Disorders of Electrophysiology Following Severe Traumatic Brain Injury

Traumatic brain injury (TBI) produces electrophysiologic abnormalities that are critical to monitor. EEG is required to diagnose most seizures (Sz) that occur after severe TBI, provides information about background activity, and informs prognostic assessment. Non-invasive scalp EEG reflects large areas of the cortex, whereas direct recording of cortex, electrocorticography (ECOG), is invasive but provides more detailed information.The clinical management of electrophysiologic disorders has limited evidence and there is considerable equipoise about the prevention and management of Sz. Nonetheless, the use of scalp EEG, subdural strip ECOG, and depth electrode recordings in conjunction with other monitoring modalities form a complementary set of data that can be informative to bedside clinicians.

Key points

•

Scalp EEG is a non-invasive method of recording background activity across the brain and necesary to detect electrographic seizures (Sz) in patients with severe TBI.
•

Electrocorticography (ECOG) records from the cortex directly, detects seizures that are not apparent on scalp EEG, and is necessary to identify spreading depolarizations.
•

Most Sz in patients with severe TBI are nonconvulsive but are common and both result from and contribute to secondary brain injury.
•

Spreading depolarizations (SD) are twice as common as Sz and act as a marker and mechanism of secondary brain injury.
•

The use of scalp EEG and ECOG monitoring provides distinct but complementary information.

Abbreviations

ABI	acute brain injury
ACNS	American Clinical Neurophysiology Society
AMS	altered mental status
ASM	antiseizure medication
BIPD	bilateral independent periodic discharges
cEEG	Continuous electroencephalography
EEG	electroencephalography
ECOG	electrocorticography
ESICM	European Society of Intensive Care Medicine
GCS	Glasgow Coma Scale
GRADE	grading of recommendations, Assessment, Development, and Evaluations
ICU	intensive care unit
ICH	intracerebral hemorrhage
IIC	ictal-interictal continuum
LEV	levetiracetam
LOC	loss of consciousness
LPD	lateralized periodic discharges
LRDA	lateralized rhythmic delta activity
NCS	Neurocritical Care Society
NCSE	nonconvulsive status epilepticus
NCSz	nonconvulsive seizures
PHT	phenytoin
PTA	post-traumatic amnesia
SD	spreading depolarizations
SDH	subdural hematoma
SE	status epilepticus
sTBI	severe traumatic brain injury
Sz	Seizures
TBI	traumatic brain injury
TOAST	Trial Of Anti-Seizure Therapy

Electrophysiologic monitoring in traumatic brain injury

The management of patients with traumatic brain injury (TBI) in neurocritical care focuses on the prevention, detection, and management of secondary brain injuries. However, most secondary brain injuries occur silently, often in patients for whom the clinical examination is confounded by coma or sedation. These secondary brain injuries require technologies that allow neurointensivists to monitor the brain, ideally using continuous measures of physiology that can detect clinically important events in real-time. Over the last 40 years, the use of electrophysiologic monitoring has been increasingly recognized as an important tool for the management of patients with acute neurologic injuries in the intensive care unit (ICU), including those with TBI.

Existing recommendations for the use of electrophysiologic monitoring in ICU patients with TBI center around the use of continuous, noninvasive scalp electroencephalography (cEEG) monitoring, as summarized in Table 1 . ^, These guidelines and consensus statements provide strong recommendations despite relatively weak levels of evidence for the use of cEEG in patients with TBI for the diagnosis of seizures (Sz) without motor manifestations, or nonconvulsive seizures (NCSz) and nonconvulsive status epilepticus (NCSE). More broadly, there are recommendations for the use of cEEG in any patient with an acute brain injury and unexplained or persistent altered levels of consciousness, or those at high-risk for Sz because of their underlying injuries. A recent eDelphi consensus process identified cEEG as a core component of multimodality neuromonitoring strategies.

Table 1

Recommendations for electroencephalography in neurotrauma

Organization	Indication	Recommendation Type	Level of Evidence	Rationale	Reference
ACNS	Moderate-to-severe TBI	Expert consensus	—	Detection of NCSz Prognosis	Herman et al, 2015
ESICM	TBI, unexplained AMS	Evidence-based consensus (GRADE)	Strong recommendation, low quality of evidence (1C)	Detection of NCSz	Claassen et al, 2013
ESICM	sTBI at high-risk (cortical contusion/hematoma, depressed skull fracture, penetrating injury)	Evidence-based consensus (GRADE)	Weak recommendation, low quality of evidence (2C)	Detection of NCSz	Claassen et al, 2015; Rubiano et al, 2020
NCS	TBI-associated ICH	Evidence-based consensus (GRADE)	Useful, benefits outweigh risks (Class 1), limited evidence (level B)	Detection of SE	Brophy et al, 2012
NCS/ESICM	ABI (including TBI) with unexplained AMS	Evidence-based consensus (GRADE)	Strong recommendation, low quality of evidence	Detection of NCSz	Le Roux et al, 2014

Abbreviations: ABI, acute brain injury; AMS, altered mental status; ACNS, American Clinical Neurophysiology Society; ESICM, European Society of Intensive Care Medicine; GRADE, grading of recommendations, Assessment, Development, and Evaluations; ICH, intracerebral hemorrhage; NCS, Neurocritical Care Society; NCSz, nonconvulsive seizures; sTBI, severe traumatic brain injury; SE, status epilepticus.

∗No guideline recommendations offered by Brain Trauma Foundation (43) for EEG; no EEG-based guidelines or recommendations for traumatic brain injury are offered by Society for Critical Care Medicine or American Academy of Neurology.

Electroencephalographic monitoring techniques

Traditional noninvasive scalp EEG relies on a series of nonpolarizable Ag/AgCl electrodes placed on the scalp to record small electrical potentials on the order of microvolts, an order of magnitude smaller than the voltage changes recorded during electrocardiography. Scalp EEG electrodes are placed at standardized locations across the scalp as defined by the International 10 to 20 System. Electrical voltage is commonly recorded at 256 Hz to resolve the standard high frequency signals observed on the scalp EEG, between 0.5 and 30 Hz. As a physiologic monitor, EEG provides both high spatial and temporal resolution with relatively low signal-to-noise ratio (SNR).

Recordings directly from the cortex, or electrocorticography (ECOG), involve invasive placement of smaller, polarizable electrodes made from platinum/iridium, which avoids the neurotoxicity of Ag/AgCl. ECOG recordings sample a much smaller region of cortex but are less susceptible to environmental or external artifacts, yielding higher SNR. ECOG also allows for the measurement of full-band EEG, or frequencies that are much slower or much faster than those traditionally recorded in the 0.5 to 30 Hz spectrum. ECOG recordings can be made either from the cortical surface via subdural electrodes arrayed on a flexible silicon strip or from within the cortex using depth electrodes on a small flexible catheter.

Placement of a subdural strip electrode involves direct visualization of the cortex during surgical craniotomy or craniectomy, and the length of the strip is aligned along a contiguous gyrus typically from an area of injury penumbra toward healthier-appearing cortex. This yields a spatial representation of potentials underlying each electrode across the ∼6.5 cm length of the strip. Strip electrode placement after acute brain injury is relatively safe, and the electrodes are removed at bedside using gentle traction.

Depth electrodes are typically placed by piercing the cortex directly via a bedside burr hole. Electrodes span ∼1.5 cm of vertical distance, whereas the cortical mantle is typically ∼0.5 cm; thus, 3 or 4 of the depth electrodes measure directly from a single region of cortex. Depth electrodes placed within multilumen bolts enable correlation with other physiologic measurements from nearby invasive intracranial monitoring devices. ^, Each EEG monitoring modality provides distinct but complementary information ( Fig. 1 ).

Continuous scalp electroencephalography: background

EEG background refers to the frequency content and organization of the electrical potentials across a spatial array of electrodes ( Fig. 2 ). The characteristics of the EEG background are predicated on state: awake, drowsy, or asleep. In patients with moderate to severe TBI, these states cannot be delineated but often “arousal” versus “nonarousal” may be distinguished. The transition between 2 states based on changes in the background frequency content or organization is referred to as “reactivity.” A normal “awake” EEG exhibits 3 principal characteristics: continuity, an anterior-to-posterior gradient with a posterior dominant rhythm, and reactivity. As the dominant frequency over the posterior regions of the head begins to slow from normal (8.5–12 Hz), the background is characterized by “mild diffuse slowing.” As the typical organization of EEG amplitude and frequencies is lost, the background exhibits “moderate diffuse slowing.” Finally, as the frequency content of the EEG becomes delta (<4 Hz) predominant without organization and, particularly when there is little to no reactivity, the background demonstrates “severe diffuse slowing.”