Waking up during surgery is a terrifying – and dreaded – experience for patients… But also for medical staff, who are always afraid of being responsible for such an event with all its consequences. This type of phenomenon is called “intraoperative awareness during general anesthesia”, defined as an unexpected awakening of the patient during surgery.
Although the origin of the phenomenon is still debated, researchers are beginning to offer hypotheses about its causes. For example, we know that when the depth of the general anesthesia, induced by the concentration of anesthetic used, is not sufficient to compensate for the surgical stimulation linked to the operation in progress, there is a risk of awakening. This is why, throughout the operation, the anesthesiologist monitors the concentration of the injected products and works to maintain sedation at an optimal level.
To monitor its depth, it can call on the observation of two types of parameters in the patient: – clinical characteristics (heart rate, blood pressure, movements, sweating), – particular patterns present in the cerebral activity of the patient via the electroencephalographic (EEG) signal.
For the anesthetist, the objective is to maintain, in the sleeping patient, a kind of complex balance where the dynamic balance between loss of consciousness, loss of painful sensation and loss of memory is in harmony. Indeed, if the anesthesia is too light, it can lead to consciousness or movement during the operation – which is very dangerous. Conversely, if too much anesthetic agent is administered, it increases the probability of death.
In the case of a high-risk anesthesia (for example on an elderly person suffering from a cardiovascular pathology), there will be a tendency to underdose the general anesthesia to avoid side effects and reduce the risk. And in general, an overdose may increase the symptoms of postoperative nausea and vomiting, thus delaying the patient’s discharge from the hospital.
The boundary between correctly dosed or too light anesthesia is often very thin in the end… Mainly due to the heterogeneity of the patients (age, weight, gender) and their physiological reaction to the products used.
The study of intraoperative awakening therefore appears to be fundamental today. First, because it responds to a real need, and falls within the framework of the improvement of monitoring and the quality of care desired by the Ministry of Health. Then, because insofar as none of the current techniques fully satisfies the detection of this type of awakening, it is obvious that a more reliable device would be welcome in hospitals.
Our work, which exploits the potential of brain-computer interfaces, brings new ideas to this field.
The number of patients affected by intraoperative awakenings is not to be taken lightly. If the incidence and occurrence of this phenomenon is debated, there is a consensus to place the number of intraoperative awakenings between 1 and 2% (in high-risk practices). This percentage may seem low, but some of the victims are not aware of their awakening and others do not testify: they are therefore not counted, hence an under-evaluation of the event.
Given the number of general anesthesias undertaken each year, the occurrence of this incident is alarming: approximately 7 million are performed per year in France, which makes more than 13,000 potential awakenings. Globally, hundreds of millions of general anesthesias are performed each year. This underlines the extent to which new solutions are necessary…
A phenomenon which is moreover still largely unpredictable since there is no satisfactory hypothesis on the biological factors which could increase the risk of awakening. Some assume a genetic origin, but without conclusive evidence. The most probable being that at a precise moment, there is an under-dosage in anesthetic or that the cerebral receptors concerned are less sensitive to the products used.
On the other hand, the harmful consequences of intraoperative awakening are well established. Such a return to consciousness can cause physical suffering and/or real panic. Patients have thus reported hearing the medical staff commenting on their current operation, feeling the incisions made, extreme pain, even the hands of the surgeon inside them… Which can cause psychological sequelae called post-traumatic stress disorder.
These syndromes can last for several years and have a serious impact on the victim’s life. Among the many known effects are documented depression, sleep disorders and personality changes, anxiety, insomnia, flashbacks, chronic fear or suicidal thoughts. These long-term psychological sequelae could affect nearly 70% of patients who reported regaining consciousness during their operation.
Intraoperative awakenings also generate a high level of anxiety among anesthesiologists and are among the top three causes of patients’ legal attacks on the hospital.
The sole monitoring of clinical signs does not allow sufficient detection of awakening, and therefore does not prevent it. They are in fact only indirect witnesses of the brain state of the patients, not subtle enough to always anticipate an awakening before it occurs.
New indexes using part of the electroencephalographic (EEG) signal at the frontal level (because it is easy and quick to place electrodes there) are being tested, such as the Bispectral Index (BIS, based on electrical activity in the brain ), the Patient State Index (PSI) or the Entropy. Although already used in hospitals, these approaches have not clearly demonstrated their superiority over clinical monitoring and their reliability is sometimes questioned by the anesthesiologists themselves.
Today, the problem can be summed up as follows: no signal-based monitoring of brain activity is satisfactory enough to effectively assess the depth of general anesthesia and even less to prevent intraoperative awakening. So we have to find new solutions, that’s what we’re working on.
For several years, we have been proposing to specifically study the motor brain activity of patients to prevent this type of awakening. For what ? Because it is now established that the first reaction of a patient in the process of waking up is to move to alert the medical staff of his terrifying situation… And it is this intention of movement that we propose to detect.
“I couldn’t breathe, couldn’t move or open my eyes, or even tell the doctors I wasn’t asleep…” (in “Patient perspectives on intraoperative awareness with explicit recall”, 2015)
However, during the majority of surgical procedures, the patient is “curarized”: among the products injected, he receives curare which causes a neuromuscular blockade in order to prevent any movement. The patient therefore finds himself in a complex situation where he wakes up and tries to warn the anesthesiologists by moving… but without being able to do so.
To detect this attempted movement, we propose to use already well-established knowledge in the field of brain-computer interfaces based on motor brain activity. In theory, the detection of a movement intention by such an interface is possible by analyzing the EEG signal from the motor cortex. Indeed, the preparation phase and execution of a movement intention lead to variations in power in the alpha and beta bands of the EEG signal (present in an awake subject, large amplitudes for the first, narrower for the seconds). These power variations can be detected by a brain-computer interface via the EEG and provide direct information on the state of cerebral activation of the motor cortex.
One of the great originalities of this research project is to exploit this phenomenon to better detect the patient’s movement intentions during their intraoperative awakening.
I have already demonstrated that stimulation of the median nerve (for the anterior aspect of the forearm and hand) is a very promising approach: indeed, it also induces changes in cortical activity visible in the signal EEG and comparable to a movement intention. When hand movement intention and median nerve stimulation are coupled, the [cerebral signature in the cortex (known as ERD/ERS%) is amplified and can be detected by a brain-computer interface] with high reliability .
It is this phenomenon that I intend to use to facilitate the detection of the patient’s attempt to move. It is even possible that it could be more generally used in the field of BCI (Brain-Computer Interface, brain-computer interface) to improve the detection of imaginations of movement, within the framework of motor rehabilitation after a vascular accident by example.
In order to design a brain-computer interface that would make it possible to detect intraoperative awakenings in patients, the BCI4IA project (Brain-Computer Interface for Intraoperative Awareness) was set up by a specifically created interdisciplinary consortium (see below). It has three major objectives:
Investigate the effect of anesthetics on the EEG signal of the motor cortex using new signal tracking techniques. If detecting a movement intention using a BCI is already possible, we do not know if the inhibitory effect of anesthetics will disturb the observed EEG signals. This data gap needs to be filled.
Achieving the ability to detect the patient’s movement intention without a “temporal marker”, i.e. being able to detect its neural signature in the EEG signal without knowing precisely when the patient will wake up.
Succeed in developing a brain-computer interface capable of operating in a clinical setting without a calibration phase. Current interfaces require prior calibration for each subject, because the characteristics of EEG activity vary greatly from one subject to another. However, in the context of the anesthesia of a patient, there will be no such individual reference EEG signals available. Classic BCI methods cannot therefore be used here.
As the number of general anesthetics performed each year increases, the need for increased patient monitoring will rapidly grow. The protocol that we are developing is not intended to be used alone, but to fit into a more global system, incorporating both monitoring of clinical criteria and new cerebral indices based on the electroencephalographic signal for the benefit of hundreds of thousands of people.
There is still a long way to go, but our work offers great prospects. Our original approach, consisting in studying motor cerebral activity in particular during stimulation of the median nerve as indirect proof of the state of consciousness, could really allow better monitoring of the patient during general anesthesia.
Indeed, it is in this way that the study of intraoperative awakening already sheds light on one of the great enigmas of our time: what is consciousness? The use of anesthetics indeed offers the extraordinary particularity of being able to suspend it in a reversible way, thus opening a new path for its understanding…
