EEG in Specific Disease States
James D. Geyer
Paul R. Carney
Erasmo A. Passaro
A wide array of medical and neurologic disorders can cause changes to the EEG patterns. These changes are rarely pathognomonic. In fact, they are usually nonspecific and highly variable, limiting their diagnostic significance. In most cases, however, they provide valuable adjunctive information, and some patterns are indeed so specific that they are virtually diagnostic of the conditions with which they are associated. More commonly, EEG can be used to rule out a diagnostic possibility; for example, an EEG can determine whether a poorly responsive patient is encephalopathic due to a metabolic condition (triphasic waves in renal or hepatic encephalopathy) or has nonconvulsive status epilepticus. This chapter will review a few specific disorders with impact on the EEG. For convenience, these conditions will be grouped into paroxysmal disorders, cerebrovascular disease, infections and postinfectious syndromes, degenerative disorders, multiple sclerosis, mass lesions, metabolic disorders, and trauma.
PAROXYSMAL DISORDERS
Syncope
EEG recorded during a cardioneurogenic or vasovagal syncopal episode will generally show a high-voltage delta activity developing ˜10 seconds after the onset of syncope, followed by attenuation of the background rhythms. While this is the classic description, there is no single specific pattern and the findings will vary.1 It is important to remember that syncope is a symptom common to numerous etiologies. When the syncope is secondary to primary neurologic causes such as a seizure, basilar migraine, or ischemia, the EEG findings will correspond to those particular disorders; however, syncope due to neurologic causes is highly uncommon.
REVIEW
17.1: An EEG consisting of high-voltage delta activity developing ˜10 seconds after the onset of a syncopal episode followed by attenuation of the background rhythms is pathognomonic for syncope.
a. True
b. False
View Answer
17.1: b. High-voltage delta slowing 10 seconds after onset of syncope is the classic finding for cardiogenic syncope, likely related to cortical dysfunction produced by hypoperfusion. However, it is nonspecific and could also be produced by vertebrobasilar migraine, brainstem ischemia, or other structural or metabolic disturbances.
Migraine
In most cases, the EEG is normal before, during, and after the headache. There have been some reports of focal slowing over the affected hemisphere, especially if the patient has a transient focal neurologic deficit during the headache,2 as seen in complicated migraine or hemiplegic migraine.
Transient Global Amnesia
In most cases, the EEG is normal in patients with transient global amnesia. In a small percentage of patients, there is mild diffuse slowing of the background rhythms.3
CEREBROVASCULAR DISEASES
Stroke
In general, the EEG abnormalities identified in stroke patients depend on the location and size of the infarction. For most cortical strokes, the EEG will show focal slowing and occasionally sharp waves. Lateralized periodic discharges (LPDs) can
occur after acute cortical stroke but generally disappear within days. Brainstem infarctions can result in significant slowing of the background and alteration of reactivity. Subcortical/lacunar strokes often show no change on EEG.
occur after acute cortical stroke but generally disappear within days. Brainstem infarctions can result in significant slowing of the background and alteration of reactivity. Subcortical/lacunar strokes often show no change on EEG.
Locked-in states following pontine infarction are usually associated with normal background activity or mild scattered slowing on EEG. The background is typically reactive.4
Carotid Endarterectomy
EEG is sometimes used to monitor cerebral activity during carotid endarterectomy surgery to ensure adequate cerebral perfusion and warn the surgeon when brain circulation is compromised. When performing carotid endarterectomy EEG monitoring, there should be at least 10 minutes of EEG recorded prior to clamping the carotid artery. Recording should continue for at least 10 minutes after the clamp is removed. Beta activity typically increases with anesthetic induction with admixed delta activity. Focal abnormalities may become apparent during anesthetic induction in ˜40% of patients. A decrease in low-amplitude beta activity is the initial EEG change after clamping. Approximately 1% of patients have a persistent focal abnormality following endarterectomy, likely indicating cerebral embolic or ischemic infarction.5
EEG changes associated with endarterectomy depend on the degree to which cerebral perfusion is preserved by collateral flow during the carotid clamp period. EEG effects were proportional to cerebral blood flow, as shown in Table 17.1.
Intracranial Hemorrhage
Subarachnoid hemorrhage is associated with diffuse slowing of the background rhythms.6 There is superimposed focal slowing near a localized hematoma. Cortical parenchymal hemorrhages typically result in focal slowing. A subdural hematoma may cause reduced amplitude of the background and act as a high-frequency filter, producing apparent loss of faster frequencies over the hematoma, but the EEG may be normal. It may be difficult to determine whether apparent focal slowing is due to cortical injury or the filtering effects of the mass lesion.
TABLE 17.1 EEG changes associated with carotid clamping during CEA | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|
|
Hypoxic-Ischemic Encephalopathy
The EEG findings depend on the severity of the hypoxic-ischemic encephalopathy. The abnormalities range from generalized slowing of the background to electrocerebral inactivity (ECI). In less severe stages, the background slowing may or may not be reactive. In more severe cases, the slowing becomes more pronounced and of lower amplitude. Intermittent epileptiform activity, burst suppression, and ECI occur in the most severe cases (see Fig. 17.1A and B).7 The recording parameters required for ECI evaluation used as are described in Table 17.2. If the recording is to be used as adjunctive testing for brain death determination, the integrity of the entire recording system should be tested (usually by tapping on each electrode).8 If EMG contamination cannot be definitively distinguished from possible cerebral activity, a short-acting neuromuscular blocking agent such as pancuronium bromide (Pavulon) or succinylcholine (Anectine) can be used in mechanically ventilated patients to eliminate muscle artifact. This procedure should be performed under the direction of an anesthesiologist or other physician familiar with the use of the drug. These agents should not be used if an apnea test is pending, since they could confound the results.
REVIEW
17.2: Which of the following parameters are incorrect for the recording of the ECI EEG? There may be more than one correct answer.
a. Sensitivity: 10 µV/mm
b. Interelectrode distance: 10 cm
c. Impedance: 100-100,000 Ω
d. Minimum of eight scalp electrodes
e. Low-frequency filter: <1 Hz
f. High-frequency filter: >30 Hz
View Answer
17.2: Answers a and c are incorrect. Sensitivity should be 2 µV/mm and impedance should be <10 000 Ω.
Sturge-Weber Syndrome
The classic description of the EEG seen with Sturge-Weber syndrome includes depression of normal background rhythms and decreased response to hyperventilation and photic stimulation over the affected side.9 These findings do not necessarily relate to degree of vascular calcification.
TABLE 17.2 Electrocerebral inactivity—EEG recording parameters | ||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
INFECTIONS AND POSTINFECTIOUS SYNDROMES