Bispectral index under propofol anesthesia in children: a comparative randomized study between TIVA and TCI

In children, only a few studies have compared different modes of propofol infusion during a total intravenous anesthesia (TIVA) with propofol and remifentanil. The aim of this study was to compare Bispectral Index (BIS) profiles (percentage of time spent at adequate BIS values) between four modes of propofol infusion: titration of the infusion rate on clinical signs (TIVA0 ), titration of the infusion rate on the BIS (TIVABIS ), target controlled infusion (TCI) guided by the BIS either with the Kataria model (TCI KBIS ) or the Schnider model (TCI SBIS ).

Sixty-six children (aged from 4 to 14 years) were prospectively randomized into one of the four groups. In the TIVA0 group, the anesthesiologist was blinded to the BIS. In each group, the percentage of time with adequate BIS values (45-55), the bias, and imprecision were calculated.

The propofol consumption was similar in the four groups. During the maintenance phase, the percentage of time spent in the targeted BIS range was significantly lower in the TIVA0 group compared to the three other groups (TIVA0 : 31% ± 22, TIVABIS : 59% ± 17, TCI KBIS : 53% ± 12, TCI SBIS : 56% ± 17). The bias was not statistically different between the four groups, but the imprecision was larger for the TIVA0 group. Compared to the Kataria model, the Schnider model was associated with shorter time delay to reach the desired BIS, to eyes opening, and to tracheal extubation.

Propofol administration using manual infusion guided by clinical signs was associated with higher risks of over- or underdosage when compared to BIS-guided administrations. When propofol infusion was guided by the BIS, no major difference was found between TIVA and TCI (either with the Kataria or the Schnider model). This study highlights the need of a pharmacodynamic feedback during propofol anesthesia in children.