Summary: Researchers detected brain-wave patterns consistent with “hidden” consciousness in about 15% of patients with acute traumatic brain injury within days of the event. Patients showing these EEG patterns were more likely to regain the ability to follow verbal commands before hospital discharge and had better functional outcomes at one year compared with patients without these signals.
Source: Columbia University
Overview
Close analysis of bedside EEG recordings reveals that nearly one in seven critically ill patients with acute brain injury show signs of covert consciousness within days of injury. Neurologists at Columbia University and NewYork-Presbyterian report that patients exhibiting these EEG signatures are more likely to recover command-following and to achieve better functional status at follow-up.
Why this matters
Predicting recovery after acute brain injury remains a major clinical challenge in the ICU. Decisions about the intensity of life-sustaining treatment are often made within the first days to weeks after injury, yet current bedside tests and imaging provide imperfect prognostic information. Because electroencephalography (EEG) is available at most hospitals and can be performed at the bedside, it represents a practical tool to identify patients who may have preserved cognitive processing despite appearing behaviorally unresponsive.
Background
Over the past decade, research has shown that some patients who seem unresponsive can nevertheless exhibit brain activity consistent with understanding commands. Early reports used MRI to detect this dissociation between behavior and brain activity, and later work using EEG found similar signals in chronically unresponsive patients. However, MRI is often impractical in critically ill patients, and there has been a need to study whether routine EEG could detect these signs in the acute setting when treatment decisions are being made.
What the researchers did
The Columbia team studied 104 adult ICU patients who were clinically unresponsive within days after acute brain injury caused by hemorrhage, trauma, or oxygen deprivation. Although none of the patients could speak or follow commands reliably, most were not paralyzed. Each day, clinicians performed standard neurological exams and repeatedly gave simple spoken motor commands—for example, asking the patient to open and close their hand or to stop moving.
Researchers recorded EEG during these command sessions and applied a machine-learning algorithm to detect reproducible differences in brain activity between different commands. A consistent, command-specific pattern on EEG was interpreted as evidence that the patient perceived and processed the instruction even if they could not carry out the movement.
Study findings
EEG signatures consistent with hidden consciousness were detected in 16 of 104 patients (15%) at a median of four days after injury. Among these 16 patients, 8 (50%) later regained the ability to follow commands before hospital discharge, compared with 23 of 88 patients (26%) who did not show these EEG patterns.
At one year, functional outcome was also better among patients who showed early EEG evidence of brain activation. Seven of 16 patients (44%) with early EEG signals achieved independent function for up to eight hours per day (a Glasgow Outcome Scale–Extended score of 4 or higher), while only 12 of 84 patients (14%) without such signals reached that level. Mortality at one year was similar between groups, with roughly one-third of patients in each group deceased.
The EEG indications of hidden consciousness were observed more often after traumatic injury and hemorrhage than after hypoxic injury, although the study was not large enough to establish firm conclusions about differences by cause.
Caveats and limitations
Although the results are encouraging, larger studies are needed to validate the predictive value of EEG-based detection of covert consciousness, particularly studies focused on specific injury types. Practical challenges remain: EEG monitoring must be started early and repeated because consciousness can fluctuate after severe brain injury; spoken commands must be precisely synchronized with EEG recordings; and computational analysis requires additional hardware and expertise.
Careful testing design is essential because poorly controlled paradigms or improper computational methods can mistake artifacts for genuine brain activation. For these reasons, standardized protocols and automated, validated analysis tools will be important before widespread clinical adoption.
Conclusions
This study suggests that routine bedside EEG, combined with advanced analysis, can detect covert brain activation in a meaningful minority of patients soon after acute brain injury. Patients who show these EEG signatures are more likely to recover command-following during hospitalization and to attain better functional outcomes at one year. If replicated in larger cohorts, EEG detection of hidden consciousness could enhance prognostic assessments and guide treatment decisions in the ICU.
Funding and disclosures
The study was supported by the Dana Foundation and the James S. McDonnell Foundation. The authors reported no relevant financial conflicts of interest. Dr. Rohaut received postdoctoral grants from several French medical organizations and the Philippe Foundation.
Study citation
Original research: “Detection of Brain Activation in Unresponsive Patients with Acute Brain Injury.” New England Journal of Medicine. Jan Claassen et al. DOI: 10.1056/NEJMoa1812757. (Closed access)
Source: Columbia University
Media contact: Jan Claassen – Columbia University
Image source: public domain