Non-Human Primate Model for Developing Closed-Loop Anesthesia Delivery Systems

ABSTRACT/PROJECT SUMMARY
Continuous monitoring of physiological state (oxygenation, breathing, circulation) is a standard practice for all
patients receiving general anesthesia and sedation. Anesthetics produce their primary effects of
unconsciousness and antinociception by acting on molecular targets and neural circuits in the brain and central
nervous system. Nevertheless, continuous monitoring of brain function is not a practice requirement. It is no
surprise that brain dysfunction following general anesthesia is highly prevalent, particularly among the elderly.
Similarly, COVID 19 patients who can be anesthetized for weeks in the intensive care unit, are often left with
profound brain dysfunction following termination of ventilatory support. Many years of research have shown that
the level of unconsciousness of a patient receiving general anesthesia can be reliably tracked using real-time
processing of electroencephalogram (EEG) recordings. In recent years, dramatic advances have been made in
sensors, actuators, artificial intelligence and control theory algorithms. A highly plausible solution is the
development of closed loop anesthesia delivery (CLAD) systems that determine in real time from the EEG the
patient’s level of unconsciousness and precisely control an anesthetic infusion to maintain the level at an
appropriate target. The Federal Drug Administration (FDA) readily acknowledges the significant enhancement
to patient care that CLAD systems can provide. To date, no system has been approved for human use due to a
lack of appropriate animal models to test adequately the reliability and robustness of these systems. Therefore
the research design of this project will be to conduct in non-human primates neurophysiological recordings (EEG,
local field potentials and neural spiking activity) while simultaneously administering anesthetics using a
computer-controlled syringe pump as the animals execute a behavior task to characterize level of
unconsciousness. The data will be analyzed by combining pharmacokinetics and pharmacodynamic modeling,
modern control theory and statistical signal processing approaches to develop and test real-time CLAD systems.
The specific aims of this research project are to develop and test in a non-human primate model, CLAD systems
for real-time control of unconsciousness using the anesthetics: propofol, dexmedetomidine, and propofol and
dexmedetomidine administered simultaneously. The broad long-term objectives are to: establish a non-human
primate model paradigm for development and testing of CLAD systems; and make the use of CLAD systems a
standard for intelligent brain state monitoring and precise second-to-second drug dosing in anesthesiology. The
health relatedness impact of the research will be a new paradigm for computer-assisted vigilance of brain state
and computer-assisted dosing of anesthetic agents. Such systems should enhance patient safety by reducing
provider errors and by fostering significant decreases in anesthesia-associate brain dysfunction as well as other
anesthesia-related morbidities (inadequate pain control, hypotension, nausea) commonly experienced by the
millions of patients who each year receive anesthesia care in operating rooms and intensive care units.

Grant Number: 5R01NS123120-05
NIH Institute/Center: NIH
Principal Investigator: EMERY BROWN