April 2002 / Volume 47 / Number 4 / Page 427
Closed-Loop Mechanical Ventilation
IntroductionClosed-loop mechanical ventilation encompasses a plethora of techniques, ranging from the very simple to the relatively complex. In the simplest form, closed-loop ventilation is the control of one output variable of the mechanical ventilator based on the measurement of an input variable. An example would be pressure support ventilation, in which flow (output) is constantly changing to maintain pressure (input) constant throughout inspiration. More complex forms of closed-loop ventilation involve measurement of multiple inputs (eg, compliance, oxygen saturation, respiratory rate) to control multiple outputs (eg, ventilator frequency, airway pressure, tidal volume). The latter type of control more closely mimics the ventilatory control and response of human physiology. This review discusses both currently available closed-loop ventilation techniques and those only available outside the United States, along with some cutting-edge techniques that have only limited use. The operation, theoretical advantages, and limitations of each technique are reviewed. When available, the literature supporting or refuting each technique will be reviewed, but, unfortunately, little has been published on certain techniques.
Commercially Available Closed-Loop Systems
Volume-Assured Pressure Support and Pressure Augmentation
Volume Support and Variable Pressure Support
Dual Control Breath-to-Breath, Pressure-Limited Time-Cycled Ventilation
AutoMode and Variable Pressure Support/Variable Pressure Control
Adaptive Support Ventilation
Mandatory Minute Volume
Closed-Loop Systems Not Currently Commercially Available in the United States
Proportional Assist Ventilation
Pressure Support Controlled by Airway Occlusion Pressure
Knowledge-Based System for Automatic Ventilatory Management
Fuzzy Logic Control of Mechanical Ventilation
Neurally Adjusted Ventilatory Assist
Closed-loop control of mechanical ventilation includes a variety of techniques, ranging from the relatively simple to the complex. Often clinicians think of closed-loop control as "complete automatic control" of mechanical ventilation. Yet that is only one facet of closed-loop control. It also happens to be the least realized and most complex.
As a simple example, pressure support ventilation (PSV) uses closed-loop control. The clinician sets a target pressure (the pressure support setting) and flow is automatically adjusted to maintain that pressure throughout inspiration. As the ventilator monitors airway pressure (the target), the control algorithm continuously modulates the flow (the output) to achieve the desired pressure. From a control standpoint, PSV is an intrabreath (control changes occur during a single breath) negative feedback controller. More complex examples of closed-loop control include the use of respiratory mechanics measurements, expired gas analysis, and direct or indirect measurements of blood gases as input signals (targets).
The complexities of closed-loop control systems and descriptions of feedback mechanisms are beyond the scope of this review. Interested readers should refer to the work of Brunner for a more complete discussion of those topics.
Herein we briefly describe the history of closed-loop control, review commercially available closed-loop techniques, and describe several systems now under consideration for use during ventilatory support.