Pilot study for classification of attempted movement from the electroencephalogram in healthy participants with a temporary paralysis of one arm induced by local administration of rocuronium

Awareness during anesthesia is defined as consciousness during surgery and
recall of intraoperative events when general anesthesia was intended, possibly
leading to intra-operative pain, panic and anxiety. Whereas the main cause for
awareness is an insufficient depth of anesthesia (Ghoneim et al., 2009),
possible undesired consequences also exist for the opposite case of too deep
anesthesia, e.g. hypotension and longer durations of recovery (Kent & Domino,
2009). This means there is only a narrow adequate range of depth of anesthesia,
hence administration should be carefully monitored. However, anesthetic depth
is not a single measurable variable, rather a complex reflection of the state
of the central nervous system. In addition to more traditional ways of
monitoring depth of anesthesia, such as measurement of changes in blood
pressure, heart rate, sweating and tear production, several types of EEG-based
monitoring devices have been introduced, e.g. the Bispectral Index (BIS, Aspect
Medical Systems, Massachusetts) and the Entropy Module (GE Healthcare,
Helsinki). Despite these developments, unintended awareness during general
anesthesia is still an unresolved issue with a current incidence of 0.1 to
0.2%, amounting to approximately 26,000 cases annually in the United States
alone (Sebel et al., 2004).

General anesthesia involves the simultaneous administration of different
components including neuromuscular blocking agents for immobilization. As a
consequence of this induced paralysis, patients trying to move in order to
alert the surgeon or anesthetist when awareness occurs fail repeatedly (Ghoneim
et al., 2009, Sandin et al., 2000). Obvious parallels can be drawn with
patients who are (partly) paralyzed by disease and for whom new methods of
communication are currently under development. In certain Brain-Computer
Interface (BCI) paradigms, frequency information in the EEG signal over the
motor cortex during (imagined) movement or planning of (imagined) movement is
used to create interaction between the user and a computer or other device
(Pfurtscheller & Neuper, 2006).

Based on this principle, we propose the development of a monitor of
intraoperative awareness by means of detecting attempts to move. In the
proposed paradigm the event-related desynchronization (ERD) and -
synchronization (ERS), features commonly used in movement-based BCI's, will be
exploited to detect a patient's urge to move and therefore function as input to
a 'brain switch' type monitor.

Previous investigations in our lab have provided insight in the requirements
for development of such a system and shown the feasibility of the paradigm
(Blokland et al., 2011). We have been able to optimize the settings of our
proposed system to such an extent that it is both fast and accurate. All
research until now has however focused on healthy, unanesthetized participants
only, performing either actual or imagined movement. To get a full
understanding of the applicability of the paradigm during general anesthesia,
we first need to determine how reliably attempted movement can be detected when
actual motor output is blocked. The aim of the present study is to investigate
how well our developed system is able to detect attempted movements from
participants of whom one arm is temporarily paralyzed by means of
administration of a neuromuscular blocking agent.

We hypothesize that attempted but pharmacologically blocked movement can be
detected from EEG (i.e., with a classification rate higher than chance),
possibly as well as imagined movement or even isometric movement.

By recording the EEG from participants instructed to attempt hand movement even
though motor output is blocked, it can be tested whether our system is able to
reliably detect these attempted movements.

In an observational study consisting of a single session per subject (lasting
at most three and a half hours), the electroencephalogram (EEG) will be
measured while the subject performs several movement tasks, both before and
after administration of a neuromuscular block (rocuronium) to one arm isolated
with a tourniquet. For a more elaborate study design please see chapters 3 and
6 of the research protocol.

Participation involves one session of at most three and a half hours, including
EEG montage and measurement, as well as administration of a neuromuscular block
to one arm isolated with a tourniquet.

Possible risks of participation in this study are:

-iv-needle: pain, hematoma, infection
-tourniquet: pain, nerve or tissue damage through local pressure or ischemia.
Important to note here is the fact that tourniquets are regularly used for up
to 2 hours during surgery if the surgeon wants to operate bloodlessly on an
extremity; with the short use we are intending, risks are minimal.
-Rocuronium: allergic reactions, systemic muscle relaxant effects
-Suggamadex: allergic reactions

All risks are considered relatively unlikely to occur.
Although the procedures might make participants feel uncomfortable at times,
they are not considered harmful.