Electroencephalography in Internal Medicine: A Practical Guide for the Modern Clinician

Dr Neeraj Manikath , claude.ai

Abstract

Electroencephalography (EEG) remains an indispensable diagnostic tool in contemporary internal medicine practice, extending far beyond its traditional epilepsy applications. This comprehensive review addresses the critical clinical scenarios where EEG provides diagnostic and prognostic value, discusses patient selection, and offers practical insights for internists. We explore established indications, emerging applications, and common pitfalls in EEG interpretation, with emphasis on conditions frequently encountered in hospital medicine.

Introduction

The electroencephalogram, first recorded in humans by Hans Berger in 1924, continues to evolve as a cornerstone neurophysiological investigation. Despite advances in neuroimaging, EEG remains unique in providing real-time assessment of cortical function with millisecond temporal resolution. For internists, understanding when to order EEG, interpreting basic patterns, and recognizing its limitations can significantly impact patient outcomes, particularly in critical care, infectious diseases, and metabolic medicine.

Fundamental Principles: What EEG Actually Measures

Pearl: EEG records postsynaptic potentials from pyramidal neurons in the cerebral cortex, not action potentials. Approximately 6 cm² of synchronously activated cortex is required to generate detectable scalp potentials.

The standard 10-20 electrode placement system uses 21 electrodes positioned proportionally across the scalp, ensuring reproducibility across patients and institutions. Modern digital EEG typically samples at 256-512 Hz, allowing detection of frequencies from 0.5 to 70 Hz, encompassing all clinically relevant brain rhythms:

Delta (0.5-4 Hz): Predominant in deep sleep; pathological when excessive in wakefulness
Theta (4-8 Hz): Normal in drowsiness; abnormal if focal or excessive while alert
Alpha (8-13 Hz): Posterior dominant rhythm in relaxed wakefulness
Beta (13-30 Hz): Frontal predominance, enhanced by benzodiazepines
Gamma (>30 Hz): Associated with cognitive processing

When to Order EEG: Evidence-Based Indications

1. Suspected Seizures and Epilepsy

Primary Indication: Any adult with suspected seizure activity requires EEG evaluation.

The diagnostic yield of routine EEG in epilepsy diagnosis ranges from 29-55% for first recordings, increasing to 80-90% after three serial EEGs. Timing critically affects yield—recording within 24 hours of a suspected seizure increases epileptiform discharge detection by approximately 50%.

Hack: Request sleep-deprived EEG for suspected epilepsy with normal routine EEG. Sleep deprivation increases epileptiform discharge detection by 30-40%, particularly for juvenile myoclonic epilepsy and generalized epilepsies.

Oyster: A normal EEG never excludes epilepsy. Approximately 12-50% of patients with epilepsy have persistently normal interictal EEGs. Conversely, 0.5-2% of healthy adults show epileptiform discharges without clinical seizures.

2. Altered Mental Status: The Internist's Dilemma

EEG proves invaluable for distinguishing among the myriad causes of encephalopathy in hospitalized patients.

Non-convulsive Status Epilepticus (NCSE):

NCSE accounts for 10-40% of status epilepticus cases in intensive care units. The Salzburg Consensus Criteria provide diagnostic framework, but EEG remains essential for definitive diagnosis. Studies demonstrate that 8-48% of comatose ICU patients without obvious seizures have NCSE on continuous EEG monitoring.

Clinical Pearl: Consider NCSE in any patient with unexplained altered consciousness, particularly with:

Subtle eye movements (nystagmoid jerks, eye deviation)
Fluctuating consciousness
Automatisms or subtle twitching
Failure to improve despite treatment of presumed etiology

Hack: In suspected NCSE, perform a benzodiazepine trial with simultaneous EEG monitoring. Clinical and electrographic improvement within minutes supports diagnosis and initiates treatment simultaneously.

3. Metabolic and Toxic Encephalopathies

EEG demonstrates characteristic patterns in various metabolic derangements, though most are non-specific:

Hepatic Encephalopathy: Triphasic waves (bilaterally synchronous, high-amplitude, positive sharp waves with anteroposterior lag) appear in 25-35% of patients with hepatic encephalopathy but lack specificity—they occur in uremia, hypercalcemia, and toxic-metabolic states.

Hypoxic-Ischemic Encephalopathy: Post-cardiac arrest EEG provides critical prognostic information. Continuous EEG background reactivity, sleep-wake cycling, and absence of highly malignant patterns (burst-suppression, generalized periodic discharges on suppressed background) predict better outcomes.

Pearl: The American Clinical Neurophysiology Society's Standardized Critical Care EEG Terminology (2021) facilitates communication about critically ill patients' EEG patterns, improving prognostication consistency.

4. Infectious Encephalitis

Herpes Simplex Encephalitis (HSE):

HSE classically produces periodic lateralized epileptiform discharges (PLEDs) over temporal regions, typically appearing 2-15 days after symptom onset. While MRI has superior sensitivity, EEG abnormalities may precede neuroimaging changes by 48-72 hours.

Hack: In suspected encephalitis with normal initial MRI, repeat EEG every 24-48 hours. Serial EEGs increase diagnostic sensitivity and guide antiepileptic therapy in patients developing seizures (occurring in 50-70% of HSE cases).

Creutzfeldt-Jakob Disease (CJD):

Sporadic CJD produces characteristic periodic sharp wave complexes (PSWCs) in 64-78% of cases, typically appearing in later disease stages. The pattern consists of 1-2 Hz generalized or lateralized sharp waves, often triggered by loud sounds or tactile stimulation.

5. Dementia Evaluation: Selective Application

When EEG Adds Value:

Rapidly progressive dementia (within 1-2 years)
Suspected CJD or prion disease
Concurrent episodic alterations in consciousness
Distinguishing dementia from depression or delirium

Oyster: EEG rarely assists in typical Alzheimer's disease diagnosis. The American Academy of Neurology guideline recommends against routine EEG in dementia evaluation unless specific indications exist.

Where Should EEG Be Performed?

Routine Outpatient EEG

Standard 30-minute recordings suffice for most indications:

Evaluating suspected seizures
Following known epilepsy
Assessing medication effects

Setting: Outpatient epilepsy centers or hospital neurophysiology departments

Prolonged Ambulatory EEG

Duration: 24-72 hours

Indications:

Suspected non-epileptic events
Quantifying seizure frequency
Distinguishing syncope from seizures

Advantage: Captures habitual events in natural environment; however, limited electrode array reduces spatial resolution compared to video-EEG.

Video-EEG Monitoring (Epilepsy Monitoring Unit)

Gold standard for characterizing spells and medication adjustment.

Duration: 3-7 days typically

Indications:

Differentiating epileptic from non-epileptic events
Pre-surgical epilepsy evaluation
Characterizing seizure semiology
Medication adjustment in refractory epilepsy

Continuous EEG Monitoring (cEEG) in ICU

Critical care application with expanding indications.

American Clinical Neurophysiology Society Guidelines recommend cEEG for:

Clinically evident seizures requiring adjustment of antiseizure medications
Acute brain injury with depressed level of consciousness
Unexplained altered mental status
Post-cardiac arrest coma
Periodic or rhythmic EEG patterns on routine EEG

Duration: Minimum 24 hours recommended; 48 hours captures >95% of seizures in at-risk populations.

Emerging Quantitative EEG: Compressed spectral array, amplitude-integrated EEG, and rhythmicity analysis enable non-neurophysiologists to screen for seizures, though expert review remains essential.

In Whom: Patient Selection Strategies

High-Yield Populations

1. Post-Cardiac Arrest Patients

All comatose survivors warrant continuous EEG within 24 hours. Seizures occur in 12-47%, predominantly within first 48 hours, and treatment improves neurological outcomes.

2. Critically Ill Patients with Fluctuating Consciousness

Even without overt seizures, 10-20% of ICU patients with altered consciousness demonstrate electrographic seizures.

3. Patients with First Unprovoked Seizure

EEG within 24-48 hours guides recurrence risk stratification. Epileptiform abnormalities increase recurrence risk from 20-30% to 60-70% over two years.

4. Autoimmune Encephalitis

Patients with suspected limbic encephalitis, particularly anti-NMDA receptor encephalitis, require EEG. Characteristic patterns include excessive beta activity (delta-brush pattern) or extreme delta slowing.

Populations Where EEG Rarely Helps

Oyster Alert:

Simple syncope: EEG typically normal; history and cardiac evaluation more informative
Isolated headache: EEG lacks diagnostic utility in primary headache disorders
Stable dementia: Unless features suggest CJD or episodic events
Uncomplicated concussion: EEG findings non-specific and don't alter management
Screening asymptomatic patients: No evidence supports EEG as screening tool

Interpretation Pearls for Internists

Recognizing Critical Patterns

1. Status Epilepticus Patterns

Generalized spike-and-wave >2.5 Hz
Periodic discharges (>2 Hz) with spatial evolution
Rhythmic delta activity with superimposed faster frequencies

Hack: Use the "2 Hz rule"—periodic or rhythmic patterns >2 Hz with evolution warrant aggressive antiseizure treatment even if clinical seizures aren't obvious.

2. Malignant Patterns Post-Cardiac Arrest

Poor prognosis patterns include:

Suppressed background (<10 μV)
Burst-suppression with generalized periodic discharges
Alpha coma (diffuse unreactive alpha activity)

Pearl: No single EEG pattern absolutely predicts outcome. Multimodal prognostication incorporating neuroimaging, biomarkers, and serial neurological examinations provides optimal accuracy.

3. Distinguishing Artifact from Cerebral Activity

Common artifacts mimicking pathology:

Glossokinetic artifact: Tongue movements mimicking frontal slowing
Muscle artifact: Can obscure beta/gamma frequencies
Electrode artifact: May resemble spike discharges

Hack: Check for field distribution—true cerebral activity follows anatomic distributions; artifact appears non-physiologic or isolated to single electrodes.

Emerging Applications

1. Sepsis-Associated Encephalopathy

Recent studies demonstrate EEG abnormalities in 70-100% of septic patients with altered consciousness. Specific patterns may predict mortality and guide prognostication, though prospective validation is needed.

2. Autoimmune Encephalitis Spectrum

EEG helps diagnose and monitor treatment response in anti-NMDA receptor, anti-LGI1, and other autoimmune encephalitides, where specific patterns correlate with antibody subtypes.

3. Monitoring Drug-Induced Coma

During therapeutic hypothermia or barbiturate coma for intracranial pressure control, continuous EEG ensures adequate burst-suppression without excessive suppression risking neurological injury.

Practical Workflow: Ordering and Interpreting EEG

Step 1: Formulate Clinical Question

Specify the indication clearly:

❌ "EEG for altered mental status"
✅ "EEG to evaluate for non-convulsive seizures in patient with fluctuating consciousness post-cardiac arrest"

Step 2: Choose Appropriate Study Duration

Routine (20-30 min): Baseline epilepsy evaluation, medication monitoring
Extended (1-2 hours): Include sleep if possible
Ambulatory (24-72 hours): Capture habitual events
Continuous (≥24 hours): ICU patients, high seizure risk

Step 3: Optimize Study Quality

Hack for better yield:

Sleep deprivation (24 hours without sleep)
Hyperventilation (3 minutes, if safe)
Photic stimulation
Medication reduction (supervised setting only)

Step 4: Correlate with Clinical Context

EEG interpretation requires clinical correlation. A "slow background" means vastly different things in post-ictal state versus evolving dementia.

Cost-Effectiveness Considerations

Routine EEG costs approximately $300-600, while continuous ICU monitoring exceeds $1,500-3,000 daily. However, identifying NCSE prevents inappropriate psychiatric diagnoses, reduces ICU stays, and improves outcomes, generating substantial cost savings.

A 2016 cost-effectiveness analysis demonstrated continuous EEG monitoring in comatose cardiac arrest survivors increased quality-adjusted life years while reducing overall costs through improved prognostication and targeted therapies.

Common Pitfalls and How to Avoid Them

Pitfall 1: Ordering EEG too late

Solution: Request within 24 hours of suspected seizure or mental status change

Pitfall 2: Accepting "normal EEG" as excluding epilepsy

Solution: Consider repeat studies, sleep-deprived EEG, or ambulatory monitoring

Pitfall 3: Over-interpreting non-specific slowing

Solution: Recognize that generalized slowing indicates encephalopathy but rarely identifies specific etiology

Pitfall 4: Missing NCSE in ICU patients

Solution: Liberal use of continuous EEG in patients with unexplained altered consciousness

Pitfall 5: Ordering EEG for unclear indications

Solution: Consult neurology when uncertain; inappropriate EEGs waste resources and delay appropriate diagnostic evaluations

Conclusion

EEG remains an essential tool in the internist's diagnostic armamentarium, particularly in critical care and neurological emergencies. Understanding when to order EEG, selecting appropriate study duration, and recognizing basic patterns enhance diagnostic accuracy and patient outcomes. While neuroimaging provides anatomic detail, EEG uniquely assesses functional brain activity in real-time, making it irreplaceable for diagnosing seizures, characterizing encephalopathies, and prognosticating neurological outcomes.

The modern internist should maintain a low threshold for EEG in altered mental status, liberally employ continuous monitoring in high-risk ICU populations, and collaborate closely with neurophysiologists for optimal interpretation. As quantitative EEG analysis and artificial intelligence tools evolve, EEG will become increasingly accessible and interpretable at the bedside, expanding its utility in internal medicine practice.

Final Pearl: When in doubt about whether EEG will help, ask yourself: "Could this be a seizure, and would knowing change management?" If yes, order the EEG—it's one of the highest-yield, non-invasive tests available.

Key References

Herman ST, et al. Consensus statement on continuous EEG in critically ill adults and children. J Clin Neurophysiol. 2015;32(2):87-95.
Claassen J, et al. Detection of electrographic seizures with continuous EEG monitoring in critically ill patients. Neurology. 2004;62(10):1743-1748.
Sandroni C, et al. Prognostication in comatose survivors of cardiac arrest: An advisory statement from the European Resuscitation Council and the European Society of Intensive Care Medicine. Resuscitation. 2014;85(12):1779-1789.
Hirsch LJ, et al. American Clinical Neurophysiology Society's Standardized Critical Care EEG Terminology: 2021 Version. J Clin Neurophysiol. 2021;38(1):1-29.
Beniczky S, et al. Unified EEG terminology and criteria for nonconvulsive status epilepticus. Epilepsia. 2013;54(Suppl 6):28-29.
Gronseth GS, Greenberg MK. The utility of the electroencephalogram in the evaluation of patients presenting with headache: a review of the literature. Neurology. 1995;45(7):1263-1267.
Taccone FS, et al. Sepsis is associated with altered cerebral microcirculation and tissue hypoxia in experimental peritonitis. Crit Care Med. 2010;38(2):550-557.
Rossetti AO, et al. Neurological prognostication of outcome in patients in coma after cardiac arrest. Lancet Neurol. 2016;15(6):597-609.

Word Count: 2,000

Disclosure: The author reports no conflicts of interest relevant to this manuscript.