CLASSIFICATION OF PATIENT SIMULATORS
No accepted classification scheme has been devised for the use
of patient simulators in anesthesia. Any classification involves some overlapping
and gray areas. The following classification and definitions are used in this chapter.
In addition to simulators, there are also computer-assisted instruction programs
and computer-based training devices. Although training devices may replicate certain
portions of the clinical domain, they do not attempt to replicate the bulk of the
work environment. Some individuals consider the screen-only microsimulator (also
known as a screen-based simulator) to be a training device and not a simulator.
Computer-assisted instruction programs and training devices are not reviewed in this
chapter. Moreover, this chapter is about "patient simulators" in anesthesia that
present the overall patient as would be seen by anesthesiologists, intensivists,
or others. Devices related purely to emergency medicine (e.g., ResusSim) and surgical
or procedural simulators that present only the technical work of surgery or invasive
manipulations (e.g., bronchoscopy, intravenous access), as well as partial-task screen-based
simulators (e.g., GasMan), are also not discussed here.
Thus, for this chapter a patient "simulator" is a system that
presents a patient and a clinical work environment of immediate relevance to anesthetists
(e.g., operating room [OR], postanesthesia care unit [PACU], intensive care unit
[ICU]) in one of the following ways:
- In actual physical reality, defined as a mannequin-based simulator (former
and still commonly used synonyms are full-scale simulator, hands-on simulator, realistic
simulator, high-fidelity simulator). These simulators can be subdivided in terms
of the way that vital signs monitoring is accomplished (interfacing to real clinical
monitoring devices or using its own virtual replica of a monitor screen) and in terms
of the primary control logic of the simulator (by individual controls and scripts
or by using physiologic and pharmacologic modeling).
- On a computer screen only, defined as a screen-only or screen-based simulator
(some prefer the term "microsimulator").
- Using "virtual reality," defined as a virtual reality simulator. In such
a device, parts or all of the patient and environment is presented to the user by
three-dimensional representations with or without "touch" to create a more "immersive"
experience. A screen-only simulator could be viewed as a very limited virtual reality
simulator.
Components of a Patient Simulator
A patient simulator system contains several components ( Fig.
84-1
). A set of outputs make up a representation of the patient, the clinical
environment, and diagnostic and therapeutic equipment. For screen-only simulators,
this representation is generated graphically on the computer screen. For mannequin-based
simulators, the representation is
Figure 84-1
Schematic diagram of the generic architecture of model-based
patient simulator systems. The simulator generates a representation of the patient
and the work environment with appropriate interface hardware, display technologies,
or both. The representation is perceived by the anesthesiologist, whose actions
are then input to the simulator through physical actions or input devices. The behavior
of the simulated patient is determined through sets of interlinked physiologic models
and control logic that are manipulated by the instructor or operator through a workstation
that allows selection of different patients, abnormal events, and other features
of the simulation. ICU, intensive care unit; OR, operating room.
generated by using a patient mannequin, plus either actual clinical equipment or
virtual replicas of monitor screens, placed (typically) in a re-creation of an actual
clinical setting. The mannequin and, where appropriate, the clinical equipment are
stimulated or actuated by interface hardware. Mannequin-based simulators often use
physical stimulation of clinical equipment in addition to electronic stimulation.
For example, the mannequin can actually be ventilated with any desired mixture of
inspired gases. Carbon dioxide and other gases can be introduced physically by the
simulator into the mannequin's lungs to provide the desired elements of gas exchange.
A real respiratory gas analyzer can thus be used to measure the inspired and expired
gases ( Table 84-2
).
Any simulator must have control logic by which changes in the
simulated patient's condition can be generated, controlled, and sent to the appropriate
output of the representation. Originally, the control logic was embedded in the
software as a fixed sequence of events, or it consisted largely of continuous input
from the instructor working from a script. Some current simulators use upgraded
types of manual control logic that allows scripting of combinations of changes in
control input. Other current simulators incorporate a more sophisticated technique
using mathematical differential equations that model a patient's physiology and pharmacology
to provide the bulk of the control logic. These models can be tailored to represent
different patients with different pathophysiologic abnormalities. However, not all
states or changes in a patient can be modeled by differential equations. For example,
ventricular fibrillation is a totally different state of heart rhythm that does not
evolve continuously from normal rhythms. No model can predict exactly when a patient
will suffer a myocardial infarction or when an ischemic heart will begin to fibrillate.
A model can only predict factors that increase the likelihood of such events. Thus,
most simulators incorporate other modeling techniques in addition to the basic physiologic
and pharmacologic mathematical equations, including finite-state models, instructor
initiation of abnormal events, and even manual modulation of modeled parameters.
In finite-state models, different underlying clinical states are defined, each of
which has appropriate entry conditions, as well as transition conditions to other
states. When an entry or transition condition is met, a new state becomes active,
which may directly trigger new observable phenomena (e.g., ventricular fibrillation)
or may alter constants in the mathematical models that then evolve in time.[4]
[5]
[6]
The control logic of most simulators is manipulated through an
instructor/operator's station (IOS) that allows the instructor to create specific
patients, select and implement abnormal events and faults, and monitor the progress
of the simulation session. The system may also have a remote-controlled hand-held
IOS in addition to the main IOS. The IOS typically provides logs of physiologic
changes and the anesthesiologist's response and may provide graphics to support the
analysis of a simulation run. Some screen-based simulators provide advice and tutorials
linked to the management of simulated events. With mannequin-based simulators, especially
for applications in which complete work environments are re-created, it is common
to obtain detailed records of the simulation and the actions taken from video and
audio recording of the personnel working in the replicated clinical environment.
Modern simulators still do not provide a number of the desired
features listed in Table 84-3
.
As shown in Table 84-4
,
different simulator systems are available. No single system is the best choice for
all uses, so the decision about the type of simulator required must be based on the
objectives and needs of the application. Moreover, in our experience, the success
of a simulator program will not be determined primarily by the type or fidelity of
the simulator used, but mostly by the enthusiasm, skill, and creativity of instructors,
as well as the time and effort devoted to preparing and performing credible simulation
scenarios.[7]
