Evaluation of a closed-loop muscle relaxation control system

Anesth Analg. 2005 Sep;101(3):758-764. doi: 10.1213/01.ane.0000167069.54613.50.

Abstract

Automatic muscle relaxation control may reduce anesthesiologists' workload freeing them for other patient care requirements. In this report we describe a muscle relaxation controller designed for routine clinical application using rocuronium and the train-of-four count. A muscle relaxation monitor (TOF Watch SX) was connected to a laptop computer running a controller algorithm program that communicates with a syringe pump to form a closed-loop muscle relaxation system. The control algorithm uses proportional-integral and lookup table components and is designed to avoid the usability restrictions of existing controllers. The controller is optimized using an objective method to avoid the uncertainties of ''hand-crafted'' controller algorithms. Controller target was train-of-four count 1 or 2 and controller performance was evaluated in 15 patients. During 39 hours of closed-loop control, 96.1% of all twitches recorded were in the target range. Average rocuronium infusion rate was 0.36 mg.kg(-1).h(-1) (sd 0.18 mg.kg(-1).h(-1)). We show that the controller remains useful even in the presence of disturbances that can arise in routine clinical conditions. The muscle relaxation controller maintained the target train-of-four count values and may serve as a basis for the design of hardware and user interfaces for closed-loop muscle relaxation control in clinical conditions.

MeSH terms

  • Aged
  • Algorithms
  • Androstanols / administration & dosage
  • Anesthesia*
  • Anesthesiology / instrumentation*
  • Automation
  • Female
  • Humans
  • Male
  • Middle Aged
  • Monitoring, Intraoperative
  • Muscle Relaxation / drug effects*
  • Muscle, Skeletal / drug effects*
  • Neuromuscular Nondepolarizing Agents / administration & dosage
  • Rocuronium

Substances

  • Androstanols
  • Neuromuscular Nondepolarizing Agents
  • Rocuronium