Refining Target-Controlled Infusion: An Assessment of Pharmacodynamic Target-Controlled Infusion of Propofol and Remifentanil Using a Response Surface Model of Their Combined Effects on Bispectral Index

Short, Timothy G. MBChB, MD, FANZCA*†; Hannam, Jacqueline A. PhD*; Laurent, Stephen MBChB, FANZCA; Campbell, Douglas BM, FRCA, FANZCA; Misur, Martin BHB, MBChB, FANZCA; Merry, Alan F. MBChB, FANZCA*†; Tam, Yuk Ho BSc, MPhil, CEng, MIET, MHKIE

doi: 10.1213/ANE.0000000000000386
Anesthetic Pharmacology: Research Report

BACKGROUND: Propofol and remifentanil are commonly combined for total IV anesthesia. The pharmacokinetics (PK), pharmacodynamics (PD), and drug interactions of the combination are well understood, but the use of a combined PK and PD model to control target-controlled infusion pumps has not been investigated. In this study, we prospectively tested the accuracy of a PD target-controlled infusion algorithm for propofol and remifentanil using a response surface model of their combined effects on Bispectral Index (BIS).

METHODS: Effect-site, target-controlled infusions of propofol and remifentanil were given using an algorithm based on standard PK models linked to a PD response surface model of their combined effects on BIS. The combination of a targeted BIS value and adjustable ratio of propofol to remifentanil was used to adjust infusion rates. The standard model performance measures of median performance error (bias) and median absolute performance error (inaccuracy), expressed as percentages, were used to assess accuracy of the infusions in a convenience sample of 50 adult patients undergoing surgery with general anesthesia. The influence of age and weight on the performance of the model was also assessed.

RESULTS: Patients had a mean (range) age of 48 (19–73) years, weight of 80 (45–169) kg, and body mass index of 28 (19–45) kg/m2. The overall model had a bias of 8% (SD 24%) and inaccuracy of 25% (SD 13%). Performance was least accurate during the early induction phase of anesthesia. There was no significant bias in BIS predictions with increasing age (P = 0.44) or weight (P = 0.56).

CONCLUSIONS: The algorithm performed adequately in a clinical setting. The algorithm could be further refined, and assessment of its accuracy and utility in comparison to current clinical practice for giving IV anesthesia is warranted.

Published ahead of print September 2, 2015

From the *Department of Anaesthesiology, University of Auckland, Auckland, New Zealand; Department of Anaesthesia, Auckland City Hospital, Auckland, New Zealand; and Department of Anaesthesia and Intensive Care, The Chinese University of Hong Kong, Shatin, Hong Kong, China.

Accepted for publication June 9, 2014.

Published ahead of print September 2, 2015

Funding: Jacqueline A. Hannam received a doctoral stipend from the Green Lane Research and Educational Fund.

The authors declare no conflicts of interest.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website.

Reprints will not be available from the authors.

Address correspondence to Timothy G. Short, MBChB, MD, FANZCA, Department of Anaesthesia, Auckland City Hospital, Private Bag 92024, Auckland, New Zealand. Address e-mail to

© 2016 International Anesthesia Research Society