agreement between cardiac output measurements with electrical velocimetry and intermittent thermodilution was clinically acceptable only before skin incision in coronary artery bypass graft surgery. The mean error was unacceptably high immediately after skin closure and was at a borderline level in the intensive care unit. Thus, the overall accuracy of cardiac output measurements with the electrical velocimetry technique during coronary artery bypass graft surgery is not clinically unacceptable.

This study was conducted to compare bispectral index, state entropy and response entropy in patients undergoing coronary artery bypass grafting.

In 66 patients, anaesthesia was maintained at two different levels using bispectral index. Doses of sufentanil and midazolam were adjusted to achieve a bispectral index in the range of 45–55 in 33 patients (BIS 50 group) and 35–44 in another 33 patients (BIS 40 group). Simultaneously, state entropy and response entropy were recorded.

The targeted values of bispectral index were achieved in both groups and the bispectral index values differed significantly during whole anaesthesia. Median response entropy and state entropy fell to 19–26 during anaesthesia in both groups. Response entropy and state entropy values in the two groups differed significantly only after induction of anaesthesia and did not differ during further anaesthesia. There was no explicit intraoperative recall in both groups. Patients in Group BIS 40 received significantly (P < 0.05) more sufentanil than the BIS 50 group (704 ± 181 μg vs. 490 ± 107 μg, respectively) and midazolam (18.5 ± 6.1 mg vs. 15.6 ± 3.8 mg, respectively). After cardiopulmonary bypass, significantly (P < 0.05) more patients in Group BIS40 needed inotropic support with dobutamine (79%) than in the BIS50 group (52%). Time to extubation did not differ between the two groups.

In patients undergoing coronary artery bypass grafting, no relationship was found between bispectral index levels and state entropy and response entropy at two different stages of a sufentanil–midazolam anaesthesia. A bispectral index level of 45–55 reduced anaesthetic medications used and the need for inotropic support.